Everything you ever wanted to know so you can build an EV

Lightweight / Folding / Portable EVs - seats optional
User avatar
ElectricGod   10 MW

10 MW
Posts: 2205
Joined: Nov 01 2015 5:24pm
Location: 3 OH 3

Everything you ever wanted to know so you can build an EV

Post by ElectricGod » Jun 01 2018 12:57pm

Hi folks,

Many people have PM'd me after looking at my Currie scooter conversion thread. I've been asked lots of questions on how to do their own builds or how I did certain things on the Currie. If you are not familiar with that thread, it is found here. This little machine weighs 75 pounds, accelerates like my Golf GTI while carrying my 240 6'4" hulk and tops out at 45mph. It's been a great build and so much fun to ride. Right now, it is getting a few changes made to it to fix some niggling details.

viewtopic.php?f=35&t=83830

This is the Schwinn S1000. They come in a few variations. This is not quite the best of the variants. For example it has a brushed motor and the forks don't use a disk brake. The factory motor will get replaced and I've found a source for the correct forks. A long time ago I put a set of S1000 forks on the Currie. This scooter will get that upgrade too. Regardless of the variant of the S1000, they are not very common machines. I've seen only 2 of them ever before. A friend in NYC has one and the other was in Denver. I've been looking for one for more than a year. i had to drive 90 miles to get it. This scooter will be the platform for this thread.

Image
Image
Image
Image
Image

People keep PMing me with questions about how to build a scooter so my intention for this thread is to make it into a tutorial of sorts where I explain the various steps and show lots of specifics so that others (mostly the new folks to EV's) can learn and then be able to duplicate what I show below on their own builds. There will be choices I make that you might not agree with or have a better idea, feel free to post them. BUT, keep in mind that the idea here is to be helpful to beginners as much as possible and to do a build with that same information.

When finished, the S1000 scooter will do 45-50mph, accelerate like my GTI does and have around 30 miles of range at near continuous full speed. Piddling around doing 15 or 20mph ought to get me more like 40 miles. The Currie has similar specs. It has 24 miles of range while doing 30-40 mph, tops out at 45 and can run there for many miles at a time.

The Currie runs on an Alien Power C80100 at 80kv and 66 volts. At my watt meter, I see 4kw at 40-45 mph or under heavy acceleration. It is quite strong and at full power, the only person that can handle the Currie is me. It keeps up with cars on the road easily. I'm adding a 3 speed switch so I can program lower power settings into the controller for other people. Low will be weak and slow (2kw and 25mph). Medium will be strong and slow (4kw and 25mph). High will be full power and full speed (4kw and 45mph). The Schwinn will be using a Revolt RV-100 outrunner. The C80100 and the RV-100 series are essentially the same sized motors, but the RV-100 series motors are a fair bit stronger than the C80100. In talking to Revolt, they have a key between the shaft and bell so that there is no need to reinforce this union like I did on the C80100. They also provide the option of using sealed bearings in steel, hybrid or all ceramic. In my C80100, I had to do all of this myself. The Revolt motor will be stronger (better magnets, more wattage) and set up from the factory with everything needed for reliable EV use. As a result the S1000 ought to be a good bit stronger and faster than the Currie for about the same weight and about the same size. The Currie is already very strong. Keeping the front wheel on the ground while accelerating always requires leaning forward. I'm looking forward to whatever new amount of insanity the RV-100 provides. :):):):):)

Here's the link to the RV-100 motors.
http://www.revolt.org.il/rv-100-model/

Revolt has a sale on the RV-100-regular motor. They are $153 each so I bought one...just to see what Revolt outrunners were like. It is currently shipping to me.

http://www.revolt.org.il/sale/rv-100-regular-sale/

As a side thread, since I currently own 2 C80100 outrunners and have an RV-100-regular coming and soon an RV-100E, I'll do a review of these motors so that people can see for themselves the build quality and make their own comparisons. I also have a CA80-160 Turnigy outrunner. I might as well throw that in the mix too, but I can tell you it is inferior to the C80100 in just about every way. I probably won't ever use it on an EV. Here is that thread.

viewtopic.php?f=30&t=94975&p=1390825#p1390825
Last edited by ElectricGod on Jun 27 2018 8:12pm, edited 2 times in total.
ElectricGod...it's a joke...don't take things so seriously folks!

XB-502 Moped:
https://endless-sphere.com/forums/viewt ... 0#p1222730

Currie scooter:
https://endless-sphere.com/forums/viewt ... 7#p1227407

User avatar
ElectricGod   10 MW

10 MW
Posts: 2205
Joined: Nov 01 2015 5:24pm
Location: 3 OH 3

Battery options

Post by ElectricGod » Jun 01 2018 3:18pm

I'll start with batteries. I'm sure someone out there In ES land is going to get all offended about what I say next. LOL!

You have essentially 5 kinds of batteries that get used in EV's. SLA, LIPO, LION, LIFE and LTO. I'll talk about each one and it's advantages and disadvantages. Don't expect loads of details. Go do some research if you want to know more. I want to give you enough information to be able to get moving.

1. SLA: This is short for Stupid Lead Acid...OK no not really...Standard Lead Acid. These batteries are cheap, heavy and low charge density. For the weight of SLA, you can build a LIPO or LION pack of the same weight and size, but have 3-4X more capacity. SLA is just the worst battery tech there is right now for EV's. They sag badly under load. They sag even worse when cold. They have a very short life span. LIPO is good for around 200-300 charge cycles and this is 2X better than SLA on a good day. Load down SLA's frequently and the life span of the SLA battery drops rapidly. They are ideal for .5C discharge rates in things that don't ever move such as a battery backup for your computer or the storage units for your solar panels. Lots of low cost EV's use SLA just becasue they are cheap and at least get the EV moving. Just ignore SLA for your EV. Everything else is better.

2. LIPO: You have 2 variants of...graphene and not graphene. Essentially what this means is the conductive elements (which BTW do NOT have graphene in them) are made differently. This allows for higher discharge rates and better life spans. If you want to know all the in's and out's go do some research. In either case, LIPO is pretty cheap and light for the capacity it provides. LIPO packs come in lots of capacities, discharge rates and cell counts. For a beginner, they are pretty easy to work with and require little knowledge about battery technology and building a pack. I use LIPO extensively in EV's. I have 14 10,000 mah Multistar 4S packs, 8 10,000mah 4S graphene packs and 20 16,000mah multistar 4S packs. I also have probably 80 8000mah super cheap lipo cells. You could say that I have a lot of LIPO! Ignore the hype that LIPO is dangerous and explodes. It can explode, but then so can gas in your car, but when did you see a car spontaneously explode last? LIPO is about like that. It can happen, but it's NOT common. LIPO doesn't like the cold very much. At 50F expect noticeable reduction in capacity and discharge rates. At 10F, expect very noticeable sagging and reduced capacities. LIPO will typically last 200 charge cycles for decent batteries with no significant loss of capacity. I have a friend that claims he has well over 400 charge cycles on a LIPO pack and has lost 20-25% of pack capacity. He never charges to 4.2 volts per cell and never runs down to 3 volts per cell. LIPO can last a long time if you are very gentle with it. Of all the lithium battery types LIPO is the easiest to get, easiest to work with and the cheapest per amp hour. It is not the lightest or the most power dense or the longest living.

3. LION: You have 2 main options here that most people will use...18650 and 21700. These numbers define the size of the battery can. 18650 is 18mm diameter by 65mm tall. 21700 is 21mm x 70mm. LION cells have a fairly sizeable electrode inside them. Make this electrode smaller and the cell can't discharge as fast, but it can have more capacity. Make the electrode larger and you get more current, but less capacity. 21700 allows for more space inside the can so that your compromises can be on a slightly larger scale. Otherwise, the chemistry and internal cell designs are identical. Expect to pay a lot for a built pack or expect to learn how to spot weld if you build your own pack. In either case a pack is made of many cells welded together in parallel and series to get the voltage, capacity and current desired. LION does better than LIPO, but still suffers loss in the cold. Ask anybody with an electric car about that! LION has good charge cycles... at least 500 cycles for cheap LION. Good LION cells should see 1000+ charge cycles. It gets used in everything from chordless tools, laptops to Teslas. Since the pack is welded together, if a cell dies for whatever reason servicing the pack is difficult. LION has the best capacity to weight and size of the 5 battery types. About the time you replace a good LION pack once, you will probably replace a LIPO pack 3 times and SLA probably 6 times. If you can afford to buy a LION pack, in the long run you will save a lot of money.

4. LIFE: I ignore this technology. It's one advantage over anything else is that it wont ever explode, catch fire or do anything categorically bad. Otherwise it's heavier than LIPO for the same capacity and costs a lot more. IMHO, just don't bother with LIFE. LIPO is so unlikely to ever have an explosive ending that I don't ever worry about it and LION is even better than LIPO. LIFE pretty much can't be made to explode even when thrown directly into a fire. The level of safety for LION is so good that more safety IMHO is just not needed. Why bother with LIFE? Otherwise, expect something similar to cheap LION charge cycles and similar resistance to the cold.

5. LTO: This battery type gets a lot of grief for 2 reasons. The weight to capacity is better than SLA and worse than LIFE. The cell at full charge is 2.8 volts so it takes more of them to build a pack at any voltage. We've gotten so used to 4.1 volts per cell that something that runs at a significantly lower voltage is somehow bad. It's not bad, just different. LTO has really good cold weather resistance. When your LION pack is sagging badly at 15F, an LTO pack is barely seeing any reduction in performance. LTO typically has 10,000 charge cycles. Expect to pass your LTO pack on to your children. 10,000 cycles is something like 30 years of battery life. None of the above battery types tolerate draining to zero without significant loss of capacity. LION, LIPO and LIFE will lose 5% or more capacity if this happens only once. Drain them below their lower limit and they lose capacity. LTO, run it dead flat repeatedly and there will be no capacity loss. Leave LTO discharged for 10 years and it will charge right back to 100% with full capacity. LTO is perfect for harsh use conditions where neglect is probable. LTO tolerates discharging them at well over their ratings far better than anything else. This is a case where LIPO and LION will have issues and at least lose capacity. They may also get hot enough to possibly catch fire. LTO...just keeps on going. IMHO, since they are so cheap, they are a far better option than LIFE and have all the advantages of LIFE. If you can tolerate the added weight of LTO over the other lithium options, then it's advantages are well worth it. For all the other lithium battery options I recommend using a BMS to protect the pack from discharging the cells too far and to protect the pack from too much current draw. For LTO, all you care about is not exceeding 2.8 volts per cell. For that a cheap balance board will do the job. LTO cells have large aluminum tabs on them. Run a couple of screws through the tabs and 2 cells are connected together. Disassembly is just as easy. For building a pack, LTO is pretty easy to work with since you need no specialized tools beyond pliers and screw drivers.
Last edited by ElectricGod on Jun 27 2018 8:15pm, edited 1 time in total.
ElectricGod...it's a joke...don't take things so seriously folks!

XB-502 Moped:
https://endless-sphere.com/forums/viewt ... 0#p1222730

Currie scooter:
https://endless-sphere.com/forums/viewt ... 7#p1227407

User avatar
ElectricGod   10 MW

10 MW
Posts: 2205
Joined: Nov 01 2015 5:24pm
Location: 3 OH 3

BMS - Battery Management System

Post by ElectricGod » Jun 01 2018 4:08pm

If you run a lithium battery of any kind it is a good idea to manage the battery pack in some way. No battery chemistry tolerates over charging very well. Lithium batteries REALLY don't tolerate it. Expect them to over heat, suffer damage, lose capacity and lose charge cycles.

A BMS is an electronic device that monitors the individual cells in a battery pack to make sure that all cells are always operating withing their specific voltage ranges. LIPO is 3 volts to 4.2 volts, LION is 2.5 volts to 4.1 volts, LTO is 0v to 2.8 volts and who cares about LIFE and SLA. I sure don't! Those voltages are limits, but if you want to extend the life span of your pack, then don't run the pack to it's limits. BMS are not always perfect. Sometimes they fail. I have 20 BMS and one failed to do it's job and allowed a LIPO pack to run a cell down to 2 volts. Cheap BMS compared to good BMS can be very different. I have 4 12S BMS that I paid $30 each for. They are usually .1 to .2 volts out of balance all the time in any given pack. They are designed for LION so that can mean a cell is at 4.2 volts while another is at 4 volts at full charge. They don't do low voltage cut off very well either. I don't use these BMS for anything that matters. I have some BMS that balance to a few milli volts very consistently and never allow the pack to run down too far. It used to be that smart BMS were really expensive, but recently that trend has changed significantly and enough so that I will probably never buy a dumb BMS again. In fact, I'm slowly replacing all my dumb BMS with smart ones.

Lets talk about smart vs dumb BMS. Dumb basically means it comes from the factory engineered to meet certain specs that can't be changed. This will be things like low voltage cut off, high voltage cut off and other things that determine good cell and pack management. They also don't tell you battery pack status. If you want to actually know the state of the cells in your pack, pull out a DMM or cell monitor. Smart BMS do everything a dumb BMS does, but you can set those parameters yourself and they will also tell you pack status via an LCD or app on your phone or an app on your PC or all of the above. For people that want to know what their pack is doing at any point in time, a smart BMS that supports a phone app is your best bet. I really recommend using smart BMS that use a phone app.

This thread covers some of my favorite smart BMS and they are pretty reasonably priced, work well and far better than any dumb BMS I've ever used for just a little more money. The 16, 20, 24 and 32S BMS's mentioned in this thread will completely replace all of my dumb BMS's.

viewtopic.php?f=14&t=88676

There will be people that will get all up in arms about this topic and tell you all about how they never use a BMS. Good for them! I too have gone for many months and never used a BMS. It can be done, but there is a cost involved. Be ready to check your pack periodically to make sure it doesn't get out of balance. Be prepared to balance charge your pack on an RC charger. If you have problem cells that don't work as well as the rest of the pack, expect to be manually monitoring your pack frequently (every time you charge). Also those problem cells will likely run down faster than the rest of the pack and that will just shorten their lives faster. The hassle factor alone IMHO makes running without a BMS just not worth the trouble. As a result, I say "always use a BMS"

What does a BMS get you? Lets assume it is working correctly. Usually they do, but I've had one dumb BMS fail to do it's job after many charge cycles. A BMS gets you 4 main things. It's like an insurance policy for your battery pack. Over load the battery pack and the BMS shuts off. Drain the cells too far and the BMS shuts off. Over charge the cells and the BMS shuts off. When charging every cell gets balanced all at the same time. Lithium batteries are expensive and so easy to damage by over charging or running them down too far...protect them! IMHO a BMS is sooo worth having. I've run several packs made of LIPO or LION with no BMS. I was constantly checking the LIPO packs and periodically checking the LION pack when charging to make sure they stayed fairly well balanced and no cell got over charged. A did a lot of balance charging on my RC charger which is slow and arduous. It's fine for a 6S pack or smaller since typical RC chargers handle 6S, but "small" to me is 12S and I never go below 16S. With a BMS, all you care about is the total voltage coming from your EV charger and let the BMS handle the rest. A BMS makes battery care and management easy and care free. With a smart BMS you get 6 things. The above 4 plus flexible configuration and pack monitoring. You get to see everything that is happening in the battery pack as it is happening. I really like smart BMS! They reduce the hassle factor of battery care to almost zero and I never have to open up the battery to see cell/pack status.

I encourage anyone running LIPO, LIFE or LION to use a BMS. For SLA, it's such cheap garbage who cares. Replace them ASAP! For LTO, all you need is a simple balancing option. Super cap charging boards are really cheap. They usually are set to 2.7 volts...close enough. With some minor tweaking of one tiny part, can be made to balance at 2.8 volts. There's also a variety of smart balance boards that work for LTO. If a BMS says it is for LION or LIFE, smart or dumb, it probably won't support LTO. In the above mentioned smart BMS thread there are BMS in there (24 and 32S) that will handle any lithium battery type you want to use.

The short version here is...use a BMS!
Last edited by ElectricGod on Jun 27 2018 8:27pm, edited 1 time in total.
ElectricGod...it's a joke...don't take things so seriously folks!

XB-502 Moped:
https://endless-sphere.com/forums/viewt ... 0#p1222730

Currie scooter:
https://endless-sphere.com/forums/viewt ... 7#p1227407

User avatar
ElectricGod   10 MW

10 MW
Posts: 2205
Joined: Nov 01 2015 5:24pm
Location: 3 OH 3

Chargers

Post by ElectricGod » Jun 01 2018 4:45pm

When are we going to build an EV?! LOL I talked about batteries and BMS, chargers are the next logical thing.

There are several schools of thought here.

1. Smart chargers: There are a few companies out there like Chargery that have done a great job making nearly fool proof chargers that are highly capable. IMHO, all I need is to set the amps, voltage and then see them on an LCD. IMHO, smart chargers are overly complicated and cost a lot, but for someone that is afraid of building your own, they are next best option.

2. Dumb chargers: They might have an LED on the cover that changes from red to green to indicate charging vs done. These devices tend to come in a range of horribly engineered designs. If the charger burns out when over loaded, that's a horribly bad design. At a rock bottom minimum, the charger ought to be able to maintain it's current limits and NOT burn out. How about shut off instead of burning out? I just avoid this category of chargers. Quality and reliability are usually poor and they are not worth having. Quite a few commercially made EV's come with some kind of craptastic dumb charger. I've fixed quite a few chargers in this category when they burned out. I have several and refuse to use them!

3. Build your own: This is me all the way! Here's a nice thread I created so you can build your own charger. Meanwell PSU's are great, very resilient and the S600 series are nearly bullet proof. They have adjustable output voltages and come in a variety of current and voltage specs. If you can't build a high wattage, high voltage charger with meanwells, you are doing something wrong. I've personally built 4 chargers with the S600-24 PSU's. There are other power supply options that work well too. Get some Dell or HP server PSU's on ebay for super cheap. Lenovo 170 watt laptop PSU's work really well too. All of these options don't burn out.

viewtopic.php?f=14&t=90292&p=1316096#p1316096
ElectricGod...it's a joke...don't take things so seriously folks!

XB-502 Moped:
https://endless-sphere.com/forums/viewt ... 0#p1222730

Currie scooter:
https://endless-sphere.com/forums/viewt ... 7#p1227407

User avatar
ElectricGod   10 MW

10 MW
Posts: 2205
Joined: Nov 01 2015 5:24pm
Location: 3 OH 3

BLDC motors - inrunners vs outrunners

Post by ElectricGod » Jun 01 2018 7:34pm

I'm going to ignore brushed motors. If you have one...I'm sorry. I have several too and they will probably never do more than collect dust. The S1000 scooter has a brushed motor on it...oh look...another paper weight!

So then, lets talk about the only kind of motor that matters on your EV...brushless motors.

I'll mention Kv, eRPM, amps, watts, phase amps, WYE and delta and a few other things below. Keep reading....it eventually all becomes clear as mudd!

There are 3 types of BLDC motors and 2 are outrunners.

1. Inrunners: This means the spinning part or armature which has the magnets on it is inside the part with the windings or stator. inrunners tend to be closed up and that means dirt and water don't get into them. Inrunners are expensive compared to outrunners of the same wattage. For the same wattage as an outrunner, they will usually be much larger and weigh 2-3X more. They commonly come wired in WYE which lowers their KV a lot. I personally own 8 inrunners that are EV worthy. 3 are BOMAs and they are the bottom of the barrel for EV grade inrunners. They work, but are not well designed and are cheaply made. You can find them on ebay for less than $200. I also have 3 HLD inrunners. 2 have 50mm armatures and the third has a 25mm armature. The 50mm version will pull you around at 60 mph quite nicely at about 5kw while moving 360 pounds of weight. I paid $375 each for them. Mine have opened up end covers to let in air. I also have 2 AstroFlight 3220 inrunners. They are quite small and good for about 6kw, but they also cost $700 each. Just don't bother unless you are dead set on an inrunner and want something really small. Lets think of a motor as a lever. If you want lots of prying force get a long lever. Inrunners can't have a long lever because it is inside the stator. A long lever means a giant motor and inrunners for the amount of power consumed and the mechanical power they create are already pretty large and heavy. Instead you want an inrunner that spins fast and can be geared down a lot. That said, I have built 2 EV's which I still use with inrunners. Heat is always your enemy and inrunners since the windings are on the outside of the motor can exhaust heat readily through their outer shell. However, the magnets on the armature are trapped inside an oven (stator) and can't get cool and will lose their strength over time making the motor weaker. Rewinding an inrunner is not easy. I'll explain eRPM later, but inrunners tend to have low eRPM

2. Outrunners: The spinning part with the magnets (armature) spins around the outside of the stator. An outrunner has a long lever by comparison to an inrunner. The lever (magnets) sits outside the stator so for a much smaller motor with much less weight, you can get the same amount of torque as an inrunner at far less cost. For example, an HLD inrunner ($375) produces a similar amount of torque at about the same wattage as a C80100 outrunner. The C80100 will cost you less than $200, weigh 1/4th as much as the HLD inrunner and be 1/3rd the size. Why would I EVER use an inrunner? Outrunners as far as I know are always wired up delta. In this build, I've already mentioned that it will use an outrunner. I'm pretty sure my days building with inrunners will end when the last of the ones I have are gone. Outrunners do have a couple of disadvantages that can be over come. The stator is inside so getting air to it is a bit harder so it keeps cool. The Revolt outrunners incorporate a radial fan in the armature which pulls air through the stator. Alien Power motors do not. In either case, the stator is mounted on a large chunk of aluminum. Plant that aluminum on more aluminum and add some heat sink compound between them. Let the EV become the heat sink for the stator. My C80100 runs at around 140 phase amps. The stator since the motor is mounted on a 1/4" thick aluminum plate does not get too hot. Outrunners tend to have much higher eRPMs compared to inrunners. The other big draw back to outrunners is really stupid. I don't know why, but everybody uses shielded bearings in them. WHY WHY WHY do they do that? Just put sealed bearings in there! As you can imagine, I get a new outrunner and I replace the bearings immediately. Dirt and water in bearings = bad bearings. Outrunners are easy to rewind.

3. Hub motors: See outrunners and then add 20-50 pounds so that you can stick it inside your back wheel. LOL. Hub motors are a just a specific kind of outrunner. They tend to have very low Kv's since the motor RPM is the same as the tire RPM. Low RPM motors tend to be less efficient than high RPM motors and hub motors are very low RPM motors. Hub motors despite being an outrunner need a lot of added mass to them since they have to carry the wheel, weight of the EV and rider and deal with road conditions. An outrunner, has to carry it's own armature mass only so there's no reason to make it heavy. A c80100 outrunner has 18 magnets while a similar wattage hub will probably have 32 that also means many more stator teeth (30 or 27).

Lets talk about those terms...Kv, eRPM, amps, phase amps, watts, WYE, delta and stator teeth.

Kv: This is a term that describes RPMs per volt. 1 Kv means at 1 volt you get 1 RPM. 100 KV means at 1 volt you get 100 RPM's. No one runs at 1 volt...that would be silly. 66 volts is common enough. At 100kv you get 6600 RPMs. The Currie runs at 66 volts and has an 80 Kv motor or 5280 RPM's. The equation is simple: KV x volts = max RPM developed. At the Kv limit of the motor, it just won't spin any faster. If you need more motor RPM's increase the voltage. Kv is a physical relationship between the stator teeth and how many turns of wire are on each stator tooth. More turns of wire per tooth lowers the Kv. Less turns raises the Kv. More stator teeth in the same phase is similar to more turns per tooth and also lowers motor kv. Look at a typical hub motor and they will have lots of stator teeth to help get the Kv down.

Magnet poles: I mentioned that BLDC's have magnets, but not magnet pole pairs. All BLDC motors have magnet counts divisible by 4. The lowest number of magnets possible is therefore 4 magnets. Pole pairs is simply the number of magnets divided by 2.

eRPM: This matters becasue the motor controller needs to stay in sync with the motor in order to properly control the motor. If the controller loses sync, the motor stops running, you get herky jerky motor behavior or other bad behavior that could blow out the motor controller. So staying in sync is very good for many reasons. Inside the motor controller is a little "brain" It can only work so fast. Some are faster than others. This little brain has to calculate what to send to the motor on a moment by moment basis and do so many thousands of times a minute. The limit at which the brain can operate and successfully stay in sync with the motor is it's eRPM limit. The eRPM of a motor is volts x Kv x pole pairs. Lets say the motor controller can handle 20,000 eRPM. Let's say we are running at 66 volts from the battery pack and our motor is 100 KV and has 18 magnets or 9 poles. This works out to 47520 eRPM. The controller will lose sync with the motor at less than 50% of full motor RPM's. NOT GOOD! So we get a new motor controller and it has an eRPM of 50,000...phew...just barely good enough with a little to spare! The point here, is you want to have a controller that can hand the eRPM of your motor at whatever is your battery pack voltage. If it can't, then you won't be riding very far before things stop working.

Amps: This is the unit of measurement for current. In EV parlance it is generally implied that this is current when charging or current exiting the battery pack.

Phase amps: This is specific to what is happening inside the motor windings. It would seem logical that you have 10 amps entering the motor controller so there ought to be 10 amps in the motor windings. While this is possible, it is also highly unlikely. Phase amps can be wildly different from battery amps. I've measured 6 battery amps and 90 phase amps before. Magic energy got made somewhere right? No...LOL...it's just the nature of the interaction between windings and iron in the stators and collapsing magnetic fields.

Watts: This is another simple calculation...volts x amps. On the Currie at the watt meter between the battery pack and controller I'll see something like 66 volts and 60 amps or almost 4000 watts. I haven't measured phase amps inside the motor while riding, but lets pretend i measure 100 phase amps. At 66 volts that 6600 watts. So then lets talk about controller watts vs motor watts. Controllers when they have a wattage rating, that usually means battery amps x battery volts. When a motor lists a wattage rating, that means battery volts x phase amps. Confused yet? And don't forget that phase amps can be wildly different from battery amps! A decent rule of thumb is assume a 2kw controller will run a motor of about the same wattage, but that you are losing some amount of torque that the motor is capable of producing. Add more controller wattage...like 50% more. I use a lot of heavily modded 12 fet controllers. They are capable of 60 amps at the battery and the mosfets inside are capable of more like 180 phase amps.

Wye vs delta: There is 2 ways to connect the ends of the motor windings together. In a brushless or BLDC motor you have 3 motor phases or 3 sets of windings. Each winding has a start and an end and the wire is wound around many teeth in the stator. AS a result, BLDC motor stators are always divisible by 3 and the dead minimum number of stator teeth is 9. I won't explain further here. Go do some research! So then you have 3 sets of windings or 6 wire ends, but the motor controller only has 3 connections for the motor phases. Those 6 wire ends need to become 3 connections. This is where WYE vs delta come into play. Lets call the 3 phases 1,2 and 3 and the start of each phase is "a" and the end is "b". 1a is the start of the first winding and 2b is the end of the second winding. WYE is all 3 ends connected together so 1b, 2b and 3b are connected together and 1a, 2a and 3a are connected to the controller. Delta is a bit different. 1a connects to 3b. 1b connects to 2a. 2b connects to 3a. These 3 connections also connect to the 3 phase wires on the controller. In either case 6 wire ends become 3 connections to the controller. There is something else that happens inside the motor with it's Kv and it has to do with how BLDC motors are powered and the motor windings. At any point in time in a motor 2 phases are powered and the third is not. With WYE, this means that 2 phases are in series when they are powered. In delta this does not happen. Each phase is powered by itself. Remember what I said about Kv and windings. In WYE, you have 2 phases in series, this has effectively made a single long wire that is twice as long as any single motor winding. This works out to lowering Kv by a factor of 1.7...yes I know NOT 2. In delta since any one winding is just it's own length, this has the effect of raising the motor Kv by 1.7 or . If you have a motor terminated in WYE with a Kv of 64 and re-terminate it in delta, that raises it's Kv by 1.7 or 108.8kv. At 82 volts that means 5248 RPM in WYE or 8921 RPM in delta. One other detail...longer wire has more resistance than shorter wire. Lets pretend that a motor winding has 1 ohm of resistance in WYE that means 2 winding in series or 2 ohms of resistance. More resistance means more heat. So then delta will have less wire resistance and therefore less heat due to resistance with the exact same wire length per phase. This is a big reason why outrunners are typically wired delta. Why use WYE? That depends on your design goals. Sometime some added heat to get something else is a good thing, sometimes it's not. just know that your motor is terminated one of those 2 ways and what that means and you are good to go.

Stator teeth: The stator is the metal thing that the wire in the 3 phases get wrapped around. The stator is made up of many thin layers of metal stacked together. This otherwise is known as the "stator stack" or just stator. The stator has protrusions that go inwards for an inrunner or outwards for an outrunner. The wire for each phase get wound on these protrusions. They are called stator teeth. The number of stator teeth is always divisible by 3 and ideally there can not be less than 3 of them to make a BLDC motor. Realistically 9 stator teeth is the minimum.

Magnets vs stator teeth: Remember earlier I mentioned the magnets always need to be divisible by 4. There is a ratio here 4:3...sort of. If you have 4 magnets, 3 stator teeth is the minimum you will have, but you could also have 6 or 9 or 12 teeth or anything else divisible by 3. There are practical limits here. For example a motor with 28 magnets wont have 3 stator teeth and a stator with 24 teeth wont have 4 magnets. 28 magnets will probably have 24 stator teeth. A motor with 14 magnets will probably have 12 stator teeth. BUT a motor with 28 magnets can actually run on 3 stator teeth and a 4 magnet motor can run on a 24 tooth stator. I did say "ratio...sort of". It's always maintaining that the number of magnets is divisible by 4 and the number of stator teeth is divisible by 3. This is the rule of thumb, but I have a few BLDC's that don't have magnets divisible by 4.
Last edited by ElectricGod on Jun 27 2018 8:57pm, edited 4 times in total.
ElectricGod...it's a joke...don't take things so seriously folks!

XB-502 Moped:
https://endless-sphere.com/forums/viewt ... 0#p1222730

Currie scooter:
https://endless-sphere.com/forums/viewt ... 7#p1227407

User avatar
ElectricGod   10 MW

10 MW
Posts: 2205
Joined: Nov 01 2015 5:24pm
Location: 3 OH 3

Motor guts

Post by ElectricGod » Jun 01 2018 7:56pm

Lets look at a few motors.

Inrunners: Notice that the windings and stator are around the armature which has the magnets on it.

This is an AstroFlight 3220. It has 8 magnets and 24 stator teeth. Inrunners always have the magnets inside the stator.

Image

HLD inrunner with the armature removed. It has 8 magnets and 12 stator teeth.

Image
Image

Outrunners:

This is the stator from the outrunner that came with the Currie scooter after I rewound it.

Image

C80100 stator (12) and magnets (14). The magnets fit around the outside of the stator.

Image
Image

Alien Power 12090 18kw outrunner.

Image
Image
Image

A couple of hub motors. You can see the ring of magnets around the outside of stator teeth. It's just an overweight outrunner.

Image
Image
Image
Last edited by ElectricGod on Jun 27 2018 9:00pm, edited 2 times in total.
ElectricGod...it's a joke...don't take things so seriously folks!

XB-502 Moped:
https://endless-sphere.com/forums/viewt ... 0#p1222730

Currie scooter:
https://endless-sphere.com/forums/viewt ... 7#p1227407

User avatar
ElectricGod   10 MW

10 MW
Posts: 2205
Joined: Nov 01 2015 5:24pm
Location: 3 OH 3

Motor controllers and their guts

Post by ElectricGod » Jun 01 2018 8:03pm

Lots of preamble so that you understand the basics of what you are messing with before you do an actual build of your own. You can live with out knowing a lot of this stuff and still build a working EV, but since I'm doing a primer for beginners, I might as well cover what is going on in any EV. Knowledge is power after all and the more you know, the more mistakes you can avoid.

A motor controller is device that applies current to the motor phases in a specific pattern. EV grade controllers typically have inputs from the motor that they use to sense the position of the motors armature. This helps the controller know what motor phases to turn on to get the motor spinning or keep it spinning reliably.

I have a variety of controllers so that's the first thing I will do is post pictures of various controllers with the covers on and off. Later, I'll explain what's happening inside and how despite the variations in design how they all work essentially the same. Finally I move onto some nuances in controller design and what they mean for you.

So first up motor controllers and there innards...
ElectricGod...it's a joke...don't take things so seriously folks!

XB-502 Moped:
https://endless-sphere.com/forums/viewt ... 0#p1222730

Currie scooter:
https://endless-sphere.com/forums/viewt ... 7#p1227407

User avatar
armandd   10 W

10 W
Posts: 67
Joined: Jan 18 2018 4:46am
Location: Malaysia

Re: Schwinn S1000 build - how to build your own EV

Post by armandd » Jun 03 2018 1:57am

I'm building my own scooter at the moment and all these is very very helpful! :D

User avatar
ElectricGod   10 MW

10 MW
Posts: 2205
Joined: Nov 01 2015 5:24pm
Location: 3 OH 3

A few common motor controllers

Post by ElectricGod » Jun 06 2018 1:32pm

Hi folks...it's been a few days. Hopefully what I have posted so far has been useful to people.

This is a typical (Probably made in China controller). They come in a variety of designs and shapes, but it's pretty safe that if you get a controller with 2 metal end caps and an extruded aluminum shell that it was made in China and the internals are going to look a lot like this. For example these Grintech controllers and this PowerVelocity controller may have a slightly different shell on them, but the internals are nearly identical.

Image
Image

Since I have lots of pictures of the PowerVelocity controller product line, I'll focus on that for our examples.

This is a bottom view of the board inside the shell. At the bottom edge of the board you will see a large solder trace that runs the length of the board. The right end of the trace isn't reinforced with added copper and solder for about the last inch. This trace is BATT+. You will also see a row of 3 solder connections grouped together and that there are 12 sets of them. These are the individual mosfets soldered to the board. You will notice that some of those mosfet solder connections go to BATT+, 3 shorter larger solder traces and another long solder trace. The large solder trace in the middle left area of the board is BATT-. You can't see it in this picture, there are several shunts that connect this large trace to the long trace that runs the width of the board and is about 1" in from the bottom of the board. I'll talk about shunts some more later. For now, they give the motor controller a way to detect current flow. This long trace is also BATT-. Notice also the 3 shorter reinforced traces. These are the 3 outputs from the controller for the motor phases.

Image

This is a top view of the same controller board and a close up of some of the mosfets. I'll talk some more later about basic controller theory, but any brushless motor controller is going to have 3 sets of mosfets. Each set of mosfets will be divisible by 2. Those 2 divisions will consist of high side and low side mosfets. Since this picture is of a 12 fet controller, those left most 4 mosfets are one of the 3 sets of mosfets and they drive one of the 3 motor phases.

Image

This is a full view of the same controller. A lot of stuff is hidden under those wires, but hopefully with a few pictures you get to see enough to get a good idea of what's inside every motor controller. You see attached to a long aluminum strip 12 mosfets. They are 3 sets if 4 mosfets that drive each motor phase. The aluminum block they are attached to is referred to as a heat spreader. In use, the mosfets get pretty warm, the aluminum block screws to the shell and dissipates that heat to the air. To the left are 2 large capacitors. There's also another one laying down on the board. They filter power entering and leaving the controller to keep it clean and stable. They are connected across BATT+ and BATT-. You will also notice a small yellow transformer next to the large capacitors. Every motor controller has a "brain" inside it. This part is called the Motor Control Unit or MCU. It does not run on battery pack voltage. Most of the time they run at 5 or 12 volts. This little transformer and the associated small caps and other parts right around it, make up a DC-DC converter. This is essentially a circuit that takes battery voltage and converts that to a stable 5 volts. It has a wide range of input voltages it can work with and still produce stable 5 volts for the MCU. About center in the board is a large black square with lots of legs coming out of it. This is the MCU or "brains" of the controller.

Image
Image

Here's a close-up of the MCU. If you go back up to that picture of the bottom of the board I posted earlier, you will notice a section in the upper right area where there are many small solder connections. This is where most of the small wires going in and out of the controller connect. Almost all of them connect to the many legs on the MCU in some way. These connections are for throttle, brakes, cruise control, 3 speeds and a few other things. The MCU takes external signals and translates them into signals that turn on and off the mosfets in the right combinations to create commutation (rotation) in the motor. The MCU is also monitoring battery pack voltage to make sure that the battery isn't over or under voltage. Via the shunts, it can measure current flow in the controller. If the current flow is too high, the MCU will adjust how it turns on and off the mosfets to bring that flow back down again. Too much current and you can destroy the mosfets. This MCU has internal programming for regen. This allows the the motor to act like an electric brake and in the process recover some of the energy spent in moving forward when you slow down.

Image

Let's focus on the shunts. Just above the orange wire are 2 vertical metal objects. These are the shunts. Effectively they are just wire, but they are a specific kind oif wire. Most wire is not calibrated at X length to have Y resistance. Shunt wire is. The shunts are 1" long and at that length will have .005 ohms of resistance. This matters to the MCU. It needs to use Ohms Law (Volts = Current x Resistance) to measure current flow. The MCU is programmed with the resistance of the shunts and then can measure the voltage on the shunts to calculate the current. In this case, 2 .0005 ohm shunts in parallel equals .0025 ohms. As an example, the MCU would measure .1 volts at .0025 ohms and calculate 40 amps current flow. Or it might measure .15 volts and calculate 60 amps. If the controller can only handle 40 amps and the MCU reads 60 amps, it will shut down the mosfets to protect them from over loading and burning out. It may shut them down just a little or a lot or completely. It just depends on the the moment by moment reading for current. This happens so rapidly that you probably don't know it is happening. The over all effect is to manage current in the controller so that it doesn't burn out. People will "butter up" their controller shunts to get it to run a bit harder and at higher current levels. This does work...for a while. What it does is lowers the resistance of the calibrated shunts. This causes the MCU to read less voltage across the shunts and think less current is flowing than what is really there. As a result, a lot of people burn out their mosfets because they don't know what they are messing with or why it works. I do controller mods, but I always use calibrated shunt wire so that the controller can "know" real current levels.

Image

Some controllers are programmable and others are not. You just saw this picture, but lets focus on the black connector at the bottom, middle area of the board. This connector allows your PC or some other device such as a bluetooth connection to be made to the MCU so that settings can be changed. For example battery low and high voltage, current limits, certain throttle settings, motor speed, regenerative braking and other details can be set in the MCU. In the very first picture, those Grintech controllers are not programmable. They are "hard coded" for certain parameters. The silver controller from PowerVelocity has a more powerful MCU in it and can be programmed. On programmable controllers, there will be a place on the board like this that allows connection to the MCU for programming.

Image
ElectricGod...it's a joke...don't take things so seriously folks!

XB-502 Moped:
https://endless-sphere.com/forums/viewt ... 0#p1222730

Currie scooter:
https://endless-sphere.com/forums/viewt ... 7#p1227407

User avatar
ElectricGod   10 MW

10 MW
Posts: 2205
Joined: Nov 01 2015 5:24pm
Location: 3 OH 3

More controller "stuff"

Post by ElectricGod » Jun 06 2018 2:57pm

That last post got loooong so I started a new one.

I want to talk about the mosfets inside every controller there is. There is a bottom limit you can have in a BLDC controller. Remember what I said previously about how there are 3 sets and each set is divisible by 2. At 6 mosfets, that means 3 sets of 2 mosfets...that's the dead minimum you can have. As a result, 6, 12,18 and 24 mosfet controllers are very common.

People commonly think that a 12 fet controller is better than a 6 fet and an 18 fet is better than a 12 fet. IE: more mosfets is better than less. This partly correct and partly NOT correct. Lets assume that we have a 6, 12 and 18 fet controller and all 3 controllers use the exact same mosfet. Then it is true that more mosfets means more power handling. However, if in that 18 fet controller, you have mosfets that can handle 10 amps and in the 6 fet controller you have mosfets that handle 80 amps, then the 6 fet controller wins. This is a real life scenario! I mod controllers and I use mosfets that have very high current ratings and a few other details that make them the "best in class" mosfets available. As a result, I have 12 fet controllers that handle the wattage and current of just about all 18 fet controllers commercially available. IE: It's all about the right mosfets! And in that scenario where you compare an average mosfet to a very good mosfet, it's not an "apples to apples" comparison. Much fewer numbers of very good mosfets are going to better than many more average mosfets.

Lots of people get confused about the amperage of their controller. Remember when I talked about battery amps vs phase amps back in the BLDC motor section. If you don't, go back and re-read it. The mosfets are working in the phases and therefore in the phase amps world. A controllers phase amps capability is defined by it's mosfets. In the TO-220 package you are limited to 75 amps per leg. If you look at that close-up of the mosfets. That is the TO-220 package. 75 amps is a maximum limit of how much current the physical legs on the mosfet can handle before they over heat and melt. Inside the mosfet, that limit may be much more or much less than this physical limit. Just assume that more is always better, but that more than 75 amps doesn't mean a whole lot since you can't use it. So then, ideally a 6 fet controller can handle a maximum of 75 amps per phase. The specific mosfets and how well they can get rid of heat will reduce this maximum phase current considerably. A hot mosfet is a dead mosfet! Run them at their limits all the time and they will get hot and die. This is the very thing that people with "buttered up" shunts typically don't understand. The controller was designed to make the mosfets last and one of those design considerations was what mosfets were used and how warm they would get.

A 6, 12, 18 or 24 fet controller all work identically. All that is really different is more or less mosfets doing the same job. The big difference between 6 and 12 fets is that in the 6 fet controller there are no mosfets in parallel. In the 12 fet controller, you have 2 mosfets in parallel for every job in the controller. 18 fets is 3 in parallel. 24 fets is 4 in parallel. So then back to that 75 amps maximum leg limit. In a 6 fet under ideal conditions, you can have 75 phase amps. In a 12 fet, 150 phase amps. In 18 mosfets 225 phase amps. In 24 mosfets 300 phase amps. This is the dead rock solid written in stone maximums for any TO-220 mosfet leg limits. You will NEVER get these phase amps!!! The moment you do, you are probably replacing the controller or at least the mosfets! Reality is that mosfets warm up the harder they are used. More heat means the closer the mosfet gets to burning up. Mosfets generate heat from 5 main sources. The actual legs on the TO-220 package heat up as they conduct more current. The mosfet itself has 3 internal heat sources and 1 external heat source. Internally they have a resistance when turned on. This resistance creates heat. Ultra low resistance mosfets generate less heat than higher resistance mosfets. Mosfets also have an internal diode. The diode when it is conducting is generating a lot of heat. Mosfets depending on the specifics of the mosfet used turn on and off in a fixed amount of time. Less time is better than more time. More time creates more heat. Externally, the MCU turns on and off the mosfets. How rapidly it does this can be a giant heat generator inside the mosfet. Cool mosfets are happy mosfets. Hot mosfets burn out. Some mosfets tolerate heat better than others. The TO-220 package can shunt away a maximum of 500 watts of heat. Any mosfet with a spec of more than 500 watts in TO-220 is lying. THat's a physical limit of the TO-220 package. However, a few mosfets are designed to tolerate 500 watts of heat, but many (nearly all) are rated for more like 375 watts. This is much more realistic since it doesn't come close to the maximum physical limits of the TO-220 package. So then, as long as your mosfets are never seeing more than 375 watts of heat...however that is accomplished...they are cool enough to keep on going. Reach that maximum heat wattage limit and they burn out soon after. Managing heat in any motor controller is VERY important!

Lets talk about how the mosfets work. This is a good article on H-bridges and how the mosfets work to turn on the motor phases. Replace that motor symbol in the circuits with a single phase in your 3 phase BLDC motor and you have exactly what is happening inside your controller. The only difference is that it is happening in 3 phases pretty much at the same time with 3 sets of H-bridges.

http://www.modularcircuits.com/blog/art ... he-basics/

Now that you've read that article, lets talk a little about hi side and low side mosfets. In the schematics, Q1 and Q3 are high side because they connect to BATT+ and Q2 and Q4 are low side because they connect to BATT-. If you ever hear someone talking about low or high side mosfets, this is exactly what they are referring to.

In the first 2 diagrams you see exactly what happens inside each one of the motor phases over and over again. You have magnets inside the motor, The faces of those magnets facing the stator are oriented N, S, N, S and alternate. There's an even number of magnets so you can be sure that half of the magnet faces closest to the stator teeth are north and the other half are south and they alternate. Power a motor winding and it will present either north or south, depending on the direction of the windings on the stator tooth. This will attract the opposite magnet poles. Reverse the wires for the phase and the phase will now present the opposite magnetic poles on it's stator teeth which will now attract the opposite poled magnets. Do this rapidly, back and forth in a single phase and you will create back and forth jiggling motion in the motor. Do this in 2 phases simultaneously and you get rotation, but very poor rotation. Do this in 3 phases where one of the 3 phases is off while the other 2 are on and alternate the flow of current and you get what we see in modern BLDC motors as smooth and efficient rotation. Never the less, it's all happening in 3 sets of H bridges...one per motor phase. There's some very specific timing for those phase switching events to happen within or else the motor and controller will be inefficient, use way more current and possibly cause a burn out. Fortunately the MCU has all that timing information programmed in so this is usually very reliable and works exceptionally well for billions of phase transitions and mosfet switching operations.
Last edited by ElectricGod on Jun 27 2018 9:16pm, edited 1 time in total.
ElectricGod...it's a joke...don't take things so seriously folks!

XB-502 Moped:
https://endless-sphere.com/forums/viewt ... 0#p1222730

Currie scooter:
https://endless-sphere.com/forums/viewt ... 7#p1227407

User avatar
ElectricGod   10 MW

10 MW
Posts: 2205
Joined: Nov 01 2015 5:24pm
Location: 3 OH 3

braking and regen

Post by ElectricGod » Jun 06 2018 3:34pm

OK...so far we have talked about batteries, BMS, chargers, motors and motor controllers and then gotten pretty deep into some controller theory. Geez...what else is there!?

Let's talk about a few other things that are starting to get to real life EV building.

You have 2 mechanical ways to speed up...The motor and you. I build stuff I don't pedal and don't have to do anything except turn the throttle to get moving. So that means the EV either works or I'm walking. So yeah...It's still the motor or you providing 100% of all mechanical advantage. Hopefully for me it's the motor reliably producing that mechanical advantage and NOT me!

You have 2 ways that won't involve wrecking to slow you down.

Bikes built with a motor often times drive via the bike chain. This means there is a free wheel at the back wheel that allows you to stop pedaling and just coast. Lots of scooters incorporate a free wheel too. This free wheel keeps you from using regen. I want regen since it extends my range and I don't ever pedal so I could care less about that. Regen needs a wheel sprocket that is fixed to the driven tire. The wheel sprocket can then turn the motor when there is reduced throttle and this uses the motor like a generator which recovers some amount of mechanical energy as charge in the battery pack. Regen also slows you down since the motor is now acting as drag on the driven wheel. IMHO, regen is good! There's really good reasons why most everybody that makes commercially available EV's uses regen. I'll talk later about regen some more.

The other way to slow down is brakes. Personally, there's only one kind of brakes to EVER have and that's disk brakes. If you look at any of my EV builds. They all have disk brakes. They get wet, slip minutely and then grab. Any other brake option just loses stopping power. Disk brakes are reliable, self cleaning and easily used for any kind of braking condition. My disk brakes are always stronger than the gripping power of the tires on dry pavement. This means I can lock up the wheels, but it also means that I can brake really hard and still NOT lock up the tires. If your brakes are weaker than tire grip, you can't stop at your maximum grip capabilities. I learned this lesson the hard way. Fortunately all that happened is I left a long scratch down the side of a van and a nice beefy pay out for body work. If my brakes were as good as or better than tire grip, that would have never happened. My brakes were inadequate for the weight they were supposed to stop. I soon upgraded them! Strong brakes are good brakes, weak brakes are going to get you hurt.

Back to regen...some motor controllers support regen, some don't. A lot of controllers that can't be programmed do NOT support regen. The non-programmable ones that do support it, don't let you set how much regen you get. I only use programmable controllers and I'm quite fond of the PowerVelocity product line. I've helped develop them and they are easily modded and hacked if you want to do that. They are also fully programmable from your phone and regen can be set with lots of flexibility. For people that can't use it, turn it off. For people like me who build intentionally to use regen, you can set it to exactly what works for you. On any of my EV's, I get 10-15% more ride time thanks to regen than I would get with it turned off. That's significant! If I got 20 miles per charge, then that's something like 2-3 more miles of range. I can live with that and it costs me nothing to use regen. How is that bad? If I did a bike build, I'd figure out how to add a fixed sprocket for the motor so I could use regen. Since this build is a motorized kick scooter, this is a no brainer. It will use a fixed wheel sprocket so that regen is possible. Right now, it has a freewheel in there somewhere so that coasting is free of any drag from the motor. GRRR!
Last edited by ElectricGod on Jun 27 2018 9:29pm, edited 2 times in total.
ElectricGod...it's a joke...don't take things so seriously folks!

XB-502 Moped:
https://endless-sphere.com/forums/viewt ... 0#p1222730

Currie scooter:
https://endless-sphere.com/forums/viewt ... 7#p1227407

User avatar
ElectricGod   10 MW

10 MW
Posts: 2205
Joined: Nov 01 2015 5:24pm
Location: 3 OH 3

Watt meters

Post by ElectricGod » Jun 06 2018 4:18pm

Good grief! Are we ever going to talk about parts and actually building stuff?! LOL...yes actually right now we are.

There's something that I think is a "must have" item on every EV that has nothing to do with going or stopping. That's a watt meter. The PowerVelocity controllers can incorporate a telemetry module. This module tells you everything a watt meter does, plus lots and lots more. It's a great option, but stand alone watt meters are also excellent. I pretty much use PV controllers so I have telemetry modules, but I also have separate watt meters too. Look at any of my builds, you will see a watt meter incorporated and there's probably a PV controller and telemetry module as well. Yes...this is overkill. BUT, what if I try out some other controller? How do I tell what is happening in the EV? AKA...just use an external watt meter. There's several cheap watt meters easily found on ebay that I really like. Let's talk about a few.

I have a couple of the TK15 watt meters. Look around some and you can find them with larger shunts for more than 50 amps. I think they max out at 90 volts which is suitable for lots of people. They are pretty basic and have a couple of programmable settings that make the battery meter and percentage work correctly. The Currie uses one of these. The TK15 is a very small watt meter and with the right shunt, that too can be very small.

https://www.ebay.com/itm/TK15-Coulomb-M ... Sww3tY3Ngz

Image

I have this meter, but without the box on it. I have one in the XB-502 moped. It works quite nicely, has lots of programming options, works wired or wireless (within limits) and can control other stuff. Wireless in an EV is IMHO, something that is going to be spotty. I've tried quite a few wireless bike speedos and they can't tolerate the EMF noise created by the motor and controller. I'm not overly confident about wireless watt meters for the same reasons. As a result, I run mine fully wired. This watt meter can't be 100% wireless no matter what you do. There's no internal battery so you have to power the display externally. By then, what's the point with going wireless at all? Still, a very nice meter, with lots of programming options and can be used to control something else. For example, set the watt meter to close or open it's relay at low battery and it cuts off the motor controller. Or connect the relay to a buzzer that goes off when you reach a battery voltage. Look around and you can find this meter in up to 400 volts like mine is. It's a 2 piece design. There's a module you stick in the battery bay that incorporates the shunt and monitoring components and the display on your handlebars. If you want ultra compact, this isn't the best choice.

https://www.ebay.com/itm/300A-wireless- ... SwPDdbFPL2

Image

I have one of these watt meters, but I haven't used it yet on an EV. It the best of the TK and TF series watt meters and the best of the color LCD meters like above. There's several things about this meter that I really like. I have several of the TK and TF watt meters and they are well made, reliable and inexpensive. They do tend to be limited in functionality. This meter is the latest from these same people and it's got everything you could want from a watt meter. This Chinese company IMHO, makes good stuff already. I bet this meter meets my expectations and exceeds them too. Then they went on step further and eliminated the shunt. Instead it uses a coil that slides over the battery wire. I'm a stickler about eliminating electrical losses. I use very low resistance mosfets, over sized power and phase wires and eliminating the shunt is just one less electrical loss and eliminates 2 major battery connections. The LCD is easily read in the sun and it is fully programmable like the above meter. I think mine goes to 400 volts and 400 amps. Look around and you will find what you need. I think, I'll be using these meters a lot in the future. They even include a temp sensor!

https://www.ebay.com/itm/Battery-Monito ... OSwGwlaJuz~

OK...so then why do I want a watt meter? you want one for at least 3 reasons and there's more.
1. Battery voltage...this is the primary indicator of how much charge you have. 3 dumb LEDs on your handle bars tell you virtually nothing. An actual watt meter tells you the real voltage you are running at. Want to see battery sag in action? Watch the battery voltage as you crank the throttle.
2. Current draw...you can see in real time how much current you are using right now. This is useful information for several reasons. I've built 2 stand up scooters so far and I can tell you maintaining the same speed standing up and squatting down that current draw is significantly different due to aerodynamics. If your pack can handle 40 amps draw, how do you know you are pulling 40 amps? The watt meter tells you this. Motor to drive wheel gearing can draw significantly more or less current depending on how that ratio is set up. Go for the gear ratio that draws the least amount of current for the maximum amount of speed and torque.
3.Ah used...this is great for figuring out how far you can go on a charge or how much battery capacity you use for any given ride. Your Amp hours is important since it helps you track how far you can go. If you have a fixed distance ride you do every day, this is really helpful. Remember batteries have a limited number of charge cycles. You now have a tool that helps you estimate how often you need to charge. On my first stand-up scooter, I rode it 10 miles round trip every day. I forget my Ah usage now, but I worked it out so that I could go 4 whole days per charge or about 40 miles with a little to spare.
Last edited by ElectricGod on Jun 27 2018 9:32pm, edited 2 times in total.
ElectricGod...it's a joke...don't take things so seriously folks!

XB-502 Moped:
https://endless-sphere.com/forums/viewt ... 0#p1222730

Currie scooter:
https://endless-sphere.com/forums/viewt ... 7#p1227407

User avatar
ElectricGod   10 MW

10 MW
Posts: 2205
Joined: Nov 01 2015 5:24pm
Location: 3 OH 3

DC-DC converters or not

Post by ElectricGod » Jun 06 2018 4:51pm

I mentioned these devices inside a motor controller, but you can get bigger ones too which I'll go into a little later.

If you are running at 36 or 48 volts, there's a fair bit of lights and horns and what not for these voltages. I never build at lower than 66 volts so this is not an option for me. The KV limits of motors just mean the EV is going to be slow at 48 volts. Why would I want a slow EV? I don't and this build won't be slow either! I want good acceleration and I want a good top speed..such at 45mph on a 48" long scooter. That pretty much means better than 48 volts. When I first built the Currie, I had a 12S or 48 volt battery pack laying around, so I dropped it in the battery bay to test things out. The scooter accelerated pretty well, but then it topped out at 33mph...YAWN! Time for more voltage and 45mph so I went to 66 volts or 16S. Much better!

By now you are wondering what that has to do with a DC-DC converter. A LOT. If you are like me you don't want slow and boring. You want fast and powerful and that means a high voltage battery pack such as 66 volts or higher. Good luck finding lights and peripherals that work at better than 48 volts! However, there are tons and loads and zillions of options that work at 12 volts becasue every car and motor cycle on the planet runs at 12 volts. They are cheap and plentiful and the options that use LED's are truly VAST.

AKA...you want a DC-DC converter to run your lights and horn and other stuff. They take a range of input voltages such as 36-90 volts and output 12 volts. Depending on the current rating you need, the "box" varies in size from 2x1.5x1.5" to 2x2x4". I buy the below Chinese converters and they are cheap, work well and are reliable. All of my EV's have a DC-DC converter. Even at 48 volts, I recommend using a DC-DC converter. You need 12 volts out so ignore anything else.

I have several of these converters. They take a pretty wide input voltage range (48-120 volts) and deliver a decent amount of current (15 amps). I can run a lot of stuff from 15 amps!

https://www.ebay.com/itm/DC-DC-Buck-Vol ... SwWw1a6xBw

I'm pretty sure I have this converter in my blue stand-up scooter.

https://www.ebay.com/itm/DC-Dc-Converte ... Sw7bla2ZhK

Look around, if you pay more than $20 including shipping, you are probably over paying.

If you need 5 volts, there's lots of small boards that take 8-40 volts in and output 5 volts and might even have a USB port for powering your phone or other 5 volt items. Of course now that you have 12 volts on your EV, any cheap cigarette lighter USB charger will work too.

To me, the world of options and devices that can run at 12 volts is sooooo worth having a DC-DC converter. Look at any of my EV's. They have LED lights all over them that cost virtually nothing since they are readily available in 12 volts.
Last edited by ElectricGod on Jun 27 2018 9:37pm, edited 1 time in total.
ElectricGod...it's a joke...don't take things so seriously folks!

XB-502 Moped:
https://endless-sphere.com/forums/viewt ... 0#p1222730

Currie scooter:
https://endless-sphere.com/forums/viewt ... 7#p1227407

User avatar
ElectricGod   10 MW

10 MW
Posts: 2205
Joined: Nov 01 2015 5:24pm
Location: 3 OH 3

LED lights

Post by ElectricGod » Jun 06 2018 6:01pm

You want to be visible. EV's are quiet and fast...or can be...so you want to be seen. LED's are a great way to throw some light out there day or night while consuming very little battery power. Don't ever think about light bulbs for anything. If your EV has light bulbs like the XB-502 moped did, consider replacing them with LED's. The LED's will be brighter and consume less power than the light bulbs ever will and they last for many years. Cheap LED's are a good option, but some LED stuff from China is way over exagerated and some stuff is pretty close to dead on.

I buy a lot of LED products from China. Some is just crap and some is pretty good and the prices for one or the other can be nearly identical.

If you find LED strips, they are usually reasonably bright and ridiculously cheap. I buy this stuff purely on price. Can you get much brighter LED strip lights? Sure you can, but then you pay 5-10X more money for it too. Why bother for 20% more light? In the USA, some colors are reserved for specific things. Yellow or orange can be used universally anywhere on a vehicle, so I use yellow strip LEDs on my sides and as accent lights to make me more visible. You can get strips of this stuff on ebay for uber cheap. Cut it to the length you need, solder together sections to make a longer strip...whatever, this stuff is uber cheap and works pretty well.

https://www.ebay.com/itm/5M-16-4ft-Ambe ... 0020!US!-1

2 LED COB modules. These are generally referred to as COB led's. Usually you get a red section and a white section or white section and a yellow section. They are arrays of tiny LEDs on an aluminum sheet. Use one section for side lights or tail lights and the other for directional lights. There are a zillion styles of these things for super cheap. I use them on all my EV's. They throw a lot of light that spreads quickly. This makes you highly visible to others, but don't do much to light up the road for you.

https://www.ebay.com/itm/2pcs-12V-Car-L ... SwjzJa1sHG

https://www.ebay.com/itm/2Pcs-DRL-LED-D ... SwRyRa-0NN

You can see one of these COB LED modules on the down tube of the Currie. It doesn't direct light in a specific direction, just throws it everywhere. Because of these lights, I've had several drivers start pulling out or come into my lane and then they see the lights and stop. Oh look, my amber strip LEDs are on too!

Image
Image

The bigger blue scooter has a set on either side of the back wheel. I get seen with these lights on!

Image
Image


This is something you will see on both of my scooters. They are a specialized LED strip light that contains 2 LED colors and 4 sections. They come in lots of different lengths. I buy on price and the length I need. If one strip is not bright enough, use several in parallel. I use 3 or 4 to get the tail end brightness I want. They include 2 sets of red LEDs for tail lights and brake lights. The brake lights are 2X brighter. They also have sections on either end useful for directional lights. Just stick them down to any clean surface.

https://www.ebay.com/itm/12-Bendable-Un ... 0020!US!-1

This is the tail and brake lights on the Currie.

Image

Right directional.

Image

Lets talk about head lights. This is a wide open field. Now you are talking about massively exaggerated lumen ratings. Everything up to now, works as advertised. In LED head lights, the Chinese do NOT know what a lumen means. Quite often they will raite something like 5000 lumens and it's really 150 lumens...truely buyer beware! But, I've found a few products that are decent.

This is a Chinese XHP70 LED light. The real CREE XHP70 light is rated for like 2000 real lumens and it is much brighter than these Chinese LED's so there is no way this is a 5000 lumen light. You are looking for 3 things here. All aluminum shell, glass lense and aluminum reflector. If the seller can't tell you what the construction is, keep looking. This is an item that you should buy somewhat on price and somewhat on making sure it is legit. I've bought 2 or 3 of this style of head light and gotten a plastic shell with a tiny LED inside becasue I bought the $3 item instead of the $23 item. They looked the same in the pictures. Silly me! you can find this lamp for a lot less than $50.

https://www.ebay.com/itm/Cree-XHP70-XHP ... 362b690923

This is one of my all time favorite lights. It's compact, well made, all aluminum construction and is a legitimately bright Chinese XHP70 LED. This seller says 2800 lumens...uh yeah right...but not ridiculously over the top with the lumens rating. I bought 2 of these like a year ago for $12 each and when they arrived, I was immediately impressed with the build quality. It's not a legit CREE XHP70 in there so it's not going to be as bright as the real part, but they are plenty bright enough and for the price...not bad at all. Look around, this exact style comes in 5 or 6 shell colors. I think I have 6 of these lights in black or silver. That should tell you a lot about how much I like them.

https://www.ebay.com/itm/1pcs-Waterproo ... SwMm1a3yac

This is the same light in a little less wattage in a bunch of colors. Just get the brightest version. It will have a Chinese XHP70 LED in it, but it's still pretty bright. This style light used to be super common on ebay. They have somewhat less availability now and the price has gone up a lot since I bought mine. Still a very good light for the price.

https://www.ebay.com/itm/1xCar-Motorcyc ... 0005.m1851

This is a great LED head light wit ha couple of oddities. Again Chinese LED's are not as good as the legit CREE XML LED's, but it's still pretty bright. The build quality is pretty good. This is the light that I found for $4 and got the plastic version. Expect to pay about $15-20 for the good version. The weird part is they run on 8.4 volts. That's 2 lithium cells in series. The light itself is well made, but you can't run it on 12 volts. You will need another DC-DC converter to drop down to 8.4 volts. Inside the shell is lots of empty space to tuck one inside. I've done this on the 2 I have. 32000 lumens...really? No not even close! More like 1000 lumens at most. It's still a lot of light.

https://www.ebay.com/itm/32000LM-Bicycl ... 0005.m1851

Castle Creations makes some nice programmable BEC's that work really well for driving LEDs. I have several of these BECS in use in several LED lamps. One of these BEC's will drive a legit XHP70.2 LED just fine while over watting it.

https://www.ebay.com/itm/Castle-Creatio ... 0020!US!-1

If you just can't get enough of LED stuff, here's a thread with random EV related stuff I've done. You'll see several posts on various LED projects I've done. I enjoy modding things and making them do far more than they originally were capable of. I mod just about everything. There's quite a few posts in this thread about LED mods I've done. The last post is for a tiny keychain light I bought on ebay and modded. It hangs with my keys all the time and I used it just last weekend. Every time I turn that thing on, I'm just stunned at how bright that XPL2 LED is!

viewtopic.php?f=10&t=91589&p=1335369#p1335369
ElectricGod...it's a joke...don't take things so seriously folks!

XB-502 Moped:
https://endless-sphere.com/forums/viewt ... 0#p1222730

Currie scooter:
https://endless-sphere.com/forums/viewt ... 7#p1227407

Addy   100 W

100 W
Posts: 193
Joined: Aug 11 2017 11:15am
Location: BC, Canada

Re: LED lights

Post by Addy » Jun 06 2018 7:10pm

ElectricGod wrote:
Jun 06 2018 6:01pm
The weird part is they run on 8.4 volts. That's 2 lithium cells in series. The light itself is well made, but you can't run it on 12 volts. You will need another DC-DC converter to drop down to 8.4 volts.
Have you tried running it with 12V? I bought a similar light from amazon which came with a 2s2p lithium pack. I tested it with my bench supply, slowly raising the voltage up to about 13V. It still worked totally fine, the wattage remained about the same. I use it on my ebike and I bought another for my friend's ebike.

http://www.amazon.ca/dp/B00CGUL65U

User avatar
ElectricGod   10 MW

10 MW
Posts: 2205
Joined: Nov 01 2015 5:24pm
Location: 3 OH 3

Throttles and switch clusters

Post by ElectricGod » Jun 06 2018 7:49pm

You are going to need a throttle for your EV. There are some decent and cheap throttles and some horrible and cheap throttles. I'll present the ones that I own and like and ignore the rest.

Before I use a throttle, I think about what it will go on. For something stand-up a thumb throttle is your best bet. The reason why is you are planted to the scooter by 2 things, your feet and hands. The more stability you have the better and on an over powered scooter like the Currie or what this Schwinn will be, you definitely want a good solid grip on the handlebars. A twist throttle, means your right hand is not planted on something that doesn't move. On my blue scooter, I started with a twist throttle. With the original set up it was weak enough that this didn't matter a whole lot. Then I upgraded to 2X the power and speed and then a twist throttle was a hazard. I want to be able to hold on to the grips solidly with no movement. My right thumb compromises my grip on the handlebars the least so a thumb throttle is pretty ideal.

This is the only thumb throttle I am currently willing to buy and as you can see it is really cheap. Despite that, they are pretty nicely made and I have 4 of them and given a couple away too. The LED's are the reason for the voltage. They are set for 36 or 48 volt battery packs. The top just pops off and then I replace the LEDs with a small 100 volt LED volt meter. I have no use for a couple of dumb LEDs to show me battery voltage!

https://www.ebay.com/itm/36V-Twist-Thro ... 0020!US!-1

This isnt the right LED meter since it's the .36" size, but you want the 3 wire meters since they work up to 100 volts. The 2 wire meters work up to 40 or 50 volts. You want the .28" size. They are a perfect fit inside the top of these thumb throttles.

https://www.ebay.com/itm/Mini-DC-0-30V- ... SwOgdYuNvr

You end up with something like this You can see the throttle in the right side of the picture on the Currie. And yes, there's lots of stuff crammed in here. Far left is a tire pressure monitoring unit. The sensors screw onto the valve stems on the tires. you can't really see them, but there's a couple of crappy switch clusters hidden by my phone that will get replaced with the below modded 7 switch cluster. The phone is mounted on a quad lock. The snap in back on the phone will come off before the quad lock lets go! Next to that is a TK15 watt meter. Above that is a GPS Bryton Rider 530 speedo. It has a twist lock mount. I pull it off whatever EV it is currently on and drop it onto all the others. One speedo does all my EV's. Next to the watt meter is a 1080P bullet camera and next to that is the thumb throttle.

Image

People will look at my modded thumb throttle and say "Dude! you just built one of these (see below URL). While it is similar, the below throttle is poorly implemented, wobbly and doesn't work as nicely. I bought one and was really disappointed. The above thumb throttle is far better IMHO and I love modding so adding an LED meter to it is something I enjoy.

https://www.ebay.com/itm/36V-48V-60V-72 ... Sw1xpa1tGJ


Full grip twist throttles:

In twist throttles you have several options that I have personally tried and like that are cheap. I use twist throttles on sit down EV's such as mopeds, but NOT on stand-up machines. These are a few that I like a good bit and the build quality is pretty reasonable.


I have one of these that I use as a bench throttle. It has worked well for a number of years. mine has a rocker switch on it instead of the key switch. The 3 wire LED meter inside is good to 100 volts. I don't know why this auction says 72 volts.

https://www.ebay.com/itm/72V-60V-48V-36 ... 0005.m1851

This throttle and the above throttle are virtually identical. You get some variation in the grip style and a switch or key switch or a 3-way switch, but these throttles with a LED volt meter in them are effectively the exact same thing. They work well enough and I own several.

https://www.ebay.com/itm/12V-84V-Twist- ... SwCsBa-mQo

https://www.ebay.com/itm/E-Bike-Scooter ... SwGp1a5tF8

If the throttle has a few LED's that show you 3 or 4 levels like this, I just walk away. Give me a real voltage meter or don't bother at all. BTW, this is a crap throttle for other reasons, so never mind on multiple counts.

https://www.ebay.com/itm/48V-Ebike-Twis ... SwWGxazuPE

Since I use a watt meter and I strongly recommend you do too, there's really no need for a volt meter on the throttle. I've purchased a few "plain Jane" throttles that I like.

I have this exact throttle on the XB-502 moped. It has no bells or whistles...just a throttle and looks pretty. It operates smoothly and makes 1/4 turn.

https://www.ebay.com/itm/Ebike-Throttle ... Sw8IJZjLZd

I have one of these throttles. The throttle works great. I can't complain about smoothness or operation, but the 2 switches laid out like they are makes for this odd looking thing hanging down off the throttle. Still, if you need controller enable and 3 speeds on the throttle, this works pretty well for that use. I wish the writing on the switches didn't exist and instead they said 0/1 and 1/2/3 or something else a bit more universal than Chinese squiggles.

https://www.ebay.com/itm/Speed-Control- ... FvmHdYD-4g


Enough of throttles...Onto switch clusters.

In a switch cluster I want 3 things as a minimum.
1. Head light switch that can handle some decently powerful LED lamps.
2. Directional switch that is very clicky and tactile so I can feel it strongly go into left, off or right positions without looking down.
3. A decent horn button that my left thumb just falls on without needing to look for it.

This is the only 3 switch cluster I will ever buy from now on. It has a nice metal shell and is very small. I haven't seen one, but beware of someone duplicating it in plastic! I have probably 10 other versions of 3 switch clusters. They all work, but pretty much suck by comparison to this one. Get this switch cluster and don't look at anything else. Trust me on this...this is the one you want! I have various other 3 switch clusters on my EV's and pretty much hate them...mostly for the directional switch and pathetic head light switch. This one just destroys everything else. For people that need headlights, directionals and horn, this is the best option and it's cheap. If I have a complaint about it, it's the toggle for the head lights is so close to the horn button. The directional switch is uber clicky which is perfect. You can easily feel the 3 positions without ever looking down.

https://www.ebay.com/itm/7-8-22mm-Alumi ... 0#viTabs_0

If you are wanting just a few more switches and don't mind modding a bit, then this is a good option. You can find them with an aluminum or plastic shell in 7 switches. The unfortunate thing about them is the plastic version looks identical to the metal version so beware. The metal shelled ones are a far better switch cluster than the plastic shelled ones. I have both and know this for fact. In either shell only one button locks. This switch cluster was intended to control a motor cycle computer, not turn on lights and other peripherals. You need to get locking switches and toggles to turn it into something useful on your EV.

https://www.ebay.com/itm/Universal-Moto ... SwBp9abtkI

These little locking switches are uber cheap on ebay. They are rated for .5 amps so double up the 2 sets of contacts to get 1 amp out of them. They are a perfect fit for the plastic gromets that hold the switches.

https://www.ebay.com/itm/20X-8-5mmx8-5m ... xyI8lR-dYN

Image

These toggle switches provide 3 positions or ON-OFF-ON. For lights, double up the contacts so you have 2X current handling or 6 amps. They are pretty clicky and tactile so for a directional switch, they are not too bad and they work well as a 3 speed switch too. They are a tiny bit snug in the plastic switch gromets, but they do fit.

https://www.ebay.com/itm/10pc-Sub-Minia ... xy4t1SiNW4

Image
Image

This is the insides of these switch clusters after modding.

Image

I kept one momentary push button for the horn. Commonly that's the green button so I used the green switch cap for the lower position on top. The rest of the buttons I replaced with locking switches. The toggle is for directional lights.

Image

I later got a couple more of them and modded them a bit differently so that I had a second 3 way switch for 3 speeds. I only needed 3 locking switches for all my lights (headlight, side lights and battery box lights) so that left me with 2 button switches that I didn't really need. One got replaced with the second 3 way toggle. Later I realized that one of them could be used at the handle bars to turn on and off a USB power port. That used up the one remaining push button locking switch. One of the switch clusters was advertised as aluminum, but I got this plastic switch cluster instead. GRRR! Be sure you get what you are expecting to get! OH well, I was giving it away to a friend anyway and the ebay seller has refunded me my money. The aluminum shell is a lot nicer than the plastic shell and it's a bit smaller too.

Image

In the above switch cluster, green is horn.The yellow toggle is directional lights. The yellow button is head lights. Red is side lights. Red toggle is low, medium and high speed. Blue is battery bay lights and grey is aux (USB power). The metal version I received will go on the Currie and the plastic one will go to my friend. I'll get another of these switch clusters to put on the Schwinn S1000 and do this exact thing to it. I might replace the yellow button with another small off-on toggle so I have more amperage for the head lights.
Last edited by ElectricGod on Jun 27 2018 10:06pm, edited 1 time in total.
ElectricGod...it's a joke...don't take things so seriously folks!

XB-502 Moped:
https://endless-sphere.com/forums/viewt ... 0#p1222730

Currie scooter:
https://endless-sphere.com/forums/viewt ... 7#p1227407

User avatar
ElectricGod   10 MW

10 MW
Posts: 2205
Joined: Nov 01 2015 5:24pm
Location: 3 OH 3

Chain and sprockets

Post by ElectricGod » Jun 06 2018 9:05pm

Good grief! Yes more stuff to talk about.

There are a few chain and sprocket options that work well for EV's.

If you are on a serious budget, I bet your small EV already came with T8F chain and sprockets. Reuse them, but keep in mind that at about 2000 watts at the motor, the chain breaks pretty consistently. It's cheap chain. What do you want? Decent grade bike chain is better and tougher!

It's just a personal opinion that will make lots of people upset, but IMHO, use bike chain for pedaling and use something much better for driving the motor. Really good bike chain can't handle 4kw and certainly the sprockets can't either. Just assume if you are doing a bike build that the below are true for you.

1. You want regen and you don't want to use a hub motor.
2. You want to be fast and powerful and run at full torque from the motor.
3. You want a mid drive motor.

Here's a side note that will annoy lots of people. I really dislike hub motors. They are heavy and they are loads of unsprung weight inside the wheel which is BAD. All that weight has to be made to spin, this eats up motor torque that should be making you go faster. Having said that, I own 5 hub motors of which 3 will see serious use...eventually. One will see 20-40kw constantly and the other 2 more like 5-10kw. Other than a few specific hub motors, most IMHO are NOT worth having. Oh and lets not forget that you can't change the gearing of a hub motor. Gearing in a hub motor comes down to 2 things. Change the battery voltage or use a different sized tire...of which neither are gear ratios. Some people swear by hub motors, I'm not one of them. Did I mention they cost a lot and weigh a lot for the wattage you get? As a general rule, just avoid hub motors. I can feel the ire and heat from a few folks already!

For everybody else that isn't building something that can be pedaled, you probably won't be using a hub motor and you will be using regen and you want to run at full power without breaking something.

Everyone, say hello to KART or 219 sprockets and chain. The sprockets come in any size you can imagine and they are built for racing so they are pretty well made and well balanced. BUT they are uber common in a zillion versions so they are very inexpensive. If you spend $40 in sprockets for your EV, that's about right. 219 chain is capable of handling 10kw or 13 hp...that's a lot on a small EV!!! I have yet to need something stronger than 219 for anything. The chain comes in 3 versions...cheap, medium and expensive. All 3 handle the same amount of loading. What's different is how freely the chain moves and if it has seals inside each link. The cheap stuff will handle just about anything you need to do, but it's a bit noisier than the better stuff. I buy the medium grade 219 and it works well. All told you are going to spend about $80 between chain and sprockets for a solution that will last virtually forever and never wear out. Just oil the chain periodically.

Here's a few options found on ebay.

This style sprocket has a standard 6 bolt pattern inside that is identical with just about all KART sprockets out there. There's split sprockets for the back wheel too, don't get them, they are not intended for long term use. The single piece ring sprocket is your best bet. Steel or aluminum wheel sprockets both last a very long time.

https://www.ebay.com/itm/RLV-219-Alumin ... Rh9fYIP35w

Your back wheel already has some kind of sprocket on it. This sprocket is ideal for using as an adapter for the KART sprocket. It hopefully already bolts to your back wheel, Drill 6 holes in it that match up the KART sprocket, bolt them together and you are good to go.

219 motor sprockets usually come with a 3/4" ID and a key way. Probably your motor has some other smaller size for the shaft diameter. The best bet is to buy some keyed 3/4" shaft and drill out the center to exactly your motor shaft diameter. Here's keyed shaft on ebay.

https://www.ebay.com/itm/KEYSHAFT-3-4-G ... SwNchaCpjr

And a few 219 motor sprockets. They are also commonly referred to as driver sprockets. They only come in steel that I've ever found. There's lots of force on those very few teeth. This auction doesn't say, but that ID is very likely to be 3/4"

https://www.ebay.com/itm/Racing-Kart-21 ... njkbSWF95g

Asuza Engineering is a good place to get 219 driver sprockets. That's all I have. They make good stuff.

http://azusaeng.com/product/b-type-sprockets-219-chain/

You are going to want to get a 219 chain breaker. I recommend this one specifically. 219 chain is very tough stuff and it's precision. A bike chain breaker is probably going to break before the chain comes apart.

https://www.ebay.com/itm/RLV-219-Chain- ... 0020!US!-1

Now you have all the parts to make your chain line robust, tough and capable of whatever you want to throw at it.

Here's a few pics of 219 in action on my EV's. This stuff is very quiet, strong and reliable. The first 2 pictures are from my blue scooter. The chain line has about 6000 miles on it. The motor sprocket shows some wear, but I bet it has another 8000 miles in it. The other 2 sets of pictures are from the Currie and the XB-502 moped. I got those chain lines a bit straighter and they are much newer so there will be hardly any wear at all.

Image
Image
Image
Image
Image
Image

A lesson learned...
219 has no side to side slop in it. Bike chain by comparison is like bending a length of copper wire! When using 219, it is very important that both sprockets are dead straight in line with each other. An early mistake I made on the blue scooter was having the motor sprocket turned a little out of line with the wheel sprocket. I was off just a little bit. The motor sprocket had maybe 20 miles on it and the chain was eating up the sides of the teeth. I had probably another 30-40 miles left in that sprocket before it was completely destroyed by the chain. I replaced the motor sprocket and got the chain line dead straight and since then it has been pretty trouble free and 6000 miles later still going strong. I made sure on the Currie and XB-502 to get the chain line dead straight. This also makes the chain a lot quieter. Not that 219 is noisy...it not. But less noise is always better.
Last edited by ElectricGod on Jun 27 2018 10:35pm, edited 6 times in total.
ElectricGod...it's a joke...don't take things so seriously folks!

XB-502 Moped:
https://endless-sphere.com/forums/viewt ... 0#p1222730

Currie scooter:
https://endless-sphere.com/forums/viewt ... 7#p1227407

User avatar
ElectricGod   10 MW

10 MW
Posts: 2205
Joined: Nov 01 2015 5:24pm
Location: 3 OH 3

Basic planning

Post by ElectricGod » Jun 06 2018 9:24pm

What's next? Getting parts together, deciding what you want to do and what your requirements are. I plan in my head usually and only plan on paper for things that I really need to draw out or it's too complex to remember. Schematics go down on paper. Some parts get drawn up or mocked up in cardboard.

I'm assuming for my expectations and goals. Yours might be different. I'm planning to build a better EV than my Currie which is pretty awesome! To improve it, means fixing a few things, more power and rear suspension. Otherwise, I'm pretty happy with how it works right now.

1. I want lots of torque so that I can accelerate as well as cars can. I want to be able from a dead stop to climb a steep hill (20 degrees or more).
2. I want good top speed...45mph or more so that I can keep up with the cars. All of my EV's can do both of these.

3. I want range...at least 24 miles while maintaining 80% of full speed. More range is better, but if I have to ride slow to get the range, then I want a bigger battery to get me range and speed at the same time. Yes a Tesla can drive 600 miles on a charge if it goes 25mph. BORING!!!

4. I want status of the EV on my handle bars. That means a watt meter and temperature meters for the motor and controller.

5. I want it to be reliable. That means good electronics and mechanics that will hold up well under daily use.

6. I want to be visible and to be heard. Lot's of LEDs and a good horn

7. I want a stable EV that rides equally well at slow or fast speeds and stops well.
ElectricGod...it's a joke...don't take things so seriously folks!

XB-502 Moped:
https://endless-sphere.com/forums/viewt ... 0#p1222730

Currie scooter:
https://endless-sphere.com/forums/viewt ... 7#p1227407

User avatar
ElectricGod   10 MW

10 MW
Posts: 2205
Joined: Nov 01 2015 5:24pm
Location: 3 OH 3

Wheels and tires

Post by ElectricGod » Jun 06 2018 9:47pm

I should probably talk about wheels and tires before moving on. If you look at any of my EV's they don't use the factory wheels usually and the tires are probably replaced too. Why would I bother?

1. Cheap tires don't grip very well. I want to stick to the road as I rocket around doing 40-60mph.
2. Cheap tires tend to be out of balance. At slow speeds you won't notice wheel balance issues, but get moving a bit and you lose control-ability due to vibration and wobbling.
3. Cheap wheels wobble and are not balanced. This creates ride and control issues.

On the blue scooter I was doing a speed test. I had the factory tires and wheels still. I got to 55 mph and the scooter was wobbling and weaving around. I pushed through it to 59.5 mph, but it got so bad that I was on the verge of wrecking. I then gradually slowed down and things smoothed back out at around 50 mph. That was 100% dangerous! I soon after bought some cast wheels and Kenda tires. That fixed 90% of the speed wobble and vibration issues I had previously. I had gotten the factory wheels and tires balanced at a motor cycle shop, but that wasn't enough. The wheels were not perfectly round and the axles were not perfectly perpendicular to the rim. It was just cheap garbage that won't get used ever again. The next speed test suffered none of the wobbling and weaving issues.

On the Currie, I originally built around the factory spoked wheels. They were round so they should have been OK. Those wheels were never intended to handle 4kw or 45mph. Both wheels broke spokes. The back wheel from motor torque and the front from braking torque. I found some cast wheels in the right size to replace them. This scooter was never intended to see what I did to it. Tires were a big issue. I knew immediately the factory tires would not hold up. I ended up buying 3 different sets of tires before getting some that could hold the speed, grip and load I was putting on them.

The XB-502 factory back end included a really weak hub motor. There was no way I was ever going to use it. I bought a cast wheel very similar to what is on the blue scooter, just the next size larger. The front wheel was cast and of reasonable quality and pretty well balanced. I reused it. Both front and rear wheels got new tires and are tubeless.

Wheel weight makes you slower, eats up power and can be reduced. Anything spinning has to be kept spinning. This is a significant down side to hub motors. They add a lot of weight inside the wheel that has to be kept spinning. A chain and sprocket weighs much less than a hub motor so it's a lighter spinning mass and requires less energy to get it spinning. There's another spinning mass. The tube inside the tire. If you've already upgraded to cast wheels, you might as well eliminate the tube and go tubeless. If you have issues with air leakage, add a little goop inside the tire. It will migrate to the leak and seal it up. Be sure to spin the tire for a couple of miles right after adding goop to get it evenly distributed. The Currie has small tires so they hold very little air in them. I have TPMS on those tires just in case I get a leak. They will lose air very quickly and don't have a lot to spare air inside. The XB-502 moped is also tubeless. I wish I had gone tubeless on the blue scooter! I saved .3 pounds per wheel in the Currie. In the XB-502 I saved about .6 pounds per wheel. I also reduced the amount of wobble and imbalance thanks to eliminating the inner tubes. The XB-502 is glassy smooth at 60mph. The Currie since it has no rear suspension and the tires are smallish is a bit rougher of a ride. At 45 mph is very stable and there is no hint of loss of control.

Conclusion: Get good wheels and tires and eliminate the tubes. Make your wheels as light as possible without loss of reliability. Good balanced wheels and tires are your best friend.
Last edited by ElectricGod on Jun 27 2018 10:39pm, edited 1 time in total.
ElectricGod...it's a joke...don't take things so seriously folks!

XB-502 Moped:
https://endless-sphere.com/forums/viewt ... 0#p1222730

Currie scooter:
https://endless-sphere.com/forums/viewt ... 7#p1227407

User avatar
ElectricGod   10 MW

10 MW
Posts: 2205
Joined: Nov 01 2015 5:24pm
Location: 3 OH 3

Range vs weight vs aerodynamics vs power used

Post by ElectricGod » Jun 08 2018 7:01pm

I don't know if anyone is following this thread or not or if they just stop by out of curiosity, but i'm going to continue anyway...

Lets talk about estimating range. That's essentially this...how much battery capacity do I need to carry me X miles.

This won't sound particularly scientific. LOL! I take an educated guess. I've built a few EV's now and have a fairly good idea based on my 240 pounds and the wattage and Ah consumed how far I can go. I estimated that the Currie would weigh around 70 pounds and in fact it's 75 pounds when completed. When I built the Currie with about 4kw max power used, I expected that 20Ah of battery pack would get me about 20 miles of range. I was pleasantly pleased to get 24 miles of range while maintaining about 80% throttle the whole way. Keep in mind standing vertically on anything is a giant wind dam and probably the worst aerodynamics possible. The only thing worse is attach an open parachute behind you. A bike with the rider leaned over the handle bars will reduce wind drag considerably and therefore ought to increase your range a fair bit. Running at lower speeds (not 35-45mph all the time) will aerodynamically save battery capacity as well since you are ramming through the air less. The blue stand up scooter weighs 120 pounds and has about 35Ah of battery capacity and got me 45 miles of range. The scooter itself is less aerodynamic and then I'm standing on it being a giant air dam and then it weighs a lot more and I still got 40+ miles range while consistently doing 35+ mph all the time.

Yes...that's not very scientific, but it will give you a good idea of what you can expect.

75 pound EV with 240 pounds of rider that is standing vertical with a 20Ah pack at 66 volts with 3000-4000 watt load gets me 24 miles of range on mostly level surfaces.

120 pound EV with 240 pound rider standing vertical with a 35Ah pack at 82 volts with 3000-3500 watt load got me 40 miles of range.

When I built the the XB-502 since it's about the same size as the blue scooter, I used that as my comparison for estimating range. I wanted less weight, better acceleration and better top speed at 82 volts while using the exact same motor. The moped weighs about 100 pounds (not lots less weight). It accelerates far better that the blue scooter and tops out at 60 on level ground. The blue scooter needs a hill to go down to get there. It has 32Ah of capacity at 82 volts. It's also much more aerodynamic than the blue scooter. I'm seeing about 5kw and about 60 amps. Based on those numbers, I expect that I'll see about 25 or 30 miles of range while riding at 40-45mph all time. 60mph draws about 60 amps. 45mph draws about 40 amps. Any more range beyond that will be great.

You might be building a 40 pound scooter and 20-30Ah is just huge for you and you need to top out at 30mph and go no more than 10 miles. 2kw is loads of power for you. 10-12Ah of capacity is going to get you probably 15 miles of range.

You might be building a full sized moped such as the RMartin I'll be starting on soon. From the factory, it had 400 pounds of SLA batteries in it. That put the whole EV at about 550 pounds. It's a moped, but it's every bit as large as a full sized motor cycle. 32Ah won't get it but 15 or 20 miles at most. I'm starting at 70Ah which I'm estimating will get me above 40 miles of range despite the greater weight. I'll be able to maintain 60mph or more. With that all in lithium, the battery weight ought to be far less than half of the SLA's. I will work towards less than 200 pounds for the EV. There's a lot of junk that can get stripped off of it to reduce weight even more. It has decent aerodynamics which will help a lot.

A friend of mine rides a KMX trike (tadpole). It weighs about 100 pounds and he weighs about 200 pounds. He has made it very aerodynamic and he sits low to the ground. I don't know his actual amps used for max acceleration and top speed, but I know that he uses about 20 amps at 40 mph at 82 volts on level ground. Wind resistance is a big source of power consumption. The XB-502 uses 2X that much amperage to maintain the same speed...mostly due to wind resistance. IF it was low to the ground and super sleek like my friends trike is, I bet I could lower my amp draw considerably. He has skinny tires and covers over his spokes. Everything...including him is inside cowlings that reduce his wind drag. The underside of the trike is flat and smooth. None of my EV's take advantage of good aerodynamics!

I know none of this is "scientific" or exacting, but you get the idea on what capacity to plan around to get the range, speed and acceleration that you want. If you are looking for exact numbers, please look elsewhere. I prefer to build around estimations and then test to see what I actually get. So far I've used this method for a few friends and my builds and the estimate is always a little conservative. IE: what I say ought to be the range is just a little less than what the actual range is.
Last edited by ElectricGod on Jun 27 2018 10:48pm, edited 3 times in total.
ElectricGod...it's a joke...don't take things so seriously folks!

XB-502 Moped:
https://endless-sphere.com/forums/viewt ... 0#p1222730

Currie scooter:
https://endless-sphere.com/forums/viewt ... 7#p1227407

User avatar
ElectricGod   10 MW

10 MW
Posts: 2205
Joined: Nov 01 2015 5:24pm
Location: 3 OH 3

Schwinn S1000 build

Post by ElectricGod » Jun 08 2018 7:24pm

I found the best forks made for the S1000 scooter. They are hard to find and therefore NOT cheap...$127. Never the less, I got a set that take front disk brakes and have suspension. They will arrive in a few days. Just buying forks, I've almost doubled the cost of the scooter.
ElectricGod...it's a joke...don't take things so seriously folks!

XB-502 Moped:
https://endless-sphere.com/forums/viewt ... 0#p1222730

Currie scooter:
https://endless-sphere.com/forums/viewt ... 7#p1227407

Spencnor   1 µW

1 µW
Posts: 3
Joined: Feb 14 2018 2:24am

Re: Range vs weight vs aerodynamics vs power used

Post by Spencnor » Jun 09 2018 1:30am

ElectricGod wrote:
Jun 08 2018 7:01pm
I don't know if anyone is following this thread or not or if they just stop by out of curiosity, but i'm going to continue anyway...
Please continue! I am learning a lot from your very informative posts. This should be a “sticky” for newbies to learn the basics. Thank you for taking the time to share your knowledge. :D

User avatar
ElectricGod   10 MW

10 MW
Posts: 2205
Joined: Nov 01 2015 5:24pm
Location: 3 OH 3

Re: Range vs weight vs aerodynamics vs power used

Post by ElectricGod » Jun 10 2018 2:07pm

Spencnor wrote:
Jun 09 2018 1:30am
ElectricGod wrote:
Jun 08 2018 7:01pm
I don't know if anyone is following this thread or not or if they just stop by out of curiosity, but i'm going to continue anyway...
Please continue! I am learning a lot from your very informative posts. This should be a “sticky” for newbies to learn the basics. Thank you for taking the time to share your knowledge. :D
you are welcome! If it gets to be a sticky or not...I don't get to choose that. The idea was to show the basics and a lot of stuff that you would otherwise find out along the way via trial and error, research and blowing stuff up. Hopefully putting this information in a single thread like this will help a lot of people.
ElectricGod...it's a joke...don't take things so seriously folks!

XB-502 Moped:
https://endless-sphere.com/forums/viewt ... 0#p1222730

Currie scooter:
https://endless-sphere.com/forums/viewt ... 7#p1227407

User avatar
ElectricGod   10 MW

10 MW
Posts: 2205
Joined: Nov 01 2015 5:24pm
Location: 3 OH 3

I wish I had a decent TIG welder

Post by ElectricGod » Jun 10 2018 2:29pm

I have wanted to get a welder for a long time. However, I don't want to settle for something cheap. I want a decent TIG welder so that aluminum, stainless steel, brass...just about any metal I come across can be welded. I'm still working on saving for the right welding machine which won't be cheap! Maybe in a few more months.

In between time, since I don't have a welder, I have to do things like buy these forks for the S1000. The forks on the scooter don't take a disk brake caliper. They are wide enough for a disk brake, but the front wheel has the less reliable and effective rim clamping brakes. If I had a decent welder, I could have taken apart the factory forks and added the little bit of metal seen in the third picture to convert them to disk breaks. At the same time, I would have ground off the mounts for the rim clamping brakes. Otherwise these 2 sets of forks are close to identical.

Image
Image
Image
Last edited by ElectricGod on Jun 27 2018 10:50pm, edited 1 time in total.
ElectricGod...it's a joke...don't take things so seriously folks!

XB-502 Moped:
https://endless-sphere.com/forums/viewt ... 0#p1222730

Currie scooter:
https://endless-sphere.com/forums/viewt ... 7#p1227407

User avatar
ElectricGod   10 MW

10 MW
Posts: 2205
Joined: Nov 01 2015 5:24pm
Location: 3 OH 3

Forks, bearings, tubes, tires and deck stuff

Post by ElectricGod » Jun 10 2018 5:36pm

I figured since I was talking about the forks, that I would put them on the scooter.

A few comparison pictures. Other than the way the brake mounts and some very minor differences, they are pretty close to the same part.

I bought the replacement forks from here:
http://wildscooterparts.com/currie-e1000-fork.html

Image
Image

The new forks have this sticker on them. I looked up www.hlcorp.com (ZOOM HL Corp). They make various bike components, kick scooters and wheel chairs.

Image

You don't get new bearings and seats with the new forks so be sure to pull them off the old forks. Tapping on a screw driver around the edges of the seat will drive it off the old forks. Just be sure to drive it off evenly and straight. It's just a tension fit on the steering tube and comes off pretty easily.

Image
Image

Now the lower bearing seat is in place on the new forks. I put some grease on the bearing seat and then dropped the bearing in place and put a bit more grease on top of it. This bearing doesn't move a lot...just 90 degrees of rotation so grease is not super important that it gets all embedded in the bearing. A little dribbled on the outside of this bearing is plenty.

Image

Same for the top bearing. I put some lithium grease in the bearing cup, dropped in the bearing and then put a bit more grease on the top bearing nut/seat.

Image

On a side note...this is a 4" magnet dish. These things are very handy. I bought 3 of them on amazon for less than $10. If you don't have any, I really recommend getting a few of them. You are going to take screws and bolts and other small parts out of things. These dishes keep your loose parts all together so they don't get lost. In this picture, I have the dish tilted about 40 degrees and everything just stays put.

Image

Back to the forks...
When you tighten up the nut, only tighten enough to remove all the slack. You can feel the nut cinch down on the bearings. This is probably a little too tight. If you try to turn the forks, you might feel resistance. They should turn super smooth. Loosen the nut a little, but not enough that you create wobble. There's a jam nut that sits on top of the bearing nut. When the two nuts get locked together, they will probably drive together a little tighter so getting the bearing nut just barely tight is important. After both nuts are tightened together, You can turn the forks again and probably feel that you've added resistance as the bearings are pressed together too tightly. Loosen up the nuts and try again until the forks have no resistance when they turn and no slop or wobble when locked together. Bearings jammed together too much will start pitting and the balls will break. You will feel the bearings crunching under the pressure they are in. This is obviously not good. Over compressing the bearings for a few seconds is not a big deal, but leaving them that way long term definitely is a problem!

This is a fairly common area that manufacturers go cheap. These are shielded bearings. They don't seal out dirt and water and as you can see, there's plenty of dirt on the bearing. Road grit and water can get in the bearing and these bearings are quite crunchy with grit. They won't last long and they will just create problems in the long run as the wheels get harder and harder to turn as the bearings deteriorate. I don't have any of the correct bearings right now, but I will be replacing them with sealed hybrid bearings. Hybrids use ceramic balls and steel races. They have much less friction than all steel bearings. The cost for hybrids is about half that of all ceramic bearings and the friction is just a little more than all ceramic bearings. They are a good option. Nichi or NSK bearings will serve you well. As a rock bottom minimum, get sealed all steel bearings.

Image
Image

I made a 3 video series some time back about changing out bearings. I'll be doing essentially this when I replace the shielded bearings.

https://www.youtube.com/watch?v=R83LZLU ... R&index=22
https://www.youtube.com/watch?v=WXtD3nt ... R&index=23
https://www.youtube.com/watch?v=4Dh6-4_ ... R&index=24

I mentioned in a previous post that tubeless was the best way to go. Here's a really good reason to NOT use tubes. The tire has rotated around the rim and this has caused the valve stem to get pulled over. Given a little more tire rotation and the valve stem will tear off and you have a flat that can't be fixed on the side of the road. Tubeless will never have this problem. When I redo the bearings, I'll pull the tubes out too. I have valve stems that have rubber seals for making the wheels tubeless.

Image

The new forks in place.

Image
Image

This is the existing deck. It's fairly smooth and made of plastic. If it gets wet which is very possible, then you feet can slip off the deck and that would be very bad. As a minimum put grip tape on top of the deck. I will go further than that and remove all the factory deck and then make my own that is wider and seals over the battery bay. I want to keep water and dirt out of the battery bay and I'll make mine a bit wider so that I can stand with my feet a little apart. It will get covered in grip tape.

Image

Since the back deck is the same as the front deck area, that will get replaced too. In the process I'll make it flat so that packages can be strapped down to it.

Image
Last edited by ElectricGod on Jun 27 2018 10:57pm, edited 2 times in total.
ElectricGod...it's a joke...don't take things so seriously folks!

XB-502 Moped:
https://endless-sphere.com/forums/viewt ... 0#p1222730

Currie scooter:
https://endless-sphere.com/forums/viewt ... 7#p1227407

Post Reply