My kick scooter project


10 MW
Nov 1, 2015
3 OH 3
I have a Ford Expedition. As anyone knows they are big and not very conducive for city driving. It's great for hauling people and stuff, but most of the time it's just hauling me back and forth at 18 miles per gallon. I have wanted a better (not gas using) solution for a good while that doesn't require that I sweat everyday going back and forth to work and elsewhere. I have considered gas powered mopeds, but most of the smaller ones are worse polluters than my Expedition for way less engine displacement. I have looked at electric EV devices for a long time, but never found anything that just grabbed me by the short curlies and screamed "BUY ME NOW!!!". I also wanted something that I could modify heavily. Well I kept looking and finally came apon adult sized electric kick scooters. Most of them were pretty lame straight from the factory with 36 volt batteries and maybe a 500 watt motor (Yawn!). I wanted a factory package that was open and easily modified and already could do 30mph or better. I then happened upon HyperPowerSports who sells a wide variety of Chinese adult kick scooters that don't cost you your first born child. I found a model that suited me and then tried to find it elsewhere or to even find reviews for it. It's been since June that I bought mine and I still don't see them sold any where else. I decided to risk it and paid with paypal. I have to say that I have spent $600 on lots of things, but nearly all of them have not given me as much fun and enjoyment as this $600 kick scooter has. It's not perfect from the factory and I'll get to that after a while, but it is mostly made from easily available components that you can find on places like ebay. I bought this model from Rick at HyperPowerSports. Rick and I e-mail 5 or 6 times a week now as I give him updates on what I am doing to his kick scooter.

This is a 48v scooter with a 1500 watt motor. It will legitimately go 30mph with a 240 pound guy riding it right out of the box.


The first thing I noticed when I got the scooter was that overall the build quality was pretty reasonable considering the price I paid for it. The next thing I noticed was after 5 minutes I had already blown the 30 amp fuse. Fortunately, there was an auto parts store right down the street so I bought 40 amp fuses of the same type and I was up and running again. Looking back It seems odd that the speed controller is limited to 40 amps, but the fuse was rated for 30 amps. Of course it was going to blow! I told Rick about this and he told me they are shipping with 40 amp fuses now. At first he looked at me as a troublesome customer because I knew way more than him about all sorts of things and asked lots of questions and nit-picked things, but over time he has come to see me as a resource. Other than the blown fuse, the scooter worked great right out of the box. My drive/ride into work is 10 miles round trip so this scooter was ideal for my purposes. The SLA battery pack weighs about 50 pounds (ugg!) and successfully got me to and from work with conservative riding. If I rode at full throttle (25-30mph) with my 240 pounds on the scooter both ways, the SLA battery was not going to get me home. Well I knew I was going to get rid of the SLA pack before I ever bought the scooter and was already designing LIPO packs.

I bought this BMS and these batteries on ebay to build 48 volt packs out of.

The BMS works, but it does a mediocre job of balancing the cells. It is not uncommon to have cells charging to over 4.2 volts and others getting to 3.9 volts when fully charged. when I built the packs, I added a 6S JST balance connector to each set of 6 cells in the pack for monitoring purposes and for balance charging if needed. Well I have needed to use the balance connectors because the BMS doesn't keep the cells balance charged very well. I still charge the batteries via the BMS, but then I also randomly pull the packs and balance charge them too. Oh well what can you expect for a $40 BMS.

The batteries are cheap LIPO's and as a result didn't last as long as I would like. When they were new with 4 identical 48v packs, I could get 24 miles out of them. 6 months later it's more like 12 miles. They don't list a discharge rating (C rating) in the auction, but the seller claims they are 3C batteries. I doubt it since the solder tabs on the cells are really light weight and probably can't handle 16 amps never mind 24 amps. With 4 packs and the factory speed controller limited to 40 amps, I am never pulling more than 1.5C from any pack anyway so that hasn't been an issue. I ordered 55 of these batteries and got several that were already showing signs of puffing up. I did some basic testing of all the cells to see if they would charge and discharge equally. Any cell that didn't make a minimum cut line got set aside. As a result I had to buy 10 more (grrr!) and 55 was already 7 more cells than I actually needed. Well considering how I was getting them for $9 each, I guess that wasn't so bad. They come with a tiny BMS on each cell so I had to take them off all of the cells. Then I built up packs. Unfortunately I don't have pictures of the pack building process and I'm not about to disassemble one now just for that purpose. I used some high density craft foam I found on ebay to create a thin buffer between each cell and to encapsulate the entire pack to protect it from shocks and jars. Heat in the batteries has never been a problem at the low discharge rates I typically do. Once the packs were all done I used 14awg silicon wires and XT60 connectors to connect them to the electrical system.

I used the factory SLA battery pack less than a week and quickly discovered that the battery connector was inadequate for the job and had overheated and burned out. As a stop gap since I had LIPO packs close to complete and was going to add a proper fuse block and a power distribution block, I replaced it with a DEANS T connector. I'm guessing the factory battery connector was really good for 20 amps...not 40 amps. I told Rick about the battery connector issue, but I don't know if he did anything about them.

Soon after I had 4 LIPO packs ready to go. The below picture is the current state of the battery compartment, but back then it looked quite different. The 4 yellow rectangles are the battery packs. you can see on one of the packs that I have a cell that is failing. I blame the cheap BMS's as the battery balance charges just fine on my LIPO charger. Each pack is 12 lipo cells separated by a 1/16" layer of foam and then wrapped on all 6 sides with more foam and covered in Kapton tape. You can see the balance connectors on each pack. The height of each pack is a perfect fit for the space.


Soon after building the LIPO packs I also installed a 4 way fuse block and a power distribution block designed for power distribution for audio equipment in cars. Both blocks have a plastic cover so it's hard to short something by accident. The fuse block has 4 30 amp fuses and is gold plated. The positive lead off each battery pack has it's own fuse. The power distribution block is rated for 240 amps which is gracious plenty for my purposes. I bought it because it has 1 large screw terminal for 4-8awg wire and 4 smaller screw terminals for 10-16awg wire. It's hard to see in the picture, but the fuse block attaches to the positive side of each battery pack and then goes to an 8 awg wire that goes to the positive side of the terminal block. The negative side of each battery terminates directly into the negative side of the distribution block. Everything is silicon coated wire and either 14 or 8 awg. Coming off the large distribution block is power to the speed controller (8awg) and a 14awg wire set going to the white screw terminal strip. The terminal strip can be disconnected via an XT60 connector. Since a lot of wires are much smaller than 14awg, I needed a way to get power to smaller things than my main power distribution block could handle. I think I paid $2 for that screw terminal strip on ebay and it will take anything 14awg or smaller. Originally all the power distribution was all mounted on the right side of the box, but I recently moved it to the left side to get the speed controller (silver box on right) in.

Soon after installing the LIPO packs (1 month after receiving the scooter) and power distribution I found I was having speed controller problems. At the factory, they hide the speed controller behind a metal plate in the upper section in the above picture. It just sits in there and bakes to death. There is no air flow and the speed controller is mounted with double sided tape to the metal plate. As a result it has no way to dump heat and failed for that reason. Rick sent me a replacement speed controller and I mounted the new controller to the floor of the box and tossed the metal plate. It was still mounted with double sided tape, but it worked fine for 6 months until I got the Kelley Controller (KBS72221E) I am using now. Once I got the replacement controller, I replaced the power wires with 14 awg silicon and terminated them in an XT60. I also noticed that the motor bullet connectors which were good for maybe 10 amps had gotten super hot and fused themselves to various other wires. I had some 3mm RC bullet connectors so I replaced those as well. I ordered some 8mm bullet connectors and now all my motor field and power connections are made up of these.

Here's the 8mm bullet connectors. The shorter ones are fine for the motor feilds, but I wanted maximum current flow for the speed controller power wires so I used these longer and more expensive ones for that. Castle Creations claims the smaller ones are good for 300 amps. I am doubtful, but I'm willing to use them for around 100 amps. The longer ones I would risk up to 150 amps. More than that and I'm going with duplicate connectors!


The scooter comes with a 3 LED "meter" on the throttle (ugg!) which I found to be completely inadequate for monitoring anything. Hunting on ebay I found this meter and also purchased a 100 amp shunt. The meter comes with a 50 amp shunt, but it can measure far more current than that. Looking back 100 amps seemed like tons of current, but now I wish I had bought a 200 amp shunt. This little meter is quite small. The top 2 buttons are basically "zero" buttons and there is a simple configuration menu inside too via the bottom button.

Here's the watt meter and my speedo. I encapsulated the watt meter in Kapton tape so it can't get wet. It has worked pretty well so far and I have ridden in the rain a few times. The buttons are slightly raised from the top surface of the meter so they have Kapton over them too and it doesn't effect them working. This is a great little meter. One thing you have to do is run your batteries down to whatever your discharged level is and then hold down the middle button so it "learns" what 0% is. The top button you use to "learn" 100%. Charge the batteries to full and press the button until it resets to 100%. Until you do both of those things the meter doesn't display the capacity correctly and the battery bar graph works oddly too. Momentarily pushing any button turns on the back light for 5 seconds.


Here's the bike speedo I bought on ebay ($12). It's, wired and water proof and the back light will stay on by holding the "mode" button for 5 seconds. Once you stop moving for a few minutes, the speedo goes to sleep and shuts off the light. I tried several wireless speedo's, but motor or speed controller RF noise kept interfering with them and so they were unreliable.

Since I only needed a little over 10 miles of battery run time to get to work and home and I had 24 miles of range, I decided more lights was a good idea. The factory lights are not wonderful anyway. I still use them, but they have been supplemented a lot. I have added these items to the factory lights.

This is a great little LED flasher module. It uses 2 relays to flash the LEDs rather than a bimetal strip like most cars have. The low current of LEDs wont heat the bimetal flashers so this is a much better choice. It also has a hazard lights mode.

I have 2 sets of side lights. The white lights are on a switch for the left thumb and the blue lights are on a switch on the throttle.

Here's the rear side lights on the scooter. The blue lights are attached to the sides of the battery box. At night I am VERY visible from the side! The light in the second picture is just the long section of the LED strip. The other section is attached to my directional lights.



In the middle of this picture is the original tail light module. The LEDs were OK, but I replaced them with higher intensity LEDs and then added 4 more. I'm probably 50% brighter in back now. I had a couple of bike LED flashers and added them too, but I never turn them on anymore now that the tail light is so bright.


This is the headlight I bought. There are cheaper versions of the same thing with an external power converter module that isn't weather resistant. This one has the power module internal and is weather resistant and quite bright. It will run on anything between 12v and 90v DC with no loss of brightness.

The light on the left is the ebay light. It's hard to see in the over exposed picture, but it's easily 4X brighter than the factor LED headlight.


Here's my throttle and switch modules. The switch module works fine as it comes from the seller, but I rewired everything and made the cable longer and used an 8 conductor cable so I could use the factory switch as well all off the same cable. The factory throttle was a basic twist style throttle with no extra switches on it. I wanted another control on the right side and didn't like the twist throttle anyway so I bought this thumb throttle. I can ride with the seat on, but I prefer standing up so I rarely use my seat. Holding a twist throttle steady while standing is not as easy as it sounds. The thumb throttle allows you to hold the grips firmly and still operate the throttle and maintain a steady speed. Also, the new grips I used are not slippery when wet or wearing gloves like the factory grips are.

Here's the throttle on my scooter. The little round button goes to the speedo. The rectangular grey button turns the blue side lights on.


Here's the left switch cluster. The right most switches came with the scooter and turned on the horn and head/tail lights. I swapped the buttons so the green horn button locks down and the light button is momentary. The right horn button turns on the back side lights. The red momentary light button acts like flashing your high beams in your car. Sometimes people don't see you coming or just didn't look before they start pulling out so now I can flash my head lights at them and get there attention. The upper left yellow switch is directionals, followed by the red slider switch for head lights and tail lights and then the green horn switch which actually operates the horn. You can see a blue wire with a smaller black one on top of it. The blue wire has 8 conductors goes down inside the battery box and the black wire connects the factory switch module to the 3 switch module. I've adjusted things a little since I took this picture and now the two switch modules sit right next to each other.


The factory motor controller is a typical cheap Chinese device that barely handles the load it was designed for. I looked up the mosfets it uses and they are rated at exactly the current draw required to run the motor at maximum RPM's...with no extra overhead and that is with 3 of then in parallel per phase. Basically, I was waiting for it to die no matter how cool or hot it was getting. Also, I wanted MORE SPEED AND POWER and there was no way the factory controller was going to give me what I wanted. I looked around and read lots of other peoples stories and went with a Kelly KLS7230S speed controller. Kelly has great e-mail based customer service. They usually respond back within a couple of minutes and they know their products really well. I received the KLS7230S controller which can do 230 amps peak, hooked everything up, put it into learning mode so it could detect my motor and then tried it out. My motor stuttered and jerked, but it did run. There's differences in hall sensors and the cheap Chinese hall sensors in my motor sent out spurious signals that the speed controller was responding to and attempting to adjust the motor fields to as a result. I haven't tried it yet, but I have 3 of these motors (2 1500 watt and 1 2000 watt) and I bought the Kelly recommended hall sensors and put them in one of the 1500w motors to see if that "fixes" the hall sensor problem with the KLS controller. Kelly being the great folks they are sent me a different controller (KBS72221E) at a discount that is more forgiving than the KLS series controllers are. In the above picture of my battery bay is the KBS controller in the upper right area. It is working like a champ so far. On the subject of Kelly motor get what you pay for. These things are built tough. I think their specs are accurate if not a little conservative. The KBS controller can't do the current that the KLS can, but still it will probably be plenty enough for my purposes. Later if I get an even bigger motor, I can always go back to the KLS. Either way...Kelly motor controllers are really nice controllers and not overly expensive. I would definitely buy more of them in the future. Everything about these controllers is built to last and be reliable. I have mine bolted to the steel box and there's heat sink grease on the underside of the controller. I doubt heat will be much of an issue.


This is the 2000w motor in a bracket I made for it. The original 1500w motor looks identical and is the same dimensions, but has smaller wires and the field coils inside are smaller. The 1500w motor also has a mounting bracket welded to it. I can feel that I have more power with the 2000w motor. This motor is probably equivalent to an astro flight 3210 motor, but at 1/3rd the cost and a little less efficient. I paid $120 for it.


These scooters have atrocious wiring inside them. There's no distribution buss or common way power or ground gets to things. Wires are just twisted together in a hodge podge manner and then badly wrapped in electrical tape. I wanted to start adding more lights and other things so I got into the wiring and it was literally a shocking mess to say the least. As a result, I ran new cables to just about everything except the key switch. The rear lights now have a dedicated 8 conductor cable for them and the left switch cluster does too. The throttle cable was sufficient for the job so I left it alone and the watt meter already had a dedicated cable for it. One of the problems with the wiring inside these scooters is some things switch on or off the ground wire and other things the +V wire. Well if that's not confusing, then what is!? I changed everything to switch off via the ground wire. The reason why is that way +V goes only to the device being powered and not up the steering tube. The wires going to the actual lights are pretty static and don't move much and are protected. The wires coming out of the battery box and up the steering tube are open and exposed in several places. Having hot wires in places where they could get damaged sounded like a bad idea to me so I used ground switching for everything. Now almost nothing "dangerous" and high voltage is exposed to possible shorting. It meant I had to rewire the tail lights, head lights, horn, switches, etc to all work via ground switching. It was a pain to do, but I think worth the effort.

Because I was rewiring everything anyway, I wanted to use a screw terminal block that everything goes to before it goes elsewhere. I originally used one of those 8 position white blocks, but I needed lots more than 8 screw terminals and it was overkill anyway so I got on ebay and found some smaller ones that were garbage and then found these which are well made. All the wires going from switches to lights or whatever are all small wires (24 to 28 awg) so I don't need a high current screw terminal block. I took these screw terminals and hot glued them back to back and then soldered their pins together to make my own mini terminal strip with 32 positions. Each set of 8 positions was super glued end to end. That left a nice "V" channel down the middle of the whole block so I dropped a long strip of metal in there and hot glued it in for extra strength. I currently have 6 free positions left on it. With the crappy terminal strip I had quite a few wires wire nutted together.

In this picture you can see the old terminal strip made from the garbage parts and the final product behind it. They are almost the same length, but the new one in the back has 32 positions rather than 16 and it's a way better quality terminal block. I doubt I will replace it again because it works so well. 100% of my wires for whatever switch or light or whatever all go into the terminal block now. Each wire hole is pretty small, but it will hold an 18 awg wire or 2 24 awg wires. The terminal block gets stuffed up under the front section of the battery box where all the wires terminate anyway.


I also added a back deck. In the original picture you can see the scooter has a tiny mud cover over the back wheel. It wasn't enough to keep water spray off your back and so I commonly got water and dirt up my back on wet days. I wanted some cargo space anyway or a place to carry a passenger and so the back deck was the perfect solution, fixes the water spray issue and gave me a natural mounting point for all the rear lights. I made it strong enough to carry my 240 pounds, but still be really light. I think it added 3 pounds to the scooter when done. The top is covered in grip tape and I have some threaded brass inserts in it so that I can mount a basket back there too. That grey wire is the cable that runs all the lights in the back. You can't see it, but right below the motor is a weather proof connector that allows me to disconnect all the back end electrical from the scooter if I need to remove the back deck. There's another one right behind the tail light that disconnects the side lights so I can work on the tail light module if needed.

Here it is from behind and from the side.


I found these connectors on ebay in 3 to 8 pin configurations. I use an 8 pin version for the tail section disconnect and a 6 pin connector for the hall wires coming out of the motor. completely assembled, they are about the size of your index finger and water proof. Considering their size and price, they are quite rugged.


The motor as it came from the factory has a single long cable that terminates inside the battery box. I wanted to be able to swap out motors or to temporarily plug one in for testing or whatever without having to pull everything apart to do it. I ordered some 10 awg silicon wire and put 8mm bullet connectors on both ends and ran it in place of the long cable coming out of the motor. I also ran a separate cable for the hall sensors which terminates onto my terminal strip inside the battery box and into a weather proof connector at the motor.

This is the motor wires tucked in next to the motor. You can see the black hall connector.


This is the original 1500w motor disconnected and pulled out. The motor wires are 8" long and easily disconnect. I wrap each field connection in a little electrical tape since it is possible that water could get into them and possibly cause issues.


Almost done...LOL!
I wanted to have everything run on 12 volts and finding stuff that runs on 48 volts is limited and higher than that is virtually impossible. 12 volt stuff is like trying to find sand at the beach. All the lights and horn originally ran on 48 volts. As long as I never changed my operating voltage from 48 volts that was fine, but I have every intention now that I have a Kelly controller that can handle lots more voltage to upgrade to 72 volts later. That meant I needed a DC-DC converter and I found one on ebay that is good up to 80 volts and converts down to 12 volts at 3 amps. I just needed to run LEDs and the horn so that's plenty of current for my needs. It's a little difficult to see in this picture, but that silver box in the upper right corner is the DC-DC converter. There's a blue 4 screw terminal block to the left of it. That is the 12 volts coming out of the converter. I put my amp meter on the converter and found even with no load it was still pulling several hundred milli-amps (grrr). Well that was no good as it would eventually run down the batteries. Originally the key switch turned off power to all the peripheral items in the scooter, but the Kelley controllers all use 5 volts to enable the controllers so that meant the key switch was now tango uniform for applying power to everything. Originally I mounted a 15 amp toggle switch under the right edge of the deck to turn everything off, but I kept forgetting to flip the switch off when I got to work or to turn it back on when it was time to go home. My solution was to go through my scrounge pile of old power supplies and electronics and find several power mosfets that could be turned on with 5 volts and could handle 10 amps and 80 volts. An hour later I had quite a few that did what I needed. I then picked the best of them and using an old CPU heat sink created a simple mosfet switch using n-channel mosfets. That's the yellow square thing in the picture. It's thoroughly wrapped in Kapton tape. Being they are n-channel mosfets, they switch the ground side of the circuit rather than the positive side of the circuit. That means the ground reference for everything going through the mosfet switch is lost when they are turned off. That meant I needed an isolated ground and 12 volts completely separate from everything else. When you measure across source to drain with the mosfets turned off you get 26 volts and infinite resistance. Keep in mind that there is other stuff (the switched load) in series with the mosfets that are providing some resistance. This why I measured 26 volts rather than 48 volts across the mosfets. Turn them on and you get 0 volts and .03 ohms. Fortunately, ground based switching was easy to do since I had just rewired everything for ground switching. So everything switched on by the mosfets has a separate ground reference from the battery source when the mosfets are turned off. Switched on and the internal resistance of the mosfets is so low that you can't tell they are there and the regular ground reference is restored. It seems like an odd way of doing things, but it works reliably and now power to everything except the speed controller is turned on and off automatically by the key switch and 5 volts. One of the wonderful things about mosfets is they consume so little power to remain activated...a few picoamps since the gate has virtually infinite resistance. Turned off they act just like an open switch and turned on they act like a closed switch. A relay by comparison has mechanical contacts where arcing can occur and they require at least a few milliamps or more to activate. Most 5 volt relays are low current devices anyway and wont have high current contacts. The mosfets are a much better choice. In this picture the gold thing on the left side is my power distribution block.


Here's a better picture of the fuse block and power distribution block.


This is the back wheel taken off. The axle bearings are of marginal quality and I have had to take them apart and re-grease them at least once in the past 6 months even though they are sealed bearings. I'll replace them after a while. This picture is really here because the rear sprocket is on a freewheel. I had to find a castle nut socket that fit the the one-way clutch. It has no seals on it and dirt, water, whatever gets into it pretty easily. I noticed it was making crunching sounds and so I knew that wasn't good.


Here it is taken apart and cleaned up. Fortunately the bearing races were not damaged


It's unfortunate that it has only 2 pawls in it and they are smallish. I expect it will be a breaking point after a while.


And reassembled. I spun it a 20 or 30 times to get the grease redistributed and to see if it sounded OK or not. That white circle is the grease coming back out of the one-way clutch. There's a tiny space in the one-way clutch where I might be able to add a grease nipple or at least a grease fill hole. That way I won't have to take it apart and I can just force grease into it. When I took it apart, it was completely dry inside (no lubricant anywhere)...just grit and some hardened black gunk on everything.


I don't know about anyone else, but after a while I forget things. I needed a single source where everything scooter could be referenced. This is pretty much how everything hooks together electrically. I have added some more stuff since I took this picture so this is a little bit dated. Feel free to "borrow as you see fit".

Scooter%20Schematics_zpsnr5kbgho.jpg last thing. I got tired of spending hours on my knees praying to the scooter god...umm I mean working on the wiring inside the battery box or whatever. There were also times when I needed to run the motor off the ground or to generally test things. I made this stand completely out of scrap wood and random hardware. It cost me electricity to make and that's it. Anyway, the bottom drawer has spare scooter parts in it and the top drawer has some tools in it. The top rotates separate from the base and the base has castors under it so I can roll the whole thing around. It's strong enough to hold my weight and the scooter.

More pics to drool over...:)
As a platform to modify, these scooters are perfect. As something you are going to ride occasionally they work really well. For a daily commuter like I use mine, you are probably going to want to do some work on it.

That picture of the watt meter and speedo wasn't so great so here's some more.



Front wheel and speed sensor

Blue side lights

Side lights connector

The factory chain tensioners don't hold very well. I even used blue locktit on the threads and they still would work loose so I replaced them with these.


Some general pics of the scooter





Stuff yet to come... (1 and 2 have been purchased)
1. Hydraulic brakes
2. New 60 or 72 volt LIPO packs with better BMS's
3. Better wheels and a Nuvinci N360 hub.
4. 150CC turnigy motor (YEAH BABY!!!!)
5. Building a battery box on the back platform
Dude, that is one serious write-up, good job.
Do you have any idea how hard it is to post a reply with quote :wink: ?
That hand-drawn schematic is a work of art !
Thanks for the bumps guys!

Most people seem to want something small that they can take on the bus or leave at their desk. I wanted a gas moped replacement that could do 30mph+ and have decent range. I can take the busses around here, but why would I when my scooter can get me there instead? Most kick scooter projects I have seen are intended as "last mile" devices. I might build one up, but I wouldn't have a practical use for it. If the weather is too bad to ride, I'll just use my Expedition instead.
I got the rear hydraulic brakes in the mail on Wednesday, so I had to put them on. Braking is stronger than the cable pulled rear brake. It doesn't look as clean since the brake line is a little longer than the cable, but it all works and the caliper lines up better with the rotor so that has reduced rolling resistance. The 2 foot brake line that came with the brake kit has 8mm fittings and the 80 inch line has 10mm fittings (grrr!). I had to make spacers by wrapping some aluminum around the banjo bolts so the larger fitting would stay centered on the 8mm banjo bolt. It works, but isn't ideal. I've considered getting 10mm bolts and tapping out the bolt holes to 10mm too, but that will be a royal pain and will require I take everything apart so I can be sure to get all the metal bits out of the caliper after retapping. I wish I would have known there was more than one size of banjo fitting for this size of hydraulic brake. 8mm banjo's on brake lines are pretty uncommon while 10mm banjo's on brake lines are everywhere and for super cheap. Oh well...this was an experiment and it worked out much better than the factory mechanical brake. I've already bought the front brake kit.

This is the parts I purchased on ebay to do the hydraulic brakes.

I found some better master cylinders with slightly larger reservoirs and a site glass. The master cylinders I bought hold 3cc of brake fluid in their reservoirs and you can't see the fluid levels. If there was a leak, the fluid would run out super fast and I wouldn't know it happened until too late.

Updated with clearer pictures...



This is what I started with.

Lol. You're just like me. We both commute with electric scooters, and end up replacing everything on them to make it work for us. Lol
MrDude_1 said:
Lol. You're just like me. We both commute with electric scooters, and end up replacing everything on them to make it work for us. Lol

Exactly...upgrade it to the max!
Parts for the front hydraulic brakes are ordered. I'll have them in bits and pieces over the next week or so. Ironically I ordered a complete brake set for the front and then found the hose elsewhere in the color I wanted and then found better master cylinders which I also bought. I could have just purchased calipers after all of this mess!

I'm looking at possibly going this route with the hydraulic lines. Bicycle hydraulic brakes operate at the same pressures as the larger stuff. They just move less fluid. Anyway, my thought is not to use the cylinders and calipers for bikes...just the brake lines. The lines are the same size as brake cables (5mm OD), but hose comes in a zillion colors and the fittings are inexpensive compared to moped class banjo's that aren't crimp-on. The bike fittings all push onto the hose and include parts to lock the hose onto the fitting that use small wrenches to install. Cheap hose crimpers cost several hundred dollars so that's ridiculous for a small project. The bike fittings make custom hydraulic hoses reasonably priced and the hose itself is really inexpensive.
sucks you have to use moped brakes on your scooter, 100lbs is heavy! I may look into hydraulic mountain bike brakes for my scooter later, they're very reasonable.
shortcircuit911 said:
sucks you have to use moped brakes on your scooter, 100lbs is heavy! I may look into hydraulic mountain bike brakes for my scooter later, they're very reasonable.

I'm getting enough braking power out of the mechanicals if I keep them adjusted all the time, but they drift out of tight braking gradually so that I don't notice the slow drift until they are pretty weak again. I want my brakes to work the same all the time. The mechanical brakes aren't very reliable or consistent.
I am working on a new and improved spark free loop key. It will consist of 2 keys and 2 loop connectors. You could just as easily use a small toggle switch for the first loop key. The first key has 2 high intensity white LEDs in parallel and a 1 watt 2.7k resistor. The second key is just a high current loop key. The first key is for charging the capacitors in your motor controller. All that arcing and sparks is from the massive amount of current that floods into the electrolytic capacitors in your motor controller as they charge. The arcing is hard on your connectors and the inrush current is hard on your electronics. The resistor in the first loop key by itself is enough to eliminate the inrush arcing you get when charging capacitors, but I wanted something that indicated that the caps were charging. That's where the LEDs come into play. Since I am using 3 watt LEDs, that means at 3 volts they are drawing about 1 amp. I have 2 of them in parallel stuck back to back with a small piece of aluminum between them. The 2.7k resistor with a 48 volt battery limits the voltage to the LEDs to around 3 volts. I tested this circuit on 4000uF of 200 volt capacitors and they charged to 80% in a couple of seconds. Initially the LEDs glow brightly, but soon they get dim. What is happening is the capacitors are so close to ground potential when discharged that the LEDS and resistor "see" the full battery voltage across themselves. As the capacitors charge less and less of the battery voltage is now being seen by the LEDs and so they glow more dimly. Finally when the caps get to nearly fully charged, there is insufficient voltage across the LEDs anymore to bias them on and they no longer glow. If you wait for this to happen you will be waiting probably a full minute. The capacitor charge curve looks pretty close to a reverse exponential growth curve. You initially get the caps charging very quickly, but then a few seconds later, they are charging slower and slower. Fortunately those initial couple of seconds are all that matters to get you to 80% charged. At that point just plug in the main loop key. There wont be any arcing anymore.

I run at 48 volts for now. 20-40 amps gets me moving and 60+ amps is common for full throttle. I use 8mm bullet connectors for my power leads since they can reliably handle 200 amps and later I will be upgrading to 72 volts and those power connectors will be perfect for when that happens. As a result of my amperage usage, I can't use an XT60 for my main loop key and I really don't want to use an XT90 for one either. I want to maintain the 200 amp capability everywhere. I'll build a loop key that uses 2 8mm bullets and embed them in epoxy. For my charge key I have built it onto an XT60 connector. I know it's serious overkill for something that might see 6 amps, but it's also a really reliable connector that wont ever fail me. The charge key is completely covered in hot glue so everything about it is water proof and you are protected from accidental shocks.

Here's the schematic for the loop keys.


Photobucket took down this pic. I guess they thought it was kiddy porn or something.
Great write up! Curious what are the wheels made out of, and what are they coated with?
NyOliver said:
Great write up! Curious what are the wheels made out of, and what are they coated with?

They look like they are CF rims. That's what I originally thought when I was looking at them before I bought a scooter. They are machined cast aluminum with a decal that looks like CF.

The front wheel is OK and balances pretty well. The back wheels have a free wheel hub that threads onto the wheel. The rear sprocket bolts to the hub. The manufacturer doesn't do a very good job making sure their tools are properly aligned and so there is some amount of wobble in the back wheel on the sprocket side. There's a sealed bearing on either side of the wheel and a 12mm axle runs through it all. The threaded part that mounts the free wheel hub is OK on both my rear wheels, but the bearing seat is machined a little off center so that's where the wobble comes from. I bought a second rear wheel in hopes it would be better than the one that came with the scooter, but it was actually worse. At 30mph or less it's not too bad, But you feel the wobble. If you spin up the rear wheel with it off the ground you can see the wheel and sprocket wobble. I'll replace both front and back wheels after a while with better ones. I want things to be the best they can be. But hey, what can you expect from a $600 scooter?
I'm still working on the loop key improvements. I have made the charge key and it works quite well. I still need to build the dual connector block with the master loop key.

I ordered some parts a while ago and they finally arrived so I could build a high current charger. A lot of people use noisy server supplies in series to get the voltage they want. I wanted something quiet and still high current so I have been looking around for a solution. I found some DC to DC Up converters on ebay that looked like they would work. I wanted something that would work with my current batteries and something that would work for the future. Right now I am running at 48 volts, but intend to go to 60 or 72 volts later. The DC up converters I bought wont do 72 volts...just 60+, but they were $9 for 2 of them so for an experiment, they were a reasonable choice for now. I have since found some that will work for 20S charging. The converters I bought are advertised as 1200 watt up converters so at 50.4 volts that's 24 amps each. I assume conservatively so I think more like 12 amps at 50 volts is more likely. Two of them in parallel gets me a reasonable amount of current and this power supply is rated for 24 amps on the 12 volt supply.

This is the up converter I used. It does a good job, runs on a wide range of input voltages and has an adjustable output voltage.

This is the large ATX connector on the power supply. Notice the green wire fourth from the right. Cut it off at the connector and the black wire to the left of it. Twist them together and the power supply will now turn on automatically. My power supply has an switch too so that's how I turn it on and off now. All the other wires can be cut off at the connectors or whatever length is useful to you.


PC power supplies mostly use a few colors to indicate what that wire is for. Black is always ground, red +5 volts, yellow +12 volts and orange +3.3 volts. All the wires of a particular color can be used interchangeably or together. In the below pictures you can see that I have used groups of each color twisted together. As a result since there are loads of ground wires, I made 5 ground points on the terminal strip. + 12 volts is plentiful too so I made 2 groups of them...and so on. I've been playing with electronics since I was a young boy so I find having + and - 5 volts and + and - 12 volts useful so I terminated those wires too. I also terminated the 3.3 volt line. Now I can use this power supply for lots of other things too. I spent maybe $20 on parts for this project. Now I have a 24 amp 50.4 volt charger. The 12 volt lines from the power supply go into the DC to DC up converters via the yellow and black wires which run in parallel to both converters. The outputs of the converters via the red and black wires go into the watt meter. The watt meter has red and black 14 awg silicon wires that then terminate at the charging connector. The terminal strip is screwed down to the top of the power supply, but everything else is stuck on with double sided foam tape. A lot of ATX power supplies wont turn on fully if they don't have a load on the 5 volt side. I added 2 3k 5 watt resisters in parallel on the 5 volt supply to give it a little load. The 2 up converters load the 12 volt circuit pretty heavily so it runs at about 11.5 volts. I bought the little LED volt meter about year ago for less than a dollar. It shows the voltage on the 12 volt supply. In the second picture you can see the blue precision potentiometers right below the toroid coils. They adjust the output voltage. I was impressed that the up converters kept a steady output of 50.4 volts with or without a load on them. My charger is quiet, delivers a decent amount of current and is totally adjustable.


What do you consider to be high current?
be aware that PC ATX power supplies can not really put out the same continuous power that they are rated for. They are overrated.
The industrial stuff is underrated... server supplies, meanwells, etc. and plenty of them dont use fans.

If you want anything in the multi-kilowatt range, you're either going to need a fan, or a HUGE heatsink.

but if you're just looking for 1200 watts or less, theres plenty of cheap options that dont use heatsinks.... heck I made a flat backpack charger by just connecting some free laptop supplies in series. It does 565watts and its totally silent.
MrDude_1 said:
What do you consider to be high current?
be aware that PC ATX power supplies can not really put out the same continuous power that they are rated for. They are overrated.
The industrial stuff is underrated... server supplies, meanwells, etc. and plenty of them dont use fans.

If you want anything in the multi-kilowatt range, you're either going to need a fan, or a HUGE heatsink.

but if you're just looking for 1200 watts or less, theres plenty of cheap options that dont use heatsinks.... heck I made a flat backpack charger by just connecting some free laptop supplies in series. It does 565watts and its totally silent.

You make several good points! I typically don't charge at higher than 1C. When I come home at night, I plug in the charger and my scooter has all night to charge. By morning it is fully charged. I currently have 24 amp/hours of battery capacity so charging at 4 amps gets me fully charged in 8 hours or so. I was charging at 2 amps until I built this charger. I easily have 5X more current than before. The PC power supply as you say wont do 24 amps continuous and I understand that. I just didn't say so. I was speaking in terms of maximums when I wrote up the charger stuff. Realistically 10 amps is more like what it will typically do for continuous load. The power supply at 2 or 4 amps has no evidence of warming up. The 2 fans in it spin at 50% and make virtually no noise at that low current level. The DC converters run at 70 or 80 degrees F. I can live with that kind of performance. So low vs high current...that's all relative right? I don't need something big. Portable, quiet and relatively small matters to me more. Since this was so cheap, I'll probably build a second one for use at work if needed. Something that gives me status and is easy to use, but has flexibility is a giant plus to me. Like I said several supply voltages is useful for all sorts of things. As a result I went with an ATX power supply since I knew it would have lots of voltage options built in and it would also handle a reasonable amount of current. I considered going with several laptop power supplies in series, but that meant a fixed and inflexible solution that I was going to have to buy parts to build since I didn't have the power supplies. I have 8 or more old ATX power supplies that work and the other parts were $20 so that was virtually free to me. Cost (virtually free) was a significant consideration. By adjusting 2 precision pots on the converters I can set my output voltage to pretty much anything I want. That would not be possible with other options. I wanted the flexibility this solution provides. Also, with several free ATX power supplies just laying around in a box, I have tons more current capability that will cost me nothing to add if I choose. In my opinion this solution was a great choice no matter how I look at it. I didn't even go with the largest (highest current on the 12v supply) ATX supply I had. I wanted to use the top of the power supply for mounting things so it could be an "all in one" power supply. I even took the cover off to see if I could mount the DC converters internally, but the power supply I selected was pretty compact (ITX) already. I have a 2000 watt ATX power supply that has dual 12 volt busses that are each rated at 30 amps and yeah that would have gotten me lots more current, but the fan takes up the entire top of the box. So much for mounting things on it. This was a reasonable compromise between size, amperage, weight and flexibility for my uses.
Cool. sounds like you have it down to me.
I know what you mean about having it fit your criteria. I have a couple of chargers, each with mine has different criteria... My home chargers are adjustable up to full 1C fast as possible charging. Since my battery is oversized, that means I can plug it in and charge that the 1C rate and 15mins later is fine for me to go to the store or work and back.

This is the first charger I ever made.. its a twin meanwell SP-500-27 setup, in a simple custom enclosure, with the fans replaced by huge overkill ones.. each charger has two fans, one in and one out. Using larger fans let it have a lower fan RPM for the same CFM of airflow.. so its pretty quiet and it can hold over 20 amps all day long. I only have 20ah of 16s... so I limit it to a max of 20A.
Heres an older pic of it:


At work though, its a quiet office, so I needed something absolutely silent. I made a very simple charger that is nothing more then 4 laptop supplies held together with silicone. I added a amp/voltmeter to go along with my CA, and its done.
You can see it in the background of this pic.

Funny thing about this charger, is I was given the supplies from work, when they were cleaning stuff out.. and here they are, back in the office.
Its nice for my backpack too, as It fits in the "laptop" slot of it perfectly. This does a solid 8.4A @ 66.2v. I often take it on trips for a quick charge when I stop.

One of my current projects is a server power supply based home charger that is stackable... It can go as high as 50amps at 120v DC down as low as 10v... with some minor limitations in between... The goal is a simple to use smart charger that will work with my goped (20A max,@66.2v), with a zero motorcycle(50A @ 116.2v), or with anything in-between. The whole thing runs off a dedicated 240v socket I have in the garage.
I like those digital sweep style gauges. Where did you find them?

I didn't think about it until just now...sitting at my desk, but we used to be 100% lenovo and over the last couple of years switched over to Dell laptops instead. I know theres a load of lenovo power bricks in some storage room somewhere here. 99% of the people I work with have a company provided laptop and there's 250 of us in my office. I should see about acquiring all of them they will let me have. The smaller bricks are like 60 watts at 20 volts and the larger ones are 90 watts at 20 volts. That's a decent amount of current either way! And totally silent is always good. I would probably take them out of their cases so I could mount them all in a single enclosure and wire in an inline fuse to each one, add a large fan, etc.

120volts DC at 50 amps. Why not just rectify the AC coming out of the wall? Do that to both phases of that 220v circuit and depending on the breaker and wiring, amperage can be whatever your feeder lines into the house can handle. Yes I know 120v AC does not rectify into 120v DC. I know a lot of people don't use BMS's on their batteries. They just set the output voltage of the power supply so that overcharging isn't possible and then balnce from time to time. I use a BMS on each of my battery packs. As a result I can supply the pack with more voltage than the total number of cells can safely absorb. The BMS limits the charging voltage to each cell to 4.2 volts despite the fact that I might be supplying 4.5 volts per cell. My old charger delivered 56 volts, There was no way I was going to deliver that to unprotected LIPO's! With the crap BMS I am currently using I never got cells over 4.25 volts despite the power supply delivering 56 volts/12=4.67 volts per cell. I'm basically saying let the BMS regulate the per cell voltage and then you are always balance charging too. You can run BMS's in parallel or series.

I wish I could bring my EV inside, but we have a policy that says no to any kind of wheeled device that's not a dolly or cart.
ElectricGod said:
I like those digital sweep style gauges. Where did you find them?
I think I found them on Ebay, but I emailed the seller... for $7 or so, I bought 5 of them. if you only need one, PM me your mailing address.

ElectricGod said:
I didn't think about it until just now...sitting at my desk, but we used to be 100% lenovo and over the last couple of years switched over to Dell laptops instead. I know theres a load of lenovo power bricks in some storage room somewhere here. 99% of the people I work with have a company provided laptop and there's 250 of us in my office. I should see about acquiring all of them they will let me have. The smaller bricks are like 60 watts at 20 volts and the larger ones are 90 watts at 20 volts. That's a decent amount of current either way! And totally silent is always good. I would probably take them out of their cases so I could mount them all in a single enclosure and wire in an inline fuse to each one, add a large fan, etc.
theres a few problems with that. IBM makes some nice compact high power supplies, but inside the plastic shell it looks like an aluminum brick. It uses that outer layer of aluminum as a heatsink. Its also connected to DC ground, so you can not let them touch. obviously not an issue if you leave it in its case...
They're also not adjustable. I used 16v rated ones, and they really put out 16.5v... but they all do it perfectly identically. they're quite nice inside, as long as you dont want to take them apart... lots of that white goop for vibration.. and most importantly they are current limiting and thermally protected. When I put them on a discharged pack, they limit their current... others can shutdown or burn up.

ElectricGod said:
120volts DC at 50 amps. Why not just rectify the AC coming out of the wall? Do that to both phases of that 220v circuit and depending on the breaker and wiring, amperage can be whatever your feeder lines into the house can handle. Yes I know 120v AC does not rectify into 120v DC.
Thats just the start of the issues.... not only is 120AC rectify to ~168, but you need some form of current limiting... and isolation. Isolation can not be understated when it comes to making high power devices safe.

ElectricGod said:
I know a lot of people don't use BMS's on their batteries. They just set the output voltage of the power supply so that overcharging isn't possible and then balnce from time to time.
This is what I do.
that and log everything so I can see if things are drifting...
if I fully charged my 16s to 4.2v, I would have 67.2 volts... instead, I charge to 4.15.. thats 66.4. I really set everything to 66.2 incase theres thermal drift in the charger, but I have yet to see that happen.

ElectricGod said:
I use a BMS on each of my battery packs. As a result I can supply the pack with more voltage than the total number of cells can safely absorb. The BMS limits the charging voltage to each cell to 4.2 volts despite the fact that I might be supplying 4.5 volts per cell.
Thats not always exactly accurate, and I wouldnt depend on the BMS that way. honestly that sounds pretty unsafe, and if you start pushing higher wattage into the pack, your BMS may overheat and fail..... and then you have nothing limiting the voltage into the pack.
IF you want it to balance everytime, set the cells to just under 4.2, then set the charger to 4.2... that way it fails safe, instead of failing and then overcharging.
current limiting...I bet in 5 minutes I could google for 5 different ways to implement current limiting with a couple of mosfets, an op-amp and a few other parts.

Laptop supplies...yes...keeping them isolated from each other was always the plan.

Regarding BMS functionality, the whole point of the BMS is to limit voltage per cell so no cells get over charged and all cells charge to within close limits of each other. Obviously there are limits to what the BMS can control. All the BMS's I have looked at have a current limiting shunt on them. Like anything it can be overcome with enough effort, but under typical situations that's not going to happen. I agree with what you are saying. There's no way I would apply 72 volts to a 48 volt BMS and expect it "limit" anything. There is common sense to this stuff. A 12S BMS needs 50.4 volts at 4.2 volts per cell to fully charge all the cells. You can safely apply 55 volts and not over charge the cells. The BMS will limit a little voltage, but not a lot.
Front hydraulic brake parts finally all arrived so I got that all in place today. Like the back wheel the caliper didn't quite line up straight so I shimmed one side of the caliper with some washers to get it inline and parallel with the rotor. Then I noticed that the caliper was rotated on the bracket a little differently so its trailing edge just grazed the rotor and made a clicking noise. It's hard to see in the pictures, but the upper right hole has been elongated a little to create a small bit of rotation and that got the caliper straightened out and clear of the rotor. I think I still have an air bubble in the line that hasn't worked itself out yet because the first squeeze feels a little mushy and then after that it feels nice and crisp. There's a small leak at the caliper that I don't have sorted out yet. I have some copper crush washers coming, but they wont be here for a while. Also the new caliper has a little less side to side travel than the mechanical one so I had to move the rotor away from the wheel by about 1/8". I tried mounting the caliper differently, but it was always too much "in" or too much "out". I used some washers behind the rotor got me the exact amount of lateral travel to get it centered in the caliper. There's still over an inch of bolt threaded into the wheel so I don't think I have compromised the security of the rotor at all.


The hydraulic brakes are finally all done. I had minor leaks at the front and back calipers. It ended up that I couldn't use a wrench to get the banjo bolts tight enough without twisting the calipers off their bolts. I pulled out my 1/4" hammer drill and got another half turn out of each bolt. After that their were no more leaks to deal with. I made aluminum crush washers and thought maybe they were too hard to properly crush and thus seal up. I am pleased that the washers weren't the problem and that it is finally all sealed up. Ah well...I got tired of waiting for copper ones anyway! The weather is getting better here, but still bloody cold in the morning and after the sun goes down. The roads are finally getting clear again. Hopefully I will be riding again very soon. I wonder how well my batteries will do. The last time I rode it was in the mid 30's all day and I could really feel the batteries sagging on the trip home from the cold.