E-Bike XB-502 conversion project

[ElectricGod]

I got the new flasher module in and rewired the handlebar wiring with a terminal block and started testing things out.

This flasher module is supposed to blink exactly the same on a single LED or 25 amps. I don't know about that yet, but I've tested it on a single 12v light bulb replacement and it flashes just fine.

This is the terminal block. It's kind of pulled out right now while I figure out what all the wires do, but it will get stuffed up under the dashboard when it's all done.

And it lives...RUN AWAY!!! I've tested out the turn signals, brakes and head lights so far. There's a switch for high and low beams. They both work. I've still got quite a few wires to sort out. There's 14 wires that come off the handlebars and go to various things. I also added a couple of spares for who knows what later. The watt meter isn't connected to battery power or its shunt so it doesn't display anything yet. I replaced the original piece of dashboard with a same shaped piece of acrylic and then spray painted it black. It had various holes in it and the speedo and battery meter were printed on it, etc. I cut small circles of masking tape and placed them where the three indicator lights are before I spray painted it. When the paint dried I removed them and laid down 6 layers of kapton tape over the clear spots on the back side. The original panel had L H and R silk screened on it and the plastic was tinted a good bit. The yellowish tint of the kapton works pretty well. I then used a labeler to make the indicator symbols...good enough. The H only turns on when the high beams are lit. I know...seriously sophisticated isn't it?

 

[ElectricGod]

It took several hours to work out what all the wires do and how they work. They cheated with the electrical. The directional lights all switch on from the positive side and have a common ground. The brake, tail and head lights all use a common positive lead and switch via the negative side of the light. It took a good while to figure out how all that worked out. When all the lights were incandescent bulbs, it didn't matter which wire was common since bulbs flow current equally well in either direction. An LED on the other hand flows current in only one direction. This created some interesting issues with the common ground vs common batt+ wiring. The headlights worked fine, but the tail lights since they were now LEDs didn't. The LED tail light bulb is 2 way...one way for the tail light and then a brighter way for brakes. The bulb uses a common ground for both brightness levels. The wiring however uses a common Batt+ which was fine with incandescent bulbs. I was either going to rewire the entire lighting system for common ground or else just fix a couple of LED bulbs to work common batt+. As you can imagine, rewiring a couple of LED bulbs is much easier than rewiring the entire moped. It was pretty easy to do...take apart the LED bulb, solder the positive end of the LED to the can of the base and then separate out the two negative ends and solder them to the 2 pads on the bottom of the bulb. It all works perfectly backwards now, but I don't have to rewire everything. I'm sure some Chinese dude had a really good reason for designing things this way, but damned if I know what it is. Everything works with a common ground...it's been the convention for as long as I've been alive. GRRR! Oh well...it's all tested and working and I've eliminated all the crappy wiring with atrocious exposed splices all over the place. There's two terminal blocks...one at the handle bars and one at the battery box. Both are 16 positions, but the one at the battery box is really only using about half of the positions. Lights, key switch, power and throttle. All the wires are now hi-temp 18 awg which is about 2X thicker than the factory wiring. I was concerned that the the 26 awg wiring everywhere might not handle my much brighter LED head lights. I'm not worried about that anymore. I also ran a separate 4 wire cable for the watt meter and another 2 wire cable for the temperature sensor that will go in the motor.

 

[Ianhill]

My watt meter had no instructions on how to calibrate, all it said was it's easy so I dug around pressing buttons and i found that if you highlight the cursor over clear and hold the ok button down the gauge will zero it's watts at that given state so a quick press zeros all the figure and hold it then calibrates the shunt to zero

 

[ElectricGod]

My watt meter had no instructions on how to calibrate, all it said was it's easy so I dug around pressing buttons and i found that if you highlight the cursor over clear and hold the ok button down the gauge will zero it's watts at that given state so a quick press zeros all the figure and hold it then calibrates the shunt to zero

I think I tried that...but it was already zeroes like this. Does yours work wireless? I can't get this one to work wireless.

 

[Ianhill]

My watt meter had no instructions on how to calibrate, all it said was it's easy so I dug around pressing buttons and i found that if you highlight the cursor over clear and hold the ok button down the gauge will zero it's watts at that given state so a quick press zeros all the figure and hold it then calibrates the shunt to zero

I think I tried that...but it was already zeroes like this. Does yours work wireless? I can't get this one to work wireless.

No I have not used it in wireless yet I'll give it ago when I bench test my hub suck it and see what I get.

 

[ElectricGod]

My watt meter had no instructions on how to calibrate, all it said was it's easy so I dug around pressing buttons and i found that if you highlight the cursor over clear and hold the ok button down the gauge will zero it's watts at that given state so a quick press zeros all the figure and hold it then calibrates the shunt to zero

I think I tried that...but it was already zeroes like this. Does yours work wireless? I can't get this one to work wireless.

No I have not used it in wireless yet I'll give it ago when I bench test my hub suck it and see what I get.

Wireless may not be a great solution. I've tried to use wireless speedos on my EV's of various kinds and they always are unreliable. The EMF from the motor or controller or both interferes with the RF signal and the speedo gets intermittent. All I have to do is let off the throttle and I get a speed reading again. Hopefully this wireless watt meter wont suffer similarly. Worst case, I just run it wired.

 

[ElectricGod]

Yesterday I spent several hours grinding off welding lumps and cutting out several things that were in the way. These bottom bracket inserts used to extend into the space more. They were for the completely useless "pedals" that came with the moped. The inserts were in the way, so I ground them off. Also, the bottom of that curved piece stuck out into the space, but did nothing. I ground them off too. I don't know why, but the moped had a kick stand and also a double leg stand that picks up the back wheel. I cut off the kick stand and ground off its welds. It was redundant and that saved me a little weight too.

Once the repainting was dried, I was able to install the battery box today. As you can see, with the 32,000mah battery pack, there's still loads of space in the box. Later when I buy LION cells, I'll weld up a 32S pack that fills the space. This was the best I could do with the turnigy lipo's.

 

[ElectricGod]

The motor and controller will fit like this. The controller will screw to the battery box. Originally the controller was up on that flat piece of metal with the 6 elongated slots. That worked for the tiny factory controller, but now that I'm upgrading from 500 watts to more like 8000 watts, it won't fit there anymore. This location gets more air around the controller since it's no longer trapped under the seat compartment.

This is a cheap 2000 watt controller I've had for a couple of years. It wouldn't fit under the seat compartment without shifting it down a good bit. There was no way the big controller was ever going to fit here. I was attempting to see if I could get enough torque out of the factory hub for it to be useful. The 2000 watt controller was a little better than the factory controller, but still, the hub could not get the moped moving from a dead stop. It's just way too weak for the size of this moped and it weighs close to 40 pounds...lots of dead weight! Anyway, that was the last straw in trying to get the factory parts to work for me.

 

[ElectricGod]

I thought I would do a side project. I was looking at gauss meters and they are $90 for a cheap one and go up from there. Admittedly they are calibrated and mine isn't, but it works well enough. I basicly wanted to be able to check north vs south and to check magnet strength. I have a stack of magnets that are supposedly N52. At least that's what I ordered, but without a real way to test them, I've never known. Now I can try them out and see if they really are N52 or not.

I used an S49E linear hall, a small 1 cell charge controller that includes a BUCK to 5 volts and a 150mah cell from a nano quad of mine. I super glued the hall to the end of a 6" long dowel and ran the 3 wires down the dowel. The +V and gnd wires go to the BUCK output to power the hall. The signal line and gnd come out to a 2 pin connector for connecting to a volt meter. These are the main components that power the gauss probe. I think I have $5 all told into this. The lipo cell cost about a dollar. I bought 5 of the buck/charge controllers for $3 and another dollar for the hall.

This is the completed gauss meter. It's 7" long. I could have made it a lot shorter, but the length makes it into a probe that can be stuck down inside places. I added a micro USB connector to the charge connections on the controller board so I can charge via any USB port. I covered the hall wires down the dowel in heat shrink, wrapped the battery and charge controller in kapton and strategically used hot glue to secure the dowel and USB port to the board and battery. A liberal use of heat shrink and a couple of small zip ties holds it all together. The linear hall with no magnetic field present outputs 2.583 volts. With the north face of an N52 magnet touching the hall it goes down to .78 volts. With the south face of the N52 magnet touching the hall it goes up to 4.3 volts. There's a simple equation to do to take the output voltage and convert it to gauss.

Gauss = 1000*(V0-V1)/k
V0 is the qiescent voltage or 2.583 volts.
V1 is the measured voltage
k is the mV per gauss or 1.4 mV/G for the S49E hall.
The north side measures .788 volts and the south side 4.29 volts. That calculates to 1280 gauss for the north side and -1221 guass for the south side.
I also have a stack 20 neo's I bought on ebay. They were advertised as N50, but I got them and just doing strength tests with them, I knew it was BS. Anyway north measures .857 volts and south 4.28 volts. Definitely weaker than the N52 magnets. I checked a few motors to see how strong the magnets are and none of them measure as high or as low as the N52's. Looks like it's working.

 

[ElectricGod]

I'm already thinking of how to MOD this thing further.

1. I've purchased a few USB meters. Several look like they have a very simple volts and current measurement chip that I can hack a little to get at the voltage sensing pin. Connect it to the hall output and now I don't need an external DMM.

2. I have another idea too. It depends on If the USB meters can measure negative voltages. If I add a voltage divider across +5v and ground made from a 100K POT and use the wiper as my ground, the divider will cancel out the quiescent voltage of the hall. The end result is the meter will display positive voltages for South magnetic poles and negative numbers for North magnetic poles. No magnetic feild reads 0. I don't care about this so much, but it does save me from doing V0-V1 since that is done by the voltage divider. Also, calibrating the quiescent point is super easy. Adjust the POT up or down a little until the meter reads 0 volts.

3. The next thing is due to the limits of halls. The hall I'm using swings lower than it does higher. IE: it measures stronger north fields than it does south fields before it maxes out. You only get a fixed amount of voltage swing out of the hall. I have some N52 magnets. They max the hall out in either direction. I tried an N30 magnet and it almost maxed out the South measurement, but didn't get close to maxing out the North measurement. If I create a cap made of heat shrink or aluminum or whatever, that will put a small gap between the hall and the magnet which will allow for strong magnets to not max out the hall. I've looked at several gauss meters and they typically bury the hall in plastic or under an aluminum cap. This not only protects the hall from damage, but it creates that gap needed to reduce the strength of the magnetic field a little.

4. If I zero out the quiescent voltage, that allows for another MOD that utilizes LEDs. Using an op-amp I can operate a couple of LEDs for showing north and south. I care the least about this one as the voltage swing on the meter effectively tells me the same thing.

5. I found a larger LIPO cell that is almost identical in size to the charge/buck board and is 300mah. I'm uessing the 150mah cell will last a week+ just sitting there. I think a 300mah cell will get me a couple of weeks of continuous use.

6. I also want to add a tiny power switch to shut off the hall and volt meter. Disconnecting the battery from the controller board just disables the board and I have to plug it into USB and start charging to get it going again so I'm not going to disconnect the cell. Of course the charge controller is supposed to protect the cell and not draw it down so that should be OK.

 

[ElectricGod]

I pulled apart the gauss meter tonight. It looks much cleaner now and now it uses a 300mah cell, has a power switch, separate 5 volt connector (long pins) and a set of short pins so I can measure the battery if needed. The striped section of heat shrink puts a 1/16" gap between the face of the hall and the magnet. Testing on an N52 with the gap is enough to keep from maxing out the hall. The wires from the hall are no longer twisted around the dowel. They run straight down the underside of dowel now. I secured the wires with a little kapton in a few places and then shrink wrapped the whole thing. On the underside I have wires taped down for now that will go to the USB meter. The bottom is totally flat so that will make a good place to mount the meter. It's all wrapped in kapton, but once I have the meter installed, I'll use clear heat shrink on it. You can see the tiny switch with the two red wires going into it. The hall and the USB meter will power on via the switch. I used a little hot glue to secure the switch. The battery and controller board are wrapped in kapton and then the dowel is hot glued to the tape in several places. It's pretty secure as long as you don't try prying the dowel loose. The USB charging port is also hot glued down to the kapton and then wrapped in a little more kapton. It's all pretty secure and I didn't need to use any zip ties...much better and cleaner setup.

 

[ElectricGod]

I'll probably add this to the USB port so I can see that the cell is charging. I'm pretty sure I have all the components in SMT packages to make this very tiny. A little scrounging of old circuit boards will get me what I need. The specific op-amp or transistor isn't that important. Pretty much any NPN transistor will work...same for the op-amp. The transistor is just providing a voltage reference...similar to a zener. This whole circuit, if I build it out of SMT parts will take about 1cm squared.

 

[ElectricGod]

I got some work done on the moped.

The first thing I did was finish the console. Not much done, but I can push the buttons on the watt meter now. I drilled 3 holes over the buttons and then added some short sections of dowel. To trap the dowel sections, wrapped some kapton tape around the bottom section of the dowel sections so that they can't come out of the holes.

Then I got back into the electrical. The first thing I did was mount the power distribution block. Then I figured out where I wanted to mount the shunt and contactor. The shunt will be enclosed inside the battery box. The contactor will mount into the corner of the battery box. I'll need to partly assemble the body pieces to see how it will look sticking out of the side of the box or out of the top. One of the two will be it's home. I like the idea of it coming out of the side rather than the top so that the lid of the battery bay is a solid piece, but we will see. This is pretty much how these components will stay. So much for all that empty space in the battery box!

 

[ElectricGod]

The factory plastic fender was too short and a POS so I purchased a new wheel fender for the back wheel. Also the original fender wasted lots of space over the back wheel. I want to control spray up into the under carriage area and maximize room as much as possible. I may have to modify my mounting brackets a little if they interfere with the chain path, but this recovered almost 3" over the wheel and blocks about 80% of any spray. It looks nicer too. Next up is to build a motor bracket.

 

[ElectricGod]

I started on the motor bracket on Saturday and got it 99% done on Sunday. The bottom is 1/4" 6061 aluminum and the sides are 1/8" aluminum. I'll need to drill a few more holes in the base for mounting the bracket to the swing arms. I also haven't figured out exactly where it will be positioned side to side yet. When I first made the bottom plate, I cut it about 3 inches too long so that I would have lots of extra metal depending on where I mounted the motor. As it turns out, had to cut off about 2.5" to get the motor in the correct spot. The bracket is all the lower profile that I can make it and still be a bracket. Once the paint dries, I'll reassemble it. I have stainless 8-32 screws coming in a couple of days. Right now it's assembled with galvanized 8-32 screws. While tapping out the holes that attach the sides, I snapped off a tap in one of the holes so there's a screw hole that can't take a screw.

 

[John in CR]

Keep up the good work EG. I think your threads would get more feedback and activity if you change your name. I've popped in to peek at some of your threads since joining, but it was despite being turned off by the name Electric God. A title like that no matter how cool it sounds is one that is earned, and even if earned (a long long way away), it wouldn't be a reference to one's self.

 

[ElectricGod]

The motor bracket is effectively done. I need to drill a few more holes for mounting it to the swing arms and the stainless screws will be here in a day or two. I also need to do some back wheel work before I can pick the final position for the motor.

 

[ElectricGod]

This was a comment on what John wrote, but I see now that he's not going to change his opinion...so never mind. It's not worth leaving a reply.

 

[ElectricGod]

I finished up the GAUSS meter today. I found a USB meter that monitors the voltage on the data pins. This meter is pretty ideal for the application. It displays the output voltage of the DC-DC converter and the hall output voltage. The only problem is the data lines only measure to a maximum of 3 volts and the hall can max out at 4.3 volts. I've added a longer extension to the probe tip so that the secondary volt meter stays within it's 3 volt range. I'm not using the other data pin volt meter so if I wanted, I could add something else to this little tool. There's a small LIPO cell sandwitched between the DC-DC/charge controller board and the USB meter. What were the main USB power leads connect to the 5 volt out of the DC-DC converter. The positive lead from the converter connects to a tiny switch so I can turn the whole thing off. The D+ voltage is from the hall. This is the quiescent voltage.

Here's the major parts I used.

49E hall, 350mah LIPO cell

DC-DC converter/charge controller
http://www.ebay.com/itm/USB-Lithium-Li-ion-18650-3-7V-4-2V-Battery-Charger-Board-Protection-Boost-Module-/192183192848?var=&hash=item2cbf02cd10:m:mp01J6C_p3WoPcNB2zU525g

USB watt meter
http://www.ebay.com/itm/332157810474?_trksid=p2057872.m2749.l2649&ssPageName=STRK%3AMEBIDX%3AIT

 

[ElectricGod]

On hold until I get the Currie going again. LOL

 

[ElectricGod]

The Currie scooter is running again as of last night, but that didn't stop me from doing a bit of work on the moped over the long weekend. I was never overly happy with the drive side of my motor bracket, so I made a new end plate and while I was at it, flipped over the bottom plate the the missing screw is now on the non-driven side. Why didn't I do that a long time ago? The new end plate has a bit more metal in it and the holes for mounting to the ears on the motor are no longer wallowed out. I think this will hold up much better. I cut off the corners on the other end plate, which didn't leave much metal around the outer portions of each hole. I figured it was a matter of time before the aluminum was stretching there and the motor was no longer a tight fit. Since I was working on the motor bracket, it was a good time to figure out how I'm mounting it to the swing arms. The part right under the motor and at the pivot will get a couple of large hose clamps to hold the front part in place. Then the back part will bolt down to the cross piece for the fender. Then a couple more bolts will go through that square tube and thread into the bottom plate of the motor bracket. 6 bolts and 2 band clamps ought to keep things pretty solid.

I also started to get the back end together. It's been months since I worked on this thing! I have to do some work on the body parts to make them fit around the new battery box. The charge port isn't an XLR connector yet and there's getting wiring from the controller to the power block and so on. I have the DC-DC converter mounted and that too needs to be wired into the terminal block and to battery power via the key switch. Things to do, things to do.

Well at least the controller is good for 4400 watts!

 

[ElectricGod]

More work done...

I got the motor mounted today. I looked at the two hose clamps and the angle bracket and it just wasn't enough so I drilled a couple of holes through the cross brace and tapped a couple of M8 holes into the 1/4" motor bracket bottom. That gets me effectively 2 screws that are locking it all in place, then 3 more into the smaller aluminum angle and the 2 hose clamps. It ought to be more than enough to keep things stable.

The motor once I had it in place was going to interfere with the wiring so I had to move the wiring a bit to keep it away from the blower. I moved the power block and DC-DC converter to the bottom side of the bracket and rerouted wiring a bit. I'll have to do more moving of things, but this will do it for now. I must have zip tied these wires and then moved them 3 or 4 times until I was happy that they were going to be in the best spot.

 

[ElectricGod]

I had to rewire the entire moped some time back. The factory wiring was garbage. I had mostly tested it to be sure it was working so I knew that the horn and lights were all good, but all of that had been done on a 3S LIPO pack. Last night I got the DC-DC converter, ignition switch and motor controller connected together. The twist throttle on the moped is aweful, but it works. Anyway, this is the first time that all systems were connected together and running as a complete unit. The motor ran, the lights worked and so on. Now I need to work out the rear wheel, chain, sprockets and brakes. It's time to hunt up hydraulic brakes again!

 

[gman1971]

HOT DAMN!!
What an amazing build man! Glad we crossed paths on E-S, even if we had a not so hot first post experience...

Thumbs up!!

G.

 

[ElectricGod]

HOT DAMN!!
What an amazing build man! Glad we crossed paths on E-S, even if we had a not so hot first post experience...

Thumbs up!!

G.

Back at you and thanks! No hard feelings here. This EV is close to done.