Zapino ZAP scooter upgrade

Status
Not open for further replies.

fubgumfaw

10 W
Joined
Sep 20, 2020
Messages
69
Hi people,

I'd add pictures, but I don't know how from my free photobucket account. Maybe I need some other picture sharing service? If you have ideas, that would be cool.

I've done lots of EV builds or helped other people with their EV builds. I've just never shared before in a forum. I usually take something worthy of upgrading (decent frame) and then work on the things that can make it into a good performing EV. I have a pretty good handle on how to build EVa, rewind motors, mess with electronics, build battery packs and so on. Most of the time I know what I am doing, but there are exceptions sometimes.

I found this Zap scooter on Craigslist and it's in pretty good condition. I bought it from a lady for $400. The SLA batteries were dead so there was some risk there might be problems with it. The worst damage on it was the rear side panes where the side lights mounted were broken. I had a wrecked Zap scooter a few years ago that I got for $50. It was thoroughly wrecked and destroyed. All the plastic cowlings were destroyed from who knows what abuse. Lights were broken or missing and someone had blown up the 12v system in it. I knew the motor was 12 phases and taking it apart, I saw how large the stator was so I had a good idea that it would be pretty strong if over watted. Too bad the factory controller is so tepid! The motor can handle lots of wattage that the factory controllers can't deliver to it!

This is soon after I got it. I took off the rear side panels to repair them.

Anyway, that wrecked Zap, I kept the motor and controller and various things that were salvageable. The rest went to the landfill.

This Zap scooter is in really good shape so it's getting upgrades in steps. I had some multistar 10,000mah packs that I made into 16S4P which is close enough to the 62v the SLA's made. At full charge I was seeing 66v on the LIPOs, but the factory controller didn't seem to have problems with a little too much voltage. I was happy to see that the scooter was 100% working and rode around about 50 miles on it as is. It was pretty boring since the scooter got to maybe 42 mile per hour and was super slow at getting there. One time going down a steep hill I saw 47 mph after going about a half mile down that hill. Going up the other side, the scooter quickly dropped off to 25 or 30 mph. Super duper lame!

I wanted 82v and planned on using 2 of whatever identical controllers I had on hand. I found out about Nucular controllers whatever that was...2 years ago and have bought 3 24 fet and 3 12 fet controllers. They were what I had that was 2 of the same controller. I was thinking maybe the 6 phase hubmotor might be capable of a lot more than 6kw like the 12 fet Nuc controllers can handle. I had one 24 fet already in an EV and another waiting to get used. That other EV lost it's controller so I could put 2 24 fet controllers on the Zap.

Like I said, I have done a lot of builds and helped others with them too. I have a lot of batteries from various sources and projects. I had an extended range Prius battery pack. It's filled with A123 LIFE cells which are 2500mah each. The pack has barely any miles on it so its cells were practically new. Under the floor of the scooter is a large space so I made a LIFE pack and got in 24S5P in the space...not too bad! It's only about 12.5Ah which is a bit smallish for capacity. The upside is it cost me $200 to build and that is almost 100% the cost I paid for the extended range pack from the grandma that didn't know what it was. I have a spot welder so after taking many of the modules apart to get the cells, I spot welded the cells together into a battery pack and then used an ANT BMS to manage it.

Those 10,000 multistar packs had friends from old EV builds or whatever. I took apart all those old battery packs down to the individual 4S packs inside, tested each one and to my surprise, most were still doing close to 10,000mah each. Any lower capacity packs got spread spread across the new 20S, 82v pack I made from them so that they were not all in parallel. I have them on another smart BMS in 20S4P. The BMS is saying I'm getting about 32Ah instead of 40Ah...oh well.

Then I had a bunch of LTO pouch cells...yeah LTO...not that great for charge density! I thought those LIFE cells were pretty pathetic for capacity, but the LTO cells are even worse! They have really good life cycles and for a forklift in a fridgid warehouse should do well in the cold. They are left overs from a forklift rebuild so who cares since they cost me nothing. I scrounged up an LTO capable BMS for them. I'm not sure of capacity since the BMS doesn't say so I'm trusting the cell specs or 11Ah. This is 28S1P. The cell voltages are a bit high as 2.8v per cell is really the upper limit and I'm running them at 2.92v each. So far no cells have bulged and I was unable to fit a 29th cell in the space. I don't really care, the cells were free to me and let's face it, they are LTO which takes abuse nicely! At least they got me a bit more capacity and filled the last space I had for batteries on the scooter.

So then now I had 3 packs capable of running at 82v, each on it's own chemistry specific BMS in parallel. This seems like a bad idea, but so far it has worked out OK as will be seen soon. At max power right now I am seeing 500 battery amps peak.

I like signals at the controller, not at the LCD like Nucular controllers can do. This lets me plug in any controller and there is no special wiring considerations to account for. I forget where I read about it, but some clever guy was using the exact same connectors on all his controllers, EV's and battery packs and wired identically to those connectors for everything. He used the same wiring scheme on his motors too. No idea who that was, but I thought it was a good idea and even found the same connectors he used and have in my own way copied him. Any EV I build or help build gets my "Standard" set of connectors and wired identically.

The dual 24 fet Nucualr controllers: I brought out the control signals in a cable that ends in an 8 pin IP68 connector. This is on both controllers. I can plug in the same throttle, regen throttle, 3 speeds, cruise and reverse signals into either controller individually if I need to. Both controllers use a 6 pin IP 68 connector for halls and temp and another for the canbus cable signals. The factory connectors on Nucualar controllers are not water proof or secure so they got cut off and replaced. I want connections to be as secure as possible. Another thing I don't like about the Nucualr controllers is the ring terminals for the motor phases. The battery wires use 6mm bullets which are really good connectors. I put them on the phase wires too exactly like I do on everything.

6 phase hub motor: I'm not a big fan of hub motors in general, but I have installed them for others and they do OK. This hub motor at 65mm tall on the magnets and stator was going to be pretty strong so I went ahead and used it. I'm not fond of 10" tires, but that's the rim on the motor...what can you do? I got a new set of Kenda tires as they ones on the scooter are something like 11 years old. Since I had the motor off the scooter to bench test with the 2 Nucular controllers, I replaced the halls, installed 2 PTC 10K temp sensors and opened up the motor shell. One of the things I don;t like about hub motors is the nearly total lack of cooling. I over watt everything and after riding around on 3kw, I knew closed up, there was a good chance the motor would get hot when running at higher wattage levels. The phase wires got 6mm bullets and the new halls and temps got connected to a 6 pin IP68 connector and wired exactly like I do on everything. THe motor got cleaned inside and out and then the magnet ring got painted with electrical paint to protect hem from possibly rusting. The motor end plates have 8 slots in them for venting. I put in new bearings too since one of the original ones was "crunchy" inside. No point in using new axle seals with the motor opened up. Opened up and 2 temp sensors to confirm with, I see the motor get to 110-120C.

I couldn't find a DC breaker with the amp capability I thought I might see so I used 2 breakers. One for each controller. They are optifuse 200 amp breakers. This isn't as good as the controller limits, but I was not expecting to see more than 200 amps per controller so I was not worried about it too much. The upside is that the entire scooter is almost entirely redundant now. I have 3 battery packs. Lose one and I keep going at lower amperage off the other 2 packs. Lose a controller or half the motor, turn off the breaker for that half and run solely from the other half/ The only non-redundant parts are the throttle and regen throttle and other control signals. THe factory DC-DC converter is rated for 60v to my surprise. It seemed to work fine at 66v. But I thought there was no way it would survive 82v so it got replaced with one good up to 90v and 25 amps. When I hooked up the 82v packs to the scooter electrical, the factory battery meter was something I had forgotten about. It soon died...oops! Oh well the Nucular LCD shows pack voltage instead of a dumb needle swing anyway.

After much time spent getting wiring, connectors and what not set up and rechecked several times, I put it all together in the scooter and started testing as a complete system. I did find a few problems to work out before I could go for a ride, but nothing major and no explosions or melted wires. I reused the scooter electrical system for the 12v lights and horn, but everything else got all new wiring. All the incandescent bulbs got replaced with LEDs including the halogen headlight.

Once I was satisifed that everything seemed like it was working OK, I took it for a ride. I had no idea what to expect from this hub motor so I started on my low speed setting and 75 phase amps. This was somewhat stronger than the factory controller setup and top speed was now 45 on level pavement. I soon wanted to try medium which set to 75% speed and 150 phase amps. I was sure that the amps would be too much for the motor, but it wasn't. The scooter topped out at 50 and accelerated nicely. After getting used to the scooter and how it handled, I finally tried my high speed setting which was 200 phase amps and 100% speed. I got a little better than 55mph which was disappointing. Acceleration was nice! I had read somewhere that this motor has 16kv. At 82v, that should have gotten me about 62-64mph. I think 16kv is wrong. I then turned on field weakening to 130% and that got me to 65 no problem. First time at this amperage was pretty thrilling as the motor pulled quite well and the top speed on this scooter was what I was expecting at 100% speed. I've since tweaked a bit more and checkened out at 70 mph and 230 phase amps.

All that motor torque on the axle flats created issues. A few days ago I discovered that the factory torque arms were spreading and about to fail completely. THe flats in the drop outs are also getting stretched. What got me to look into this was an odd clunking sound I would hear when I went to accelerate or use regen. The axle was rotating back and forth instead of staying put. I have bought some 4140 steel so I can machine stronger torque arms. I think that will hold the motor torque.

I have discovered another problem. At full acceleration on high, I see peaks of 250 battery amps and 309 phase amps per controller. This is at the upper limits of the combined battery packs. There is a stuttering thing that happens when I crank the throttle wide open at full power. I think maybe my BMS's are going into overload and shutting off. Maybe the breakers are opening their contacts, but never quite pop open? I still have to find out what this is about.

Last Saturday, I did 2 charges. The first one was just to test out my latest controller settings at max power. As long as I stay under that stuttering level, I get pretty good acceleration. I have not measured it yet, but I have had fast cars and I know what a bike feels like accelerating. I think I'm currently at about 0-60 mph in 5 seconds or less. No idea what it will be like once I overcome that stuttering and have solid torque arms on the motor since I can't currently find out. The second charge was to see how far I got on a charge. I want 70 miles or better so I can commute to work and home on a single charge. On my medium setting, I'm seeing 48 miles. On low...well who cares as I won't be using that very much if ever myself. High and medium both tell me the same thing. Yank out that brand new LTO pack and replace it with a better battery tech and probably pull the LIFE pack too. I have a lot of LION cells that I was trying to not use, but it seems that is exactly what I need to use in the scooter to get the range I need.
 
Imgur won't let me post pics inline for some reason. I guess it's the free account or something...

So here's my 6 phase hub motor now. I machined vent slots in the shell. I painted the magnet ring with red electrical paint so they hopefully wont rust. These motors lack a proper temp sensor in them. They have temp switch...pretty useless for motor temp! I added 2 PTC 10K sensors in the phases...one per controller.

https://imgur.com/Aaa6pJy
https://imgur.com/buWMudV
https://imgur.com/tedvKHg
https://imgur.com/B7wkvfC
https://imgur.com/wGqJDo6
https://imgur.com/RSHJXes

This is the 3 original packs I put in it. Front one under the floor is 24S5P of A123 LIFE cells for about 12Ah. Middle pack is 20S4P of LIPO for about 32Ah. The back pack is 28S1P of LTO for about 11Ah.

https://imgur.com/GtlYaRW

I got 12" cast wheels for it. I really just cared about higher speed tires, but I already had the best there was in 10" and they are rated for 60mph. I was doing 75mph so the tires were a concern. In 12", much higher load ratings and speeds are pretty common. This is a Dunlop D604 tire on the new wheel. I fitted a much larger rotor and spaced the caliper out further for improved front end braking.

https://imgur.com/nFPeaz6
https://imgur.com/vanFAwZ

The back wheel was a bit more work. I needed to machine out the center section of the wheel and add the 12 bolt flange pattern the motor has on it.

https://imgur.com/JUJQxwN
https://imgur.com/YhJLYVa
https://imgur.com/Kqhuxnc
https://imgur.com/ahGRP9c
https://imgur.com/Z6rLZYI

My latest addition. A 20S10P pack made from 200 4800mah 21700 cells on a XiaoXiang BMS. I estimate real capacity under load is more like 40-43Ah.

https://imgur.com/O2J8AXl

This isn't quite current. This is the underside of the seat compartment. The scooter uses 2 motor controllers so they each have a watt meter on them. The white wires are the BATT+ wires to the controllers. Inside the compartment are dual 300 amp Optifuse breakers.

https://imgur.com/cvnJtNv
 
Both controllers report peaks of 309 phase amps. Battery amps until I took it apart again were peaks of 250 amps per controller. So that's 25kw peaks in the phases per 24 fet Nuc controller. 16kw per controller with semi-aggressive riding is pretty common. The factory torque arms were garbage and incapable of holding the 50kw peaks I was seeing in the motor. The factory torque arms and the drop-out flats didn't last long with that much power on them! I think I got like 2 hours of use before they were gone.

https://imgur.com/RZYCB2J
https://imgur.com/anY5scQ

I machined these torque arms from 1/4" thick steel. They are already showing signs of wallowing out a little. I need to make another set that clamp to the shaft flats for best possible holding power. While I'm in there, I'll widen the drop-out flats so the torque arms can fill inside the drop-outs and grab as much shaft flat as possible.

https://imgur.com/zjwjtJD
https://imgur.com/Lu7vV6e
 
What I call the front battery pack is under the floor. I found a bunch of 10S mobility scooter packs. They were low use and something in the BMS firmware is wrong so they got pulled and I was able to get 12 of them for a low price. They are filled with LG MH1 cells in 10S4P. Testing them out, they showed new cell capacities. This was pretty cool since I paid about $1.40 per cell.

This is the scooter packs after I stripped off the outer layers and the onboard BMS.

https://imgur.com/9SSXXUR

I cut them in half as that made a "block" that would fit inside my frame space pretty easily. A little soldering reconnected them together again in 20S4P. In retrospect, I could have fit more capacity in the space if I would have harvested the cells and packed them into the space and then rewelded them into a pack. Oh well, this worked out well enough.

https://imgur.com/ZbcyiMH

This is an early mock-up of the a few of the half sections next to the LIFE pack I originally built for the space from A123 cells. The A123 cells are rated for 2500mah in 24S5P or about 12Ah at 82v. They were recovered from a semi-used Prius extended range pack I took apart. They still tested at close to new capacity, but the LIFE cells were simply not a great solution since the capacity for such large cells is so low.

https://imgur.com/uJKlRHd

This is the final pack or 20S12P or about 34Ah. The new pack is pretty close to the same size and weight to the LIFE pack and about 2.8X more capacity! IF I had stripped the cells out of the blocks and rewelded a new pack from them, I probably could have gotten 3X or more capacity over the LIFE pack.

https://imgur.com/00iVQB7

This is the LIFE pack I had in the front space.

https://imgur.com/SQ21I6w
https://imgur.com/4K2M6hs
 
The rear pack was originally made from 11Ah LTO cells in 28S1P. The cell voltages (2.92v) are a little high, but LTO is pretty tolerant of abuse and you can't ever make LTO catch fire so I wasn't worried about it too much.

This is the LTO pack I had over the back wheel.

https://imgur.com/tiltXlb
https://imgur.com/UZIu4lt

It has now been replaced with this 20S10P LION pack made from 21700 cells which I built just a few weeks ago. This is about 42Ah or just short of 4X more capacity than the LTO pack!

https://imgur.com/O2J8AXl
 
I have 3 packs in this EV for a total of about 108Ah now. Originally they were made of LIPO, LIFE and LTO each on a different kind of BMS. The 3 different apps and issues with the one BMS made me standardize on 3 JBD BMS's instead of bodging it all together like I did originally.

I had these BMS's on the original 3 packs:

The LIPO pack is still in the EV. It's made of 10Ah Multistar packs in 20S4P. I made up a bunch of Y cables so that all those balance connectors could come together to this one 20S BMS. It's a bit of a mess and the LIPO packs are of various ages and levels of use, but it works and the banks all discharge at pretty close to the same rates so...whatever. I'll probably replace it after a while...

https://www.aliexpress.com/item/1005001705251281.html?spm=a2g0s.9042311.0.0.31614c4dQvxJQw

The rear LTO pack was on 4 of these smart BMS's. The app is not so great, but the boards work well for balancing just about anything. The hardware implementation is pretty good. You can use them individually or daisy chain them together over fully opto-isolated canbus up to 16 boards in series. The LCD is a there for viewing status only. My BT was unreliable. I think I have a defective BT module. I'd have to unplug it from the chain to get it to work and then it would stay working just 5 or 10 minutes before it was down again.

https://www.aliexpress.com/item/32861093603.html?spm=2114.12010612.8148356.26.64f62eb1Q4LBaa

The LIFE pack was on an ANT BMS...which are pretty nice. I like them quite a lot! The LION pack going in its place will get a JBD BMS like the other 2 packs.

https://www.aliexpress.com/item/32856349172.html?spm=2114.12010612.8148356.6.79513de29lCICG
 
Finally! Tracking from ICGOGOGO for the last BMS came through today. Hurry up and wait...LOL

I'm also waiting for some proto board to arrive. I have been talking to Vasily from Nucular about a problem with multiple controller setups. My first 2 Nucular 24 fets died mysteriously of small electronics failures. There was no evidence of over heating and no mosfets blew. It had to be related to the fact that CANBUS, enable and 12v lines are not electrically isolated or protected from each other. With multiple controllers, you connect the CANBUS, enable, ground and +12v lines together between the LCD and controllers. This creates possibilities for voltage spikes, current loading and who knows what causing issues. I have purchased some components from Mouser that will help control these things.

I don't want to power up the controllers until I have that protection in place. It's easy to build, just need the protoboard to put it on.
 
The scooter has been out for rides a few times now.
This is the second set of 24 fet Nucular controllers that have been on it and both did this violent stuttering thing. The answer, turn down phase amps...which isn't an answer, but at least it makes the violent ratcheting or stuttering much reduced when I crank the throttle.

I had posted in the KWELD thread since I have one of those and a kind soul pointed out how to make my imgur pictures work right. Editing... :)

I did some machining and made new torque arms. This is 1/2" thick steel plate. I made both arms at the same time as much as possible. I'll let the pictures do the talking...

lJsB8Nx.jpg

hMCR5BA.jpg

[imghttps://i.imgur.com/yGyrhdL.jpg][/img]
nIolcJ5.jpg

u9q61rR.jpg

HtPl7RV.jpg

kjspkMr.jpg

5LPScKT.jpg

juZMyVZ.jpg

cXVQtSL.jpg

jxZhpnQ.jpg

zbrDfPz.jpg

ZfSszEq.jpg

776H8Rw.jpg

KtnzTBd.jpg

j7HCw48.jpg


The swing arms needed some modding to fit the new torque arms so they got several things ground off them, some rework at the drop-outs and a new paint job.

qNJEk5Y.jpg

MSDHC0X.jpg


They are working great so far. Double M5 screws clamp the torque arms to the entire motor shaft flats on both sides.

aqzUwaR.jpg

9WbtKSC.jpg
 
ridethelightning said:
awesome job! ill be following this.

Do you suppose there is irony in this? This poor guy has been trolled by a single individual and has had his account locked out for the second time now because of a single troll that continues to go unpunished for his actions! I'd say "69" is exactly what he is getting! AKA f-cked by the moderators and f-cked by the troll!

fCbC8Fi.png
 
The front end looks odd with no forks on it.

mWkHpaE.png


I took apart the shocks since they were too soft/squishy. I found that the left one had a leaking seal and neither had very much hydraulic fluid in them. They were basically air filled and that explains why they had so little damping. Air being compressible is a good way to create springiness, but not a good way to create damping. For that you need something that does not compress and these shocks were nearly empty of hydraulic fluid. I have purchased new shock fluid and seals and they should be here Monday. Gotta love how fast Amazon is!

r6SSh0c.jpg

tFQXVkN.jpg
 
rishiimaze said:
ridethelightning said:
awesome job! ill be following this.

Do you suppose there is irony in this? This poor guy has been trolled by a single individual and has had his account locked out for the second time now because of a single troll that continues to go unpunished for his actions! I'd say "69" is exactly what he is getting! AKA f-cked by the moderators and f-cked by the troll!

fCbC8Fi.png

69? id say cut to the DVDA. its F^^D

I am looking forward to more on this thread. there really arn't many threads with guys as handy as this involving hubmonsters and crazy power, and its very similar to my next project that's lined up.
 
btw I looked at my phase to phase resistance on my hubmonsters, I have got a pretty decent Fluke but the readings were pretty wired. some showed 0.1ohm, others like 6mohm and fluctuated a lot, maybe its too low for my meter to read? I have no experience in this dept.
 
rishiimaze said:
ridethelightning said:
awesome job! ill be following this.

Do you suppose there is irony in this? This poor guy has been trolled by a single individual and has had his account locked out for the second time now because of a single troll that continues to go unpunished for his actions! I'd say "69" is exactly what he is getting! AKA f-cked by the moderators and f-cked by the troll!

So funny to see someone sign up from a foreign VPN just to defend one of EG's fake accounts..

This thread is now locked and only kept for the technical reference about the hubmonster.
 
Status
Not open for further replies.
Back
Top