PWD's Yamaha XV250 Virago build

[pwd]

In case you haven't seen this series of posts about the 3shul:

Yamaha YZ450F

I even have these microballoons: I haven't used this, but something similar a long time ago. But I thought I would be able to get the fiberglass down there as I had time to go back and forth many times. I was thinking it would be better in the end to get the strength of the fiberglass...

endless-sphere.com

Thank you, I did see that but will be watching the discussion there too.

 

[pwd]

Added a side bar to the center battery tray. The top cells can be added with the tank removed.



Refitted to the bike. The center battery pack will be added from the side and then have that metal bar bolted on.


I finally settled on where to mount the controller. This will get it good airflow and allow me to fit all 30 cells in the battery tray,



The bottom half of the controller mount will bolt into the factory engine mount as well. I still have to add an upper bracket for the controller and will have to extend the phase cable by about 12cm. This photo is showing the controller "swung-down" without a top bracket in place.

 

[pwd]

I've been trying to go full steam ahead on this build while I can. The used 3shul cable harness had been chopped a bit so I refitted the throttle/regen connector and refitted a longer motor hall / temp connector to plug into the motor.

I then realized I could test out the motor and controller or at least a "bench" test after some temporary wiring and finding a battery pack to use. Since the controller was previously used on a different motor, I was having trouble doing FOC detection. It then occurred to me to try reverting motor settings to default and trying again - after doing that and following this video:

I got the motor and throttle working in a couple of hours.

Although the real test will be on the road; it brings me some relief that the used motor and controller do work up to this point.

There is a lot of work to go on the battery pack and battery circuit but I am starting to see the light.

Some observations /personal notes so far: prismatic cells aren't really much easier to work with; just different. I should have bought inexpensive crimping tools years ago - they make things so much easier. Same goes for a basic hot air / reflow gun - even if just for heathshrink items.

 

[ZeroEm]

That is a good looking Bike. Just wanted to let you know.

 

[Tony01]

Nice. I think I took my license test on one of those in a course. What’s the ending weight? I bet it’s about the same as before but way faster….

 

[ZeroEm]

1976 took my license test in a 1973 impala, 350ci maybe 130 hp, 10 mpg. Had my learners permit and was driving on it for a year, lol.

 

[jkrienert]

This photo is showing the controller "swung-down" without a top bracket in place.

Really digging the accessibility of the controller for troubleshooting, while still being in a location where its going to get maximum passive airflow during riding. That is some clever, elegant, and functional design.

 

[pwd]

Nice. I think I took my license test on one of those in a course. What’s the ending weight? I bet it’s about the same as before but way faster….

By the time I'm finished it might be a bit heavier; but should have about twice the power.

 

[pwd]

That is a good looking Bike. Just wanted to let you know.

Thanks, although it will be notably different looking when its running; keeping the original tank in place will help keep some of it's original look.

 

[pwd]

Among other things like installing balance leads on each sub pack; I have decided on a place for the contactor but realized I had no good way of mounting it there. I added a bracket under the lower sub-pack so I can bolt on the contactor and keep the cables fairly short.




I removed the battery trays again (I'm getting good at this) so I could have good access to grind/weld.


Next on the list is to complete assembly of the final sub-pack that goes on top directly under the fuel tank.

 

[pwd]

All the sub-packs are assembled but I needed to add some way to secure the top pack. The top tray has gaps and voids that the pack can slide around on. I changed my plan to have the cells in two rows. It is now a simpler, single row of 7S for the top pack.



Here is how it looks will all 30 cells on the bike (but not secured):


Some more observations: the battery pack and steel trays I build are heavy - my gut feeling tells me the bike may be a tad heavier than with the ICE by the time I get it road worthy and I may have to look into stiffer fork springs depends how the first road test goes.

I've been working on the BMS wiring. I've chosen Molex connectors for each sub-pack to make things easier to connect/disconnect as I work on the bike.



I did have a small thermal event when trying to fix one of the pins on the bottom sub-pack - two pins shorted and caused most of the balance leads to melt smoke up. Note to self - remove balance leads from the cells before trying to de-pin the connectors. I finished re-making the balance leads that melted - thankfully nothing else was damaged and I had enough extra Molex pins.

Life isn't slowing down so I'm in a bit of a crunch to get a road test in before my next (planned) major life event.

 

[harrisonpatm]

Note to self - remove balance leads from the cells before trying to de-pin the connectors.

It's a mistake you'll only make once! Luckily, those relatively long and thin balance connectors would never be able to carry so much current to cause a thermal event; the wire would melt before the battery went into runaway. Still, best to avoid if you can.

 

[SlowCo]

Will be some extra work but I would adjust/reweld the top rack to narrow it and fit to the width of the battery cells. And put some rubber strips under and around the battery cells to provide a little cushening and protect them from vibration.

 

[pwd]

I have been putting as much as I can into this project to get it rolling.

It's a mistake you'll only make once! Luckily, those relatively long and thin balance connectors would never be able to carry so much current to cause a thermal event; the wire would melt before the battery went into runaway. Still, best to avoid if you can.

It seems it took a second time for me to learn.

I've completed mounting and installing the BMS but had another thermal event (but got lucky again) when I was trying to squeeze the BMS lead for battery negative past some cells. This let out an audible *bang* sound as it made contact with an uncovered cell terminal. It scared the crap out of me, melted a copper lug and some of the aluminum on the cell. After assessing the damage, I was able to surgically replace the melted lug on the cable (split it open with a rotary tool and carefully pried it apart) since I didn't have any extra cable and didn't want to re-make the whole thing. I also cleaned everything off with isopropyl alcohol and make not that I got lucky that none of the cells were significantly damaged from this.

After that mistake, I decided to take the time to make sure all the cell terminals and cable harness lugs were covered - they aren't perfect since you can still see a bit of the aluminum post base but its a big improvement from having nothing.

I decided to put the 13.8V step down converter and fuse block in the old starter battery tray which goes under the seat. I cut a few holes in it to run the leads out to the different 12V circuits and called it a day. The pre-charge resistor, and time delay pre-charge relay will also sit inside this tray with all the "rats nest". I'm quite proud of how the pre-charge setup works and glad I decided to spend the few extra coins on a "Bueler BU510TD" adjustable timer relay - it works very nicely.



For now, I have not wired in a key-switch but do have the main contactor connected to the bike's kill switch and controller ignition hooked up to the side-stand switch.

While I was testing the ignition circuit I noticed I had a dead cell... ironically it was one of the replacement cells that the seller had sent - thankfully it was on the top row of cells which saved me from having to take everything apart once again. After cutting the cable ties and replacing the cell - the voltage was back where it should be and I'm glad I noticed before I started charging or using the bike.

I ordered some more terminal boots to cover up the posts on the controller, since that would be a very expensive mistake if I were to short something there.

With the starting and contactor circuit wired up, I added some connectors to the charger cable leads and have the bike going through it's first charge cycle.... just about ready for it's first test ride.



It's finally coming together.

 

[bananu7]

I think building batteries "in situ" like that is generally speaking, quite an unsafe idea. I feel semi-comfortable with cells in a controlled environment when i'm putting them together. After that, I want everything to be protected by the BMS and/or a switch. That minimizes the chances of anything being shorted during assembly, and ensures the cells are protected from a runaway at all points. The battery basically becomes it's own self-contained, self-managing unit. As such, there's also no needs to make balance leads removable, as it shouldn't really ever come out in the open, unless for very rare cell replacement.

As my bike has seen a couple of months of use at this point, I can say that I'm very happy I went this way. The battery has two terminals and nothing else in the open, they are always protected from overcurrent and shorts, it can be disabled quickly and precharges automatically. There are no external components such as precharge circuit or a contactor, which makes wiring simpler and more robust.

 

[pwd]

I'm happy to report I've been riding the bike when I can. The first ride was very encouraging and I slowly turn the current levels up. I'm running about 700A max motor current and 250 max battery current. The top speed is less than expected at around 87 km/h.

I ran for a couple weeks without regen. I have since re-purposed the clutch lever to engage variable regen with another throttle box. I had to add some springs to the lever since the spring on the throttle box felt too weak. The variable regen feels like a luxury and really takes the bike to a great level of control.

One issue I've been having is the chain has been skipping. The first time I removed it, cleaned it, re-lubed it and tensioned it and it worked well. After a couple days it began skipping again so I re-tension it. Perhaps it is stretching or is worn too far for the stress I'm putting on it. Next in line is a new x-ring chain to keep maintenance to a minimum and going down from a 45T to a 39T rear sprocket to get a bit more top speed.

The bike is still running with uncovered battery cells and BMS but will run it like that for the rest of the season. Next year I will have to design some covering with pliexiglass that I have on hand. The bike feels great overall.

 

[pwd]

Although I don't have time to work on this bike - I still have plans to address the chain skipping issue. What I've gathered so far is that:

I may need to re-inforce the motor mount to prevent flexing - add some bracing at the top of the mount to the frame

the motor/front sproket *may* be sitting up too high causing the suspension to "squat" under load- I may try removing the square tubing spacers which will lower the sprocket and possibly take away some flex due to the small guage square tubing



once the skipping is corrected (I'm hopeful that it can be) - this bike will be able to take the full beans

 

[adsc]

It's a difficult decision, hub vs mid. With hub you don't have all these chain problems, but you have new problems like having to fit a new brake and like you said the weight of the rear wheel. Also mounting it is of course non-trivial itself. And I don't think QS makes a high powered 16 inch hub motor, so you'll have to go 17 inch, which is not natural for a chopper.

But you gain space and have much less noise.

 

[CRF265]

That top speed seems alot slower than that motor/controller/battery combination is capable of. It looks like you are running 14:45 ratio which is a final ratio of 7.55. I am running the qs138 90h and have found this ratio on this motor to be too short for the road.

Going to a 39t rear will improve top speed heaps and be a ratio of 6.55 - expect a top speed of 110kph with your 30s setup and wheel torque should actually be higher at 70kph+

 

[pwd]

It's a difficult decision, hub vs mid. With hub you don't have all these chain problems, but you have new problems like having to fit a new brake and like you said the weight of the rear wheel. Also mounting it is of course non-trivial itself. And I don't think QS makes a high powered 16 inch hub motor, so you'll have to go 17 inch, which is not natural for a chopper.

But you gain space and have much less noise.

I though a lot about using a hub-motor on this bike but I am enjoying the mid-drive even if I have to deal with the chain etc... once I have the skipping issue solved; the motor adjustment should be complete. The little bit of chain/gear noise is just enough to hear the rpm changes which retains a small amount of audible feedback I wouldn't get with a hub (I have a mini moto with a hub + FOC and it is nearly silent). Also, I like the idea of keeping the weight in the center of the bike and having a normal rear swingarm and wheel that is just as easy to service as it was when it was ICE.

That top speed seems alot slower than that motor/controller/battery combination is capable of. It looks like you are running 14:45 ratio which is a final ratio of 7.55. I am running the qs138 90h and have found this ratio on this motor to be too short for the road.

Going to a 39t rear will improve top speed heaps and be a ratio of 6.55 - expect a top speed of 110kph with your 30s setup and wheel torque should actually be higher at 70kph+

Funny enough, I have a 39T sprocket sitting here that I ordered months ago - I figured the taller gearing will add more stress to the chain so once the chain skipping is resolved I will try out the 39T.

I've begun the process of lowering the motor position to help with the chain skipping issue - from what I read it will also provide some anti-quat attributes that will help. I will be able to remove those weak steel square-tube spacers and mount directly on the motor supports or I can use some aluminum flat bar to raise it up just enough so the chain doesn't rub. Since the lower battery tray uses the same bolt as the motor mounts; I will add some spacers (which should be much stronger and distribute the load better).

 

[adsc]

Yeah, I'll probably also have to go with a mid drive in the end. My issue is getting it street legal in Switzerland. Apparently, if you make any changes to the structural parts of the frame, you have to do a very expensive (thousands of dollars) test at a proof center to make sure it's still structurally sound. I hope I can avoid that if I go with a mid drive and just build a mounting plate that interfaces with the normal frame mount points.

I thought about making clamps for the hub motor that can be mounted to the swingarm without having to modify it. Would lift the bike up 2 cm at the rear, but I think since the wheel is mounted there, those clamps would also have to be tested. So I think I have much better chances if I can keep the whole rolling frame original. Still very questionable if I'll ever be able to ride it legally.

 

[pwd]

I got the spacers and aluminum bar in place to test and looks like it will work OK.




With the chain tensioned, the chain *just* clears the plastic chain guide. This give the front sprocket a lower position in relation to the swing arm pivot to (hopefully) help with squat under acceleration. Next on the list is to clean up/grind the motor supports and get some paint on that area while I have the motor out - something I should have done from the start before I added everything to the bike. I also added a small bit of metal on the chain side motor support plate to reduce any flex in that area.

 

[bananu7]

With the chain tensioned, the chain *just* clears the plastic chain guide.

Make sure you didn't overdo that. With a tight chain and a sprocket high the chain can eat the underside of the swingarm. Slightly resting on the guide when bike is upright (and thus barely clearing when suspension is compressed under rider weight) is typically optimal at least for long-travel offroad bikes.