wr250f conversion

Ha, nah. I pulled the battery apart on friday and it isnt too bad. Poor design, bottle necked and melted some copper. The nearest 4 cells probably got a bit toasty. Short of throwing away 180 cells i have carefully cleaned it up and measured capacity and IR. They check out fine but they will no doubt have a shorter cycle life. No self discharging. I mulled on it a bit and stripped the 3 internal "packs" down to the copper and am beefing up the series connections. About half way done doing that. Then i gotta re-glass the pack. I will be working on some other parts for this bike though. Need to spin some PCB's and design a cluster that can talk directly to the controller.

You make any progress with your YZ?

Doug
 
I suppose i didnt really talk much about the drag racing. I like drag racing as a metric of what a vehicle actually goes like, regardless of weight, power, aero etc. My conclusion on this bike is its pretty darn quick. As mentioned i had some mid track dramas, probably due to the battery copper turning into a heater :lol: but even so, faster to the 330ft then my tiger 800. Was only .1sec slower to the 1/8th then the tiger, albiet well into the heater portion of the run. Late track still running 13's at 140kph. My tzr runs 13's at 160kph. Tiger makes 90hp and weighs 255kg, TZR makes 50hp and is about 140kg. WR is about 138kg. My gut feel is i was pretty much right on the tzr for power, probably 40ish HP at the wheel. I will get it on the dyno but those are some comparisons to actual bike numbers.

On the road it completely obliterates both tiger and tzr for feel. Those times are when launching the bikes. For the tzr and tiger, clutch and throttle dance is critical, clutch on tzr lasts 2 dozen or so and tiger spins the rear wheel. You never launch that hard on the road. The WR.... Smack it to WOT and it will pull a 2s 60ft every time 8) .

Doug
 
Im so in the controller stuff that I forget there are some other things to deal with :wink:
If I got it right you lay the fiberglass as it is on the cells, and shape it around the cells?
So when you want to get in, you have to hack trough the fiberglass?

dougf said:
You make any progress with your YZ?

Doug

Hmm, not really. I have more or less given up with getting the sevcon to do what I want.
I think I should update my thread with a rant about alibaba sellers :evil:

Your tests on the strip sounds promising, there should be power potential :D
 
Yeah. Will be going again soon. I sort of semi vac bagged the battery last time. I was afraid to pull a hard vac on the cells so i only did enough to hold the glass down. It worked ok and was pretty tough but looked a bit crap. I think this time i will wrap it in shrink tape instead of the vacuum. I will also fill and fair around the cells to stop it pulling in. Yes, getting into it requires a hour or 2 with a dremel carefully slicing into the not cells part of the battery then ripping the glass off. It wasnt built to come apart. The original design was going to be an aluminium enclosure but it became clear that i wasnt going to be able to weld to the precision required for it to work. Perhaps i could build a jig and try again. Is probably heavier then the glass though and my bike is already chunky.

I got my controller from the DUN electric store. I didnt buy in aliexpress but msged the seller directly as my controller was customized to use the encoder. As far as aliexpress sellers go they were helpful and delivered.

Shame about the sevcon. It is a problem ive seen with the stuff i build for work though. More features = hard to commission. If the tech cant do it then "your product sucks". I would definitely try and get in touch with some of their payed support. Failing that, send it to me, ill dyno tune it on my bike and send it back 8)

Doug
 
Your bike is such a ripper!! What a beastly powertrain!

Part of me wants to re-build your pack for you into something that let's you see triple digit HP, but I'm sadly way too busy with lab work.

I hope you do though, and let that machine set some records and destroy all the ICE bikes it faces.
 
Ha, yeah that would be cool, but maybe with a more suitable bike. This is supposed to be a dirt bike, not a widowmaker :lol: . Id say im 100% going to build a road bike at some stage and yeah, i want north of 100kw :twisted: . Thats a future project. I want to put a bow tie on this bike and enjoy it.

Doug
 
dougf said:
I would definitely try and get in touch with some of their payed support. Failing that, send it to me, ill dyno tune it on my bike and send it back 8)

Doug

I tried "kit elec shop" where I bought it recently, but so far no reply.
I find it a bit odd, they seemed very helpful and interested at first. And seemed genuine, not just "buy my stuff"
But later when I asked questions, no reply or worthless replys.
And it is not like I have asked about everything, I have sent 3-4 questions.

I think you are in Australia? It seems a little far, and it has to pass customs on the way there and back..
I also doubt it can handle your encoder. It can handle A and B, but I havent seen anything about Z.

Have you determined phase resistance, inductance and pole pairs on your motor?
 
I Msg'd the DUN store to see if they can make or have a 1081800.

The local Sevcon guys refused to sell me a drive. One didn't reply at all. So screw them. I've had pretty good support from both QS and Fardriver.

Yeah In Australia. Import taxes can be reduced because It would be sent to AUS for "repair" and then returned. Was a bit of a jest though, there will be places with dynos in the states who could give you an hr on a dyno for not much. Though I feel like what you need is someone who has experience with those drives.

I measured the inductance of my motor when I first got it. Cant remember the exact numbers. I will take the LCR meter home tonight and test it if you like. The ohm DC resistance is basically 0 and my fardriver complains about it on first powerup but that could be due to the precharge circuit being a bit slow.

Doug
 
1800Amps phase on a bike let's you have powerband flexibility.
 
Got a reply from DUN. They aparrantly do have stock. There was a shortage. They are going to confirm after the holiday (6th of June).

Yeah, the APT was maxed out. I should be able to put this bike on my dyno now and bring up each phase point until power output change doesnt match power input change. I promise i will do all this once i get the battery back together. Should be running again this week.

The RS485 out of the controller isnt Modbus, looks like they use hex AA as the leading sync. Ive only had a brief glance so far. I will update when i know more.

Doug
 
dougf said:
I Msg'd the DUN store to see if they can make or have a 1081800.

The local Sevcon guys refused to sell me a drive. One didn't reply at all. So screw them. I've had pretty good support from both QS and Fardriver.

Yeah In Australia. Import taxes can be reduced because It would be sent to AUS for "repair" and then returned. Was a bit of a jest though, there will be places with dynos in the states who could give you an hr on a dyno for not much. Though I feel like what you need is someone who has experience with those drives.

I measured the inductance of my motor when I first got it. Cant remember the exact numbers. I will take the LCR meter home tonight and test it if you like. The ohm DC resistance is basically 0 and my fardriver complains about it on first powerup but that could be due to the precharge circuit being a bit slow.

Doug

I ordered a 961800 about 2 days ago, it will be annoying if it was possible to find 1081800 :?
I got a bit fed up after I had ordered and payed for two controllers from (the only) sellers who clamed to have them in stock.
So far I have got the money back from one of them, -$100 that got lost in fees.

I am in Sweden, I can find dynos locally but the only sevcon representative I have found was a forklift company you need an account to get in touch with.

You dont have to go measuring, I just wanted to compare if you had the numbers. I was told 4 pole pairs, and I think I was able to confirm it was correct. I am pretty sure I have seen 5 pole pairs in some specs too.. I measured 130uH inductance from phase to phase, it turned out to be 65 I should use. And about 5mohm phase to phase.
 
Yes 4 pole motor. Phase inductance varied from around 42uH to 60uH (depending on motor angle) when i tested with a pocket lcr. I had a bit of a crazy day and completely forgot to grab the benchtop LCR meter. I need to dash back to work tomorrow morning, ill grab it then.

Doug
 
j bjork said:
It has been a long time now, any news?
Have you done any more controller tuning? :wink:

Nah unfortunately not. Had a few bad weeks with other stuff. I have finished repairing my battery though, just needs sanding and painting. Once that is back in the bike I can start reverse engineering the RS485 from the controller. I want to be able to use the controller's SOC instead of the external system. Its probably not as accurate but for a dirtbike I really dont mind. There might be other benefits too.

I heard back from DUN again, no stock of 1081800's.

When I get the bike back online will post some updates.

Doug
 
how is the direct drive with the qs 180 working for you? i have also ordered a 180 90h and 721800 and am still deciding if i should do direct drive.
 
Seems to work. The front sprocket is at least a few inches further forward which requires a significant amount of chain slack. Not as bad as 80s macios but more then a wr250. I have mostly been using 11/60 gearing but have used as tall as 11/52. Torque at the rear wheel is good and top speeds over 120kph which is plenty for me. I havent noticed any unwanted behaviours in handling. Personally i didnt want the hassle of a second reduction. It is important to make sure the chain across the top isnt getting pulled into the swingarm otherwise the chain guide will last about 3 rides.

Doug
 
could you show more pictures of your motor mount? my motor arrived today and im doing a direct drive like you.

Thanks
 
Nice craftsmanship.
I'm curious about the machined plastic used for the battery box ?
 
TysonScott said:
could you show more pictures of your motor mount? my motor arrived today and im doing a direct drive like you.

Thanks

Hi Mate, these are the only 2 I dont think I posted earlier. I basically machined supports to go onto the swingarm pivot which then went to threaded rod going all the way across the motor.
machining.jpg
on motor.jpg
Its not a particually good solution if Im honest, I ended up removing some of the cooling fin closest to the threaded rod and the aluminium ends up pretty thin between the threaded rod and the motor. I then just support the front of the motor in a similar fashion with threaded rod. Because of the ADR requirements I cannot modify the frame without it becoming significantly harder to get registered. If your just doing a dirtbike then I would probably get a mate with an AC tig to help you put more appropriate mounting points on the bike.

I would also not dismiss the jackshaft out of hand, having the primary reduction will mean stock rear sprockets but it also will allow you to move the motor forward to where most of the conventional motor mounts are located.

Doug
 
PaPaSteve said:
Nice craftsmanship.
I'm curious about the machined plastic used for the battery box ?

The white plastic is PVC foam board. Used in sign making. It machines really easily, doesnt change shape like hdpe and is incredibly light.

Doug
 
A small update I suppose. Bike is going again. Confirmed with a shed burnout. I wrote some software to help me decode the rs485 comms between the pc and controller. For anyone interested you need to send a command which is buried in my laptop right now, afterwards the drive will stream back information. The stream is formatted in hex and your software will need to convert to decimal to get human numbers. The structure is 0xAA as leading sync, followed by an ID bit going from 0x00 to 0x17 (basically 24 different id's), followed by data. At a glance 0x01 has voltage in the 3rd and 4th byte. It might be using a checksum at the end, I still need to check. It took me an embarrasingly long time to write the VB and afterwards I was sort of over it so I put it aside. Will do some actual science this weekend. But it looks like it should definitely be possible to read the SOC information from the drive and display it on a screen.

Doug
 
dougf said:
PaPaSteve said:
Nice craftsmanship.
I'm curious about the machined plastic used for the battery box ?

The white plastic is PVC foam board. Used in sign making. It machines really easily, doesnt change shape like hdpe and is incredibly light.

Doug

Thanks for the tip :thumb:
 
j bjork said:
Nice to see that you are back.
How long have you come with the tuning and determining what different settings do and how they effect other settings etc.?
I'm trying to get the throttle and brake to react like I want, but I'm not there yet.

I put a reply in your thread, basically havent done anything since melting the battery at the drags. I have ambitious plans to make a dash that talks to the controller for basic SOC and possibly on the fly adjustment of Ride modes and settings. I have changed my left switch block to a factory triumph tiger switch set which has an enter button and up down arrows. Should allow me to change riding modes and maybe some parameters all with my left hand while riding. Im also tempted to have the throttle go to this board to allow completely custom throttle curves, or velocity based throttle responce, or speed based max power. I find the elec motor a bit brutal on slippery rocks. Im temped to try an E-clutch setup too. For the cost of an extra ADC input, seems worth a shot. Yet to see if I can pull it off though. Hopefully this weekend Ill get some time to sit with the bike and decode the 485 completely.

Doug
 
Back
Top