Aussie EV Autocross Special

[galderdi]

I now have the results from Sunday. I managed 2nd in class and 19th outright. Not great on face value.

But those places are a direct result of my failure to finish my second run. For that run I inherited a time of double the fastest run (2 x 72.29 = 144.58). If I had managed to equal my previous run (88.34) I would have been 1st in class and 10th outright. Usually the second run is a couple of seconds quicker than the first, but I won’t speculate too much. I am not dismissing the real results, but I am using this adjusted result as a gauge to arrive at a target for future events. If I can achieve a moderate increase the power, make a few other minor adjustments, and maintain some reliability I am fairly confident I can push for 1st in class and an outright position well inside the top ten (maybe 6 or 7). This is supported by my results in the shorter test with a 5th, 8th, 8th and 7th placings.

So while I am not overly happy with the actual result I am quite happy when I read between the lines. And by next year it should be a whole new ball game.

 

[galderdi]

Here is a still pic from Sunday's khanacross

 

[kiwiev]

Well done greg

If you could increase your acceleration a bit more by the looks of you video im sure you could move up the ranks. Which you are rolling the car looks tight on the track mate.

Cheers Conrad

 

[galderdi]

Thanks. I agree it felt solid and controlable, all be it a little frantic at times. But I think softening the front a touch will help. I think through the tighter corners I can pick up some time just by getting used to the car. Aside from threading the needle through the concrete gate I never felt at risk. I did get a surprise from some torque steer, you can see it at the beginning of the first run. But it was managable. I attribute most of these results to my initial approach, which was to mostly copy the geometry from the Pulsar. I figured the Pulsar felt pretty good so if I could leverage off their design I couldn't go too wrong. The ackerman angle adjustment really seems to have worked well. When cornering at normal speeds the tyres no longer complain at all. When I am pushing the limits they complain but thats more a case of being at the limits of the car's cornering ability, not a problem with the geometry.

Under braking the rear brakes seem to have a little too much bias. But I don't currently have anyway to compensate for it. But I don't think it is severe enough to worry about anyway.

I have been comparing the videos from my old Pulsar to the Batt mobile and the acceleration from 50-90 is where I am hurting. Going by my G force meter the old car would accelerate about twice as fast from 50-90. I have some plans for some tweaks and testing that might see some improvements. I did some experimenting on various runs with different gear change strategies. At the end of the day I think the lack of torque forced me to run through the full set of gears as quickly as possible. Skipping gears resulted in slower times. It was also necessary to change gears a lot during the run, before and after each corner. With some extra torque I might be able to save some time by reducing gear changes.

Assuming I can increase the current and therefore the torque I beleive I will need a way to control the launch as wheel spin will become an issue.
So I am planning to implement another dial so I can choose a lower power for the initial launch and then revert to 100% for the rest of the run. I initially thought about connecting it to first gear but I don't think that would be ideal. I think the simplest idea is a button on the centre console. Press the button during the launch then release it for the rest of the run. I just want to get past that first 1-2 seconds where too much power just equates to tyre smoke.

As you can tell I got some great data from the day. All in all I have a positive feel about the whole thing.

 

[kiwiev]

Cool

Its the complete opposite in the Sonic off the line its slow but 30 to 100 its bloody quick on my hill climb I compared my video to a mate in a ICE car and I really made up time in the corners due to no gear changes and I could focus completely on the track.
I bet those concrete barrier looking intimidating when you are up close and personal

I wonder if you can do series parallel switching like the Zombie 222 does

I don't know much about it but its like a horsepower torque switch with you battery pack maybe hillhater can explain it.

Cheers Kiwi

 

[galderdi]

The only concrete that worried me was that gate. I think I could have been doing 30% faster through there but if I tried it I think my testicles would be found hiding up somewhere between my lungs.

Maybe the torque switch idea would work but to be honest I am not sure I will need it now. My "loose connection" was so loose I beleive it was to blame for the poor performance ever since switching to Lithium Ion. The system was operating through the smallest contact patch. I can't imagine the resistance that introduced and I am quite surprised it actually ran at all! The performance is now fairly similar (maybe identical) to the old lead acid batteries. And even when pushed hard I don't feel any heat from the batteries.

I am now 50/50 regarding the posibility that the controller's current setting is now the bottle neck (not the batteries). In other words it is entirely possible the batteries are happy to produce more current than the current I am seeing but they are hitting the barrier set by the controller. So I might get some good results if I simply turn it up. So if I can work out a way to test it I will prove it one way or the other. The difficulty is finding an environment were I can put the system under enough load to draw maximum current. It might need to wait until the next event. But thats Aug 7th. I really don't want to wait that long. I might see if I can secure some dyno time as an alternative. Or maybe I will tow my BT50 with the BT50 handbrake slightly on to produce some drag. But the strain on the drivetrain scares me.

 

[galderdi]

Is there any scenario where a Lithium Ion cell could be approaching its current limits without any detectable heat?

 

[oatnet]

And even when pushed hard I don't feel any heat from theI am now 50/50 regarding the posibility that the controller's current setting is now the bottle neck (not the batteries). In other words it is entirely possible the batteries are happy to produce more current than the current I am seeing but they are hitting the barrier set by the controller. So I might get some good results if I simply turn it up.

That is exactly right. Amps=torque and your motor is is not getting enough amps to begin to show you what it can do. Put the 600a it needs into the motor and it will be a whole different animal. I know that the torque curve on the FB1 motor sheet is almost linear, but to my butt-dyno it feels like a geometric progression - maybe it is the torque-to-weight ratio I am feeling.

I repeat my advice to you from April 3rd:

"I'm telling you man, dial up the amps and it will be better than you expected. I'm pretty sure you said you had cooling on the batteries, so the batts can handle it, the controller can handle it, the motor can handle it; assuming your drivetrain can handle it, it's all good. Since I don't think your motor is cooled, I'd check it after a few runs at each new power level, to get a sense of the thermal load. "

Is there any scenario where a Lithium Ion cell could be approaching its current limits without any detectable heat?

You are nowhere near the limits of these cells, and severely limiting your fun by being too tender with them. As you dial the amps up, if you are concerned look for resistance, expressed through voltage sag and heat.

-JD

 

[galderdi]

And even when pushed hard I don't feel any heat from theI am now 50/50 regarding the posibility that the controller's current setting is now the bottle neck (not the batteries). In other words it is entirely possible the batteries are happy to produce more current than the current I am seeing but they are hitting the barrier set by the controller. So I might get some good results if I simply turn it up.

That is exactly right. Amps=torque and your motor is is not getting enough amps to begin to show you what it can do. Put the 600a it needs into the motor and it will be a whole different animal. I know that the torque curve on the FB1 motor sheet is almost linear, but to my butt-dyno it feels like a geometric progression - maybe it is the torque-to-weight ratio I am feeling.

I repeat my advice to you from April 3rd:

"I'm telling you man, dial up the amps and it will be better than you expected. I'm pretty sure you said you had cooling on the batteries, so the batts can handle it, the controller can handle it, the motor can handle it; assuming your drivetrain can handle it, it's all good. Since I don't think your motor is cooled, I'd check it after a few runs at each new power level, to get a sense of the thermal load. "

Is there any scenario where a Lithium Ion cell could be approaching its current limits without any detectable heat?

You are nowhere near the limits of these cells, and severely limiting your fun by being too tender with them. As you dial the amps up, if you are concerned look for resistance, expressed through voltage sag and heat.

-JD

Thanks for the advice. I admit I didn't take the advice last time because I assumed my drop in performance was due to the batteries. It turns out I had introduced a loose connection (adding resistance and reducing current). So I will certainly be cranking it up as you suggested. I certainly do have cooling on the batteries. I am really getting excited now. Can't wait for the next event. The results should be chalk and cheese.

 

[oatnet]

I wonder if you can do series parallel switching like the Zombie 222 does I don't know much about it but its like a horsepower torque switch with you battery pack maybe hillhater can explain it.

You need 2 motors for series/parallel switching and a controller (like the Z2K) which can interrupt power and manage the relays.

At low rpm, the motors run in series so that the both get the the same amps for maximum torque. However, this splits the voltage giving them a low max RPM, limiting top end. Switching them to parallel gives them the same voltage, so they can give usable torque at a higher RPM. This of course splits the amps between them, which sounds like a bad thing, but at higher RPMs back-emf is limiting the amps you can push into the motor anyhow.

 

[oatnet]

Thanks for the advice. I admit I didn't take the advice last time because I assumed my drop in performance was due to the batteries. It turns out I had introduced a loose connection (adding resistance and reducing current). So I will certainly be cranking it up as you suggested. I certainly do have cooling on the batteries. I am really getting excited now. Can't wait for the next event. The results should be chalk and cheese.

Awesome. You are gonna have a lot of fun!

-JD

 

[galderdi]

I wonder if you can do series parallel switching like the Zombie 222 does I don't know much about it but its like a horsepower torque switch with you battery pack maybe hillhater can explain it.

You need 2 motors for series/parallel switching and a controller (like the Z2K) which can interrupt power and manage the relays.

At low rpm, the motors run in series so that the both get the the same amps for maximum torque. However, this splits the voltage giving them a low max RPM, limiting top end. Switching them to parallel gives them the same voltage, so they can give usable torque at a higher RPM. This of course splits the amps between them, which sounds like a bad thing, but at higher RPMs back-emf is limiting the amps you can push into the motor anyhow.

It sounds amazing and the results they are achieving certainly confirm the theory. But I think, or at least hope, it would be unnecessary for my purpose. I just need to improve my acceleration from 50-105kmph anything over that would be wasted in my sport. And indeed the extra components and weight would likely have a negative impact in handling. But never say never. If I explore all my existing options and I'm still not satisfied I will continue to explore other options.

Its pretty obvious I have been a little cautious. But I am fairly new to all this. I only started this build in March 2015. I would rather make sure to do incremental adjustments and check with you guys along the way rather than big BANG!!!! changes.

 

[oatnet]

I just need to improve my acceleration from 50-105kmph anything over that would be wasted in my sport.

That may simply be a matter of gearing, along with an increase in motor amps.

Amps are directly correlated to torque. As RPMs increase, backemf limits the amps we can push into the motor. I know of 2 ways to address it.

1) If you increase the motor-side volts, a given amount of back-EMF happens at a higher rpm. This gives you a given amps/torque at a higher RPM, extending your top end. The torque curve stays exactly the same, in just gets pushed up the RPM band. This works great on brushless motors, but unfortunately the brush/stator interface of a brushed motor limits the voltage we can run before experiencing flashover. IIRC your controller is voltage constrained anyhow, so this o-toon is not available to you.

---> 2) Instead, we have to work on keeping RPMs as low as possible to keep in the torque band we need. So in an electric car the best way to accelerate to pass someone is to UPshift into a higher gear, so the motor is running at a lower RPM where it can get more amps=torque. This is counter to what we have learned from gas cars, where we DOWNshift to get higher RPM to access more torque.

That said, you need the amps=torque to back it up. When I was testing with low amps, the UPshift was not effective - I think I wasn't running enough amps to overcome drag and inertia in a dramatic fashion. However, UPshift was effective after I dialed the amps up.

Torque comes at cost, whether you get it from higher voltage or lower gearing. At low RPMs, the motor delivers more torque but it is also less efficient, which is expressed as more current wasted as heat, which results in lower wh/m. Hopefully your FB1's thermistor is connected to your controller so it can automatically back off current if the motor gets too hot. If not, as you crank up amps and gear down, check temps by hand, often, until you get a sense of how hard you can push it before it becomes a problem.

-JD

 

[galderdi]

Thanks again. Yes I do find it works most effectively when I quickly work through the gears. It feels like 1st is at about the right compromise. 2nd and 3rd are a struggle at the moment but with the extra amps I am expecting them to come good. Eta on the answer is 6 hours.

 

[kiwiev]

Thanks again. Yes I do find it works most effectively when I quickly work through the gears. It feels like 1st is at about the right compromise. 2nd and 3rd are a struggle at the moment but with the extra amps I am expecting them to come good. Eta on the answer is 6 hours.

Cant wait to hear the results mate exciting stuff, thanks for the great advuse JD

Cheers Kiwi

 

[galderdi]

Ok so the amp setting on the controller seems to have about 90 degrees to go before its on full. Bad news is I can't seem to safely put it under enough load to max the amps. I tried towing a full size wagon with the handbrake on. I will turn it up some but I won't be able to check the results until august. It didn't help that the batteries were fairly low on charge after last weekend.

 

[oatnet]

Ok so the amp setting on the controller seems to have about 90 degrees to go before its on full. Bad news is I can't seem to safely put it under enough load to max the amps. I tried towing a full size wagon with the handbrake on. I will turn it up some but I won't be able to check the results until august. It didn't help that the batteries were fairly low on charge after last weekend.

BackEMF is probably limiting amps quicker than you can see. If you are using a standard CycleAnalyst, you can check max amps and volt minimum to see what amps you pulled and how much your pack sagged when you pulled them. If you are just running an amp meter, you are lacking the instrumentation you need to make an informed evaluation, shooting in the dark.

How hot did the motor feel? Putting additional load - weight and handbrake - has its hazards. Even at low amps, you can burn your motor in a few seconds by holding it at stall - at 0rpm most of the current gets transferred to heat. One should never try to "hill hold" an EV with the throttle, for this very reason.

A better approach in a lightweight vehicle like ours would be to launch in 3rd/4th gear. This allows you to spend more time in low RPMs that launch you powerfully, but with so little mass/inertia to overcome, RPMs come up quickly enough for backEMF to limit amps before you cook the motor. I'm sure our gear ratios are different, but 1st gear is useless to me, and 2nd gear is best for just spinning tires at the tap of a throttle.

I can program specific amp limits with a Zilla, but it sounds like you have to twist a potentiometer to set your amps? How frustratingly analog! This will be imprecise, because amps are so dependent on RPM, and you don't want risk burning your motor by hovering at low RPM, so you will never get a clean reading. I guess I'd attempt tuning by doing a bunch of 100 meter runs, turn up the amps each time until 3rd-gear-launches pull nicely, but checking the motor temp with your hand after each run to make sure you haven't gone too far, in case 3rd gear is too tall.

I think your motor, battery, and chassis are up to snuff, but a CycleAnalyst, a pack sized charger (or cheap PSU), and Soliton jr or Zilla controller will significantly enhance your experience and abilities. Budget constraints may not allow that, so you may have to limit your expectations to what you can test and achieve. TradeOffs, but I highly regret not upgrading my VW bus conversion to a Zilla before I disassembled/sold it, I just didn't understand that the controller was what was holding me back.

 

[oatnet]

I just did a search on this thread, and it looks like you said your controller is a Curtis 1231c. If that is only a 500a controller, 500a is nothing to the motor and batteries, turn it ALL the way up. I'd start being more conservative about my driving style at 600a-700a. At 1,000a, I'd want extreme cooling, Helwig brushes, and keep my runs under 10 seconds and let the motor cool down in between.

-JD

 

[galderdi]

Thanks. Yeah I am really looking forward to turning the amps up. I have spotted a piece of road to use for testing. It is public road so its not techincally legal. But I plan to go out there at 6am and its an unused dead end anyway. So it should be fine. I will be going straight from there to a Japanese car show. So after that I should have some answers sunday night.

 

[galderdi]

[youtube]cXuownQlQi0[/youtube]

Some more footage from the last event

 

[galderdi]

[youtube]HXkbNuCl7-4[/youtube]

 

[galderdi]

[youtube]RMlKAiNBcqk[/youtube]

 

[oatnet]

Nice vids Gald, thanks for posting. What gear are you launching in, and do you have the amps turned all the way up on the Curtis? Did you feel a difference when you cranked up the Curtis? What is the highest amp drain you ave been able to observe?

Although we are running the same motor and similar weights/voltages, it doesn't Look like it is leaping off the line like the way mine feels like it is. It looks more the way mine felt with the Kelly controller - which is a cheap knockoff of a Curtis. If you are wanting more, I'd think that is your bottleneck, hard data from a CycleAnalyst/Analogger could prove that out.

-JD

 

[galderdi]

I had not turned up the controller when these videos were taken. At that time the controller's amp settign was about 90 degrees from the maximum. I have only seen 240amps (on the battery side of the controller.
I currently launch in 1st. But if I get significant increases in my torque when I tweak it I will try taking off in 2nd. At the moment running quickly through all the gears seems to get the best results.

I have turned it up since then but have not had the conditions to test it. Hopefully in the next week or two I will get that chance.
Unfortunately I don't have any device to provide data logging. That might be on the cards once I take care of all the other priorities.

 

[kiwiev]

Awesome video mate especially looking down on the course great driving the car looks quick and solid.

I really think you should take it to a Dyno and twick controller on there then you have hard data, you can monitor current with a clamp on meter.

Cheers Kiwi