LDU parameter tuning

Topics concerning the Tesla front and rear drive unit drop-in board
jon volk
Posts: 572
Joined: Wed Apr 10, 2019 7:47 pm
Location: Connecticut
Been thanked: 2 times

Re: LDU parameter tuning

Post by jon volk »

Dont feel too bad. I broke both of my axles at the same time.

Dramatic changes in load will cause an overcurrent so that’s plausible. Wheel hop will trigger it for me.

Formerly 92 E30 BMW Cabrio with Tesla power
P.S.Mangelsdorf
Posts: 753
Joined: Tue Sep 17, 2019 8:33 pm
Location: Raleigh, NC, USA
Has thanked: 89 times
Been thanked: 92 times

Re: LDU parameter tuning

Post by P.S.Mangelsdorf »

jon volk wrote: Sun Feb 20, 2022 8:51 pm Dont feel too bad. I broke both of my axles at the same time.
I knew the solution I had wouldn't last, I'm just a bit irritated that I didn't investigate the clunking earlier, and ended up stranded 3 miles from home. Thankfully, I've got some great family friends that came to give me tow home.
If at first you don't succeed, buy a bigger hammer.

1940 Chevrolet w/ Tesla LDU - "Shocking Chevy" - Completed 2023 Hot Rod Drag Week
Golf_mkIII_E
Posts: 32
Joined: Tue Apr 20, 2021 8:55 pm

Re: LDU parameter tuning

Post by Golf_mkIII_E »

Tesla driveshaft adapters are not the only possible answer to this problem.

I had been stuck on the driveshaft chapter in my build for quite some time. In the end, I welded together my cut VW Golf driveshaft head to head to a cut Tesla drive shaft. It was only for testing the fitment of it.

After that I brought the driveshaft to a famous buggy builder in the area, who made 2 completely new driveshafts out of one piece. They are holding up really good. :D

Image

Image
P.S.Mangelsdorf
Posts: 753
Joined: Tue Sep 17, 2019 8:33 pm
Location: Raleigh, NC, USA
Has thanked: 89 times
Been thanked: 92 times

Re: LDU parameter tuning

Post by P.S.Mangelsdorf »

Golf_mkIII_E wrote: Sun Feb 20, 2022 9:42 pm Tesla driveshaft adapters are not the only possible answer to this problem.
True, however I was already planning on going this route because the suspension I'm using (2005 Pontiac GTO/Holden Commodore) uses the flange type axles, and mixing axle types was sort of sketchy. This makes it super easy to get custom axles. "Please make me a pair of axles this long, with a GTO flange on one end and a Porsche flange on the other"
If at first you don't succeed, buy a bigger hammer.

1940 Chevrolet w/ Tesla LDU - "Shocking Chevy" - Completed 2023 Hot Rod Drag Week
User avatar
Bratitude
Posts: 783
Joined: Thu Jan 02, 2020 7:35 pm
Location: Canada
Has thanked: 57 times
Been thanked: 168 times
Contact:

Re: LDU parameter tuning

Post by Bratitude »

Yeah getting custom axels shafts made is really quite an easy job. Indexing head and a fly cutter on a mill. nothing to fancy.
definitely worth building in universal ways. 108mm cv joints (930) are used every where. But cutting to match the Tesla inner race splines is easy. Cheaper if you already have them!
https://bratindustries.net/ leaf motor couplers, adapter plates, custom drive train components
P.S.Mangelsdorf
Posts: 753
Joined: Tue Sep 17, 2019 8:33 pm
Location: Raleigh, NC, USA
Has thanked: 89 times
Been thanked: 92 times

Re: LDU parameter tuning

Post by P.S.Mangelsdorf »

Bringing this thread back to life now that my car is on the road and I solved a few issues and still have some gremlins.

Solved: slamming an overcurrent trip early in an acceleration run
Solution: idcmax somehow ended up too low. Not sure if adjusting scaling or what, but I had it at 700. Now that I've raised it to 5000 as in the recommended base parameters, I no longer have the issue. Fixing the axle helped as well.

Solved: strange surging and limited speed after overcurrent trip in traffic
Solution: my brake signal to the LDU came from the third brake light wire. My hazards flashed the third brake light. Thus when I threw on the hazards because I'd just stalled in the middle of the road, the LDU saw brake-throttle-brake-throttle-brake-throttle-brake-.... etc.

Unsolved: low speed oscillation around 5-10mph.
This one I need help with. I'm using r5.14 with the fweakstrt even higher than some example parameters, and I'm still getting a violent oscillation when traveling at 5-10 mph and low-moderate throttle input. The only way to get it to stop is to give it a lot more throttle (always works) or back off and roll in carefully (sometimes works). Its even worse on hills, which sucks because my little cul-de-sac is a hill, and I like to drive like a sane human in my neighborhood (read: slow and smooth). I will note that 5.14 HAS solved over current trips when starting off slowly, which is fantastic.

I've attached my parameters below. Any ideas beyond give it even more fweakstrt?
Attachments
params.json
(1.45 KiB) Downloaded 57 times
If at first you don't succeed, buy a bigger hammer.

1940 Chevrolet w/ Tesla LDU - "Shocking Chevy" - Completed 2023 Hot Rod Drag Week
jon volk
Posts: 572
Joined: Wed Apr 10, 2019 7:47 pm
Location: Connecticut
Been thanked: 2 times

Re: LDU parameter tuning

Post by jon volk »

Try raising fweak to 300 and also increasing boost to 1600. Fweak first, then the boost. Do you have one or two Volt packs? I can’t recall.
Formerly 92 E30 BMW Cabrio with Tesla power
P.S.Mangelsdorf
Posts: 753
Joined: Tue Sep 17, 2019 8:33 pm
Location: Raleigh, NC, USA
Has thanked: 89 times
Been thanked: 92 times

Re: LDU parameter tuning

Post by P.S.Mangelsdorf »

I've got two packs. Worth noting that the most recent testing was between 75% and 50% SOC

I'll give that a shot. The weather here is miserable so its going to take a few days.
If at first you don't succeed, buy a bigger hammer.

1940 Chevrolet w/ Tesla LDU - "Shocking Chevy" - Completed 2023 Hot Rod Drag Week
P.S.Mangelsdorf
Posts: 753
Joined: Tue Sep 17, 2019 8:33 pm
Location: Raleigh, NC, USA
Has thanked: 89 times
Been thanked: 92 times

Re: LDU parameter tuning

Post by P.S.Mangelsdorf »

jon volk wrote: Tue Jan 31, 2023 1:18 am Try raising fweak to 300 and also increasing boost to 1600. Fweak first, then the boost. Do you have one or two Volt packs? I can’t recall.
I was able to test this today. Definite improvement in the oscillation issue, and some drop in performance (via butt dyno). There is still a bit of oscillation at low speed, low throttle, so I tried raising the fweakstrt by 10, and it came back nearly as bad as it was. I wonder if some of the issue is the delta between fweak and fweakstrt? I plan to experiment with that the next time I've got clear weather and time to work on the car. (I'm supposed to have clear weather during the week, but also have lots of real world work to do.)
If at first you don't succeed, buy a bigger hammer.

1940 Chevrolet w/ Tesla LDU - "Shocking Chevy" - Completed 2023 Hot Rod Drag Week
User avatar
Boxster EV
Posts: 434
Joined: Fri Jul 26, 2019 9:32 pm
Location: UK
Has thanked: 35 times
Been thanked: 26 times

Re: LDU parameter tuning

Post by Boxster EV »

I had irritating albeit infrequent on/off throttle OC issue for months running version 5.14r. Since using 5.20r not had one issue. I’ve also been able to run a higher boost value (2000) without any OC events.

Thanks to Johannes for the ongoing development.
Porsche 986 powered by a Tesla large drive unit. Backwards. Build documented here and Instagram @tesla_porsche here.
P.S.Mangelsdorf
Posts: 753
Joined: Tue Sep 17, 2019 8:33 pm
Location: Raleigh, NC, USA
Has thanked: 89 times
Been thanked: 92 times

Re: LDU parameter tuning

Post by P.S.Mangelsdorf »

I'll give 5.20 a shot and report back.
If at first you don't succeed, buy a bigger hammer.

1940 Chevrolet w/ Tesla LDU - "Shocking Chevy" - Completed 2023 Hot Rod Drag Week
User avatar
Boxster EV
Posts: 434
Joined: Fri Jul 26, 2019 9:32 pm
Location: UK
Has thanked: 35 times
Been thanked: 26 times

Re: LDU parameter tuning

Post by Boxster EV »

P.S.Mangelsdorf wrote: Mon Feb 06, 2023 12:50 pm I'll give 5.20 a shot and report back.
Some of your parameters deviate quite significantly from the standard Damien default. Might be worth doing the firmware upgrade and trying a fresh known good parameter set. You’ll probably need to drop boost and Fslipmax a little here:
params.json
(1.49 KiB) Downloaded 66 times
Porsche 986 powered by a Tesla large drive unit. Backwards. Build documented here and Instagram @tesla_porsche here.
P.S.Mangelsdorf
Posts: 753
Joined: Tue Sep 17, 2019 8:33 pm
Location: Raleigh, NC, USA
Has thanked: 89 times
Been thanked: 92 times

Re: LDU parameter tuning

Post by P.S.Mangelsdorf »

Boxster EV wrote: Mon Feb 06, 2023 9:08 pm Some of your parameters deviate quite significantly from the standard Damien default.
Mine are based on Jon's set and tuning guide.
If at first you don't succeed, buy a bigger hammer.

1940 Chevrolet w/ Tesla LDU - "Shocking Chevy" - Completed 2023 Hot Rod Drag Week
User avatar
Boxster EV
Posts: 434
Joined: Fri Jul 26, 2019 9:32 pm
Location: UK
Has thanked: 35 times
Been thanked: 26 times

Re: LDU parameter tuning

Post by Boxster EV »

P.S.Mangelsdorf wrote: Tue Feb 07, 2023 1:08 pm Mine are based on Jon's set and tuning guide.
Worth a try. :) Some of your parameters seem even more aggressive (throtramp 75 and regenramp 100...) than the "aggressive" ones listed here https://openinverter.org/parameters/view.html?id=16
Porsche 986 powered by a Tesla large drive unit. Backwards. Build documented here and Instagram @tesla_porsche here.
P.S.Mangelsdorf
Posts: 753
Joined: Tue Sep 17, 2019 8:33 pm
Location: Raleigh, NC, USA
Has thanked: 89 times
Been thanked: 92 times

Re: LDU parameter tuning

Post by P.S.Mangelsdorf »

Alright so I'm making some progress. Thanks for pushing me on comparing the parameters, somehow all of my gain settings were off base. The regen settings felt fine before, now they're too aggressive but fixing the gains made a huge difference in general drivability.

I'm now working on tuning out a WOT overcurrent. I've got it down now to where if I'm already moving, 98% of the time if I stand on it I do not get an overcurrent. However, if I do a full launch (brake and throttle on the floor, dump brake), as soon as full power has ramped in, I'm getting an overcurrent.

I think that I've gotten quite close to fixing it, and certainly have learned a lot about the respective parameters. I don't have a full parameter dump to share as I've been tuning with my phone, but I do have a question about fweak and fslipmax.

Based on this from Johannes, I understand both effect overall current:
johu wrote: Sun Dec 15, 2019 11:07 am Lowering fweak OR raising slip means increasing motor current. Thats why you have to compensate on with the other.
My question is, is it better to optimize one or the other? Should I be targeting a balance? What does maximizing each do?

(Goal for the car is a street-able drag car, so 1/4-mile time/acceleration is a big priority.)
If at first you don't succeed, buy a bigger hammer.

1940 Chevrolet w/ Tesla LDU - "Shocking Chevy" - Completed 2023 Hot Rod Drag Week
User avatar
Boxster EV
Posts: 434
Joined: Fri Jul 26, 2019 9:32 pm
Location: UK
Has thanked: 35 times
Been thanked: 26 times

Re: LDU parameter tuning

Post by Boxster EV »

I'm not really a WOT type of person, although I do like a spirited drive. However, from my practical experience on the symptoms you describe, I don't see any benefit messing with Fweak if you are running 5.20R as it has Jon's compensation mod embedded in the code (higher Fweak in lower rev range and lower Fweak after a certain RPM).

Are you running 5.20R? Earlier versions of the firmware caused me WOT OC at top end (4,97R) and on/off regen (5.14R). 5.20R seems extremely reliable and robust.

In the past, when I've experienced OC events under the conditions you describe, it's usually because Boost or Fslipmax have been too aggressive. However we're not comparing apples with apples as my car is probably lighter and I'm only running one Chevy Volt pack.
Porsche 986 powered by a Tesla large drive unit. Backwards. Build documented here and Instagram @tesla_porsche here.
P.S.Mangelsdorf
Posts: 753
Joined: Tue Sep 17, 2019 8:33 pm
Location: Raleigh, NC, USA
Has thanked: 89 times
Been thanked: 92 times

Re: LDU parameter tuning

Post by P.S.Mangelsdorf »

I'm running 5.20R now. Backing fslipmax down to 3.00 has seemingly worked well. I've been able to lower fweak to 270 without a trip at 85-90% SOC. Boost is at 1600. I haven't had the time to push further, I only have about 30 minutes to play with the car in the evening. Hoping to pull a full parameter set this evening.
If at first you don't succeed, buy a bigger hammer.

1940 Chevrolet w/ Tesla LDU - "Shocking Chevy" - Completed 2023 Hot Rod Drag Week
gerhard25
Posts: 1
Joined: Sun Feb 26, 2023 6:43 pm

Re: LDU parameter tuning

Post by gerhard25 »

Hello, can someone help me with the control of the motor?
the podnom is going totally crazy. nominal voltage to 0, then it works to some extent but slowly. but he doesn't do the recuperation every time.
It would be cool if someone who knows what I'm talking about can turn to me for a moment.
User avatar
Boxster EV
Posts: 434
Joined: Fri Jul 26, 2019 9:32 pm
Location: UK
Has thanked: 35 times
Been thanked: 26 times

Re: LDU parameter tuning

Post by Boxster EV »

gerhard25 wrote: Mon Feb 27, 2023 3:16 pm Hello, can someone help me with the control of the motor?
the podnom is going totally crazy. nominal voltage to 0, then it works to some extent but slowly. but he doesn't do the recuperation every time.
It would be cool if someone who knows what I'm talking about can turn to me for a moment.
Suggest you create a specific thread with the issue providing as much detail as possible. As a minimum provide details of your parameter file, low and high voltage power source. Pictures of your set-up would also help.
Porsche 986 powered by a Tesla large drive unit. Backwards. Build documented here and Instagram @tesla_porsche here.
User avatar
gesso
Posts: 8
Joined: Sat Feb 05, 2022 5:12 pm
Has thanked: 1 time
Been thanked: 4 times

Re: LDU parameter tuning

Post by gesso »

I updated from 5.13(what came on my board when it arrived) to 5.20R today. Oh man, what a world of difference.

On 5.13 I had low speed, low throttle surging/bucking and would hit OC at large throttle low speed every time (even with boost at 1300). The car was mostly driveable with careful throttle management, just not really able to push it from a stop.

On 5.20R, the car feels lovely. No low speed issues any more and no OC when launching. I upped the boost to 1600 and then 1800 and still no OC. 1800 gets a little wheel spin with a full throttle launch, very excited to try some more aggressive settings.
My only issue so far is that subjectively it feels like it's not launching quite as hard (even with full throttle), but the kw readings say otherwise, so maybe it's just the poorly calibrated butt meter.
P.S.Mangelsdorf
Posts: 753
Joined: Tue Sep 17, 2019 8:33 pm
Location: Raleigh, NC, USA
Has thanked: 89 times
Been thanked: 92 times

Re: LDU parameter tuning

Post by P.S.Mangelsdorf »

gesso wrote: Sat Mar 04, 2023 7:46 am My only issue so far is that subjectively it feels like it's not launching quite as hard (even with full throttle), but the kw readings say otherwise, so maybe it's just the poorly calibrated butt meter.
I've experienced the same, and I'm trying to find a 5.20 tune that brings back the violence without an oc.
If at first you don't succeed, buy a bigger hammer.

1940 Chevrolet w/ Tesla LDU - "Shocking Chevy" - Completed 2023 Hot Rod Drag Week
Dennis
Posts: 8
Joined: Mon Oct 23, 2023 5:56 pm

Re: LDU parameter tuning

Post by Dennis »

Hello everyone, I’m new here.
I have installed an Openinverter board to a Tesla LDU for my EV conversion. I was immediately able to control the unit. I have two questions that I can't resolve. I have read a lot on the forum and I have tried to understand Johannes video The Great Parameters Tutorial well but my first language is not English and can’t understand some things. If any of you can help me with two inquiries I will appreciate.

The first inquiry is the following:

The accelerator pedal turned the LDU into spin when it had been pressed almost halfway. After I watched the tutorial video, I set the throttle to min 0 and max 4095. Now it travels about 1/4 of the way when I press it. How can I adjust it so that just with a little foot pressure the motor responds to spin?

The second question is as follows:

At first, I had to swap A and B (pins 10 and 16) to get the spin on the LDU. It works perfectly but without inverting them, there is no spin. So, in Forward (pin 7) the motor rotates as if in reverse and in reverse (pin 5) the motor rotates as if in forward. Why does that happen? The Openinverter board comes set for a LDU in reverse rotation?
User avatar
johu
Site Admin
Posts: 5682
Joined: Thu Nov 08, 2018 10:52 pm
Location: Kassel/Germany
Has thanked: 153 times
Been thanked: 960 times
Contact:

Re: LDU parameter tuning

Post by johu »

Dennis wrote: Mon Oct 23, 2023 6:06 pm The accelerator pedal turned the LDU into spin when it had been pressed almost halfway. After I watched the tutorial video, I set the throttle to min 0 and max 4095. Now it travels about 1/4 of the way when I press it. How can I adjust it so that just with a little foot pressure the motor responds to spin?
Known behaviour and will be worked on. First 30% travel are reserved for regen by default. You can change this by reducing the "regentravel" parameter
Dennis wrote: Mon Oct 23, 2023 6:06 pm At first, I had to swap A and B (pins 10 and 16) to get the spin on the LDU. It works perfectly but without inverting them, there is no spin. So, in Forward (pin 7) the motor rotates as if in reverse and in reverse (pin 5) the motor rotates as if in forward. Why does that happen? The Openinverter board comes set for a LDU in reverse rotation?
Also a known issue discovered some months ago. OI firmware runs the field the wrong way around. That's why the encoder needs to be swapped. And afterwards forward becomes reverse...
Support R/D and forum on Patreon: https://patreon.com/openinverter - Subscribe on odysee: https://odysee.com/@openinverter:9
Dennis
Posts: 8
Joined: Mon Oct 23, 2023 5:56 pm

Re: LDU parameter tuning

Post by Dennis »

Thanks you Johu 👍
Dennis
Posts: 8
Joined: Mon Oct 23, 2023 5:56 pm

Re: LDU parameter tuning

Post by Dennis »

Hi Johu. I have other inquiries.
1. In this great video ( ) at time 59:12 to 1:16:23 Damien explained the principle of a cooker resistance in PARALLEL with the HV fuse to prevent damage by “back emf” or “spring tearing” analogy in the inverter transistor when performing tests. And, he mentioned that the main contactor is controlled by the inverter using pin 3 and pin 6 for precharge relay coil to prevent damage in driving mode. Isn't it supposed that to have the protection explained with the cooker resistance, during driving, the precharge resistor must be in parallel with the contactor and the HV fuse that are in series? Openinverter diagrams illustrate the precharge resistor in parallel only with the main contactor.
2. The other inquiry is about the throttle. Which is pod2 wiper signal pin when use second throttle redundancy?
Post Reply