Tongsheng TSDZ2B assist too high

[HackyHackerson]

I have a 48v TSDZ2B that I've put on a Ozark Trail G1 Explorer. It came with EKD01 display. Lately the bike has been adding 300-500 watts assist in power level 1.

I discussed the problem with an AI, and it mentioned that there is a crank arm zero that might be an issue. The crank arms have been off, and I honestly can't remember if I put them back on the same way. I never really paid attention to the assistance level before the arms came off, but it seems like it was lower. My range is suffering pretty badly.

I've also updated to the latest firmware for the EKD01 display. Both of those things happened about the same time, so I'm unsure whether I have hot rodded firmware now, or if the crank arms are in the wrong orientation.

Taking the crank arms off and testing is complicated by the fact that I don't have a puller at the moment. I can, and have, loosened the crank arm screws and ridden the things loose, but doing so multiple times for the sake of testing is something I would like to avoid if possible.

TL;DR - Is there a zero orientation for crank arms on a TSDZ2B? If so how would I determine what it is visually? Or, is it likely that the new firmware is hot rodded?

Thanks in advance for any help.

 

[gromike]

Get a crank arm puller and move the arms 90 degrees and see. Otherwise, I don't know. Never ever had an issue with too much assistance in a TSDZ2b. Can you set the power of the assist level on the display setup, or via firmware program?
Don't ever ride with loosened crank arms.

 

[HackyHackerson]

Get a crank arm puller and move the arms 90 degrees and see. Otherwise, I don't know. Never ever had an issue with too much assistance in a TSDZ2b. Can you set the power of the assist level on the display setup, or via firmware program?
Don't ever ride with loosened crank arms.

Yeah, I can set assistance level. It's just that it doesn't seem to make a difference. I'm hitting 300 plus watts with most of my strokes.

There aren't too many hidden menu options with my display that I can change.

Does rider weight and pulling a trailer perhaps impact output? I'm 200 plus pounds and I'm pulling a trailer quite frequently.

 

[docw009]

I believe the AI is talking about the torque sensor calibration that the TSDZ2B does on powerup. All you do is leave the pedals alone for 10 seconds, although they could be at 3-o-clock and 9-o-clock for equal weighting, The motor then knows what the T/S output should be with no pedal force and then calculates the assist from there.

I've found though that disturbing the calibration by having wight on the pedals results in less assist, not more assist. My display is the VLDC5 which doesn't show power, but based on how my other bikes feel, I doubt my TSDZ2 can put out much more than 400 watts. That's being optimistic too.

 

[HackyHackerson]

I believe the AI is talking about the torque sensor calibration that the TSDZ2B does on powerup. All you do is leave the pedals alone for 10 seconds, although they could be at 3-o-clock and 9-o-clock for equal weighting, The motor then knows what the T/S output should be with no pedal force and then calculates the assist from there.

I've found though that disturbing the calibration by having wight on the pedals results in less assist, not more assist. My display is the VLDC5 which doesn't show power, but based on how my other bikes feel, I doubt my TSDZ2 can put out much more than 400 watts. That's being optimistic too.

It seemed to say that the 'timing' or crank location could be the problem. I know that when an AI has incomplete information it can hallucinate, though. I thought I'd see if I could get information to back it up here.

If I had a puller I'd simply test it.

There are some other tricks I've read about. Some say to actually push the bike backwards while powering it on. I intend to test some of the things I've read.

I've gotten it to output less than 100 watts per stroke in testing. But that was only achieved in first gear with a high, at least for me, cadence.

I also hoped that the displayed wattage was combined output, but apparently that's not the case.

 

[docw009]

There are some other tricks I've read about. Some say to actually push the bike backwards while powering it on. I intend to test some of the things I've read.

If you want inconsistent assist, resulting in lower power, that will do it for sure.

As for pulling the pedals, think about it. This is a power up calibration done in the first 10 seconds after every power up. You can't be pulling the pedal arms for this every time. Sometimes, I do start off too soon, and can tell the torque is off since I'm having to press harder. I'll often try find a section where I can coast, take the feet off the pedals, and do a power reset, But if the bike hits a bump, it's won't calibrate, so it's best to stop.

 

[HackyHackerson]

If you want inconsistent assist, resulting in lower power, that will do it for sure.

As for pulling the pedals, think about it. This is a power up calibration done in the first 10 seconds after every power up. You can't be pulling the pedal arms for this every time. Sometimes, I do start off too soon, and can tell the torque is off since I'm having to press harder. I'll often try find a section where I can coast, take the feet off the pedals, and do a power reset, But if the bike hits a bump, it's won't calibrate, so it's best to stop.

I might not explaining myself very well. The AI seems to imply that the crank arms need to be on in a certain orientation in order to work with the torque sensor properly. If they're out of position the sensor thinks that the rider is applying considerable pressure to the pedals. Essentially one side of the square on the square taper is the correct place to mount the crank.

It makes sense in theory, but I was wondering if someone who works on the thing knows for certain.

I would have to order a puller in order to test crank orientation.

It might be that the display isn't reading accurately. The watts jumped up north of 800 when I set it at level 5.

At 48v the TSDZ2B is 750;watts peak, correct?

 

[E-HP]

It might be that the display isn't reading accurately. The watts jumped up north of 800 when I set it at level 5.

The EKD01 never displays anything accurately. Voltage and power are totally unreliable. I was thinking about adding a separate meter, but I just ignore the display instead.
Luckily the numbers are too small to see anyway.

 

[docw009]

I might not explaining myself very well. The AI seems to imply that the crank arms need to be on in a certain orientation in order to work with the torque sensor properly. If they're out of position the sensor thinks that the rider is applying considerable pressure to the pedals. Essentially one side of the square on the square taper is the correct place to mount the crank.

OK. This is your concern, and chatGP must have found this post and ignored the answer.

TSDZ2 crank position relative to the torque sensor

Hi, anyone having any idea about what's the best position of the crank(s) in relation with the torque sensor? If you look at the pictures you see that there are #2 main positions of the square interface on the axle in relation with the torque sensor (depending on how you rotate the axle inside...

endless-sphere.com

A crank puller is inexpensive, around $10 on ebay or amazon. If you need to prove it to yourself, buy a puller and rotate your cranks 90 degrees.

 

[HackyHackerson]

OK. This is your concern, and chatGP must have found this post and ignored the answer.

TSDZ2 crank position relative to the torque sensor

Hi, anyone having any idea about what's the best position of the crank(s) in relation with the torque sensor? If you look at the pictures you see that there are #2 main positions of the square interface on the axle in relation with the torque sensor (depending on how you rotate the axle inside...

endless-sphere.com

A crank puller is inexpensive, around $10 on ebay or amazon. If you need to prove it to yourself, buy a puller and rotate your cranks 90 degrees.

While I appreciate any sort of feedback I can get at the moment I'm at a campsite and not in a position to get anything delivered to me easily.

Anything that I get has to be delivered via UPS or FedEx. If it comes via USPS it will be sent back. I've already had a package with brake parts from Amazon sent back playing delivery roulette.

So, hopefully that post is accurate. I assume that if it is it must have been the firmware update that juiced the motor.

I suppose I'll look into alternative firmwares. Mine says N15, and I think that there's an N17.

Beyond that I might have to look into the OSF for the motor.

Thank you.