TSDZ2 48V 500W No power to motor

[chocomel2]

Hello, I was hoping the fine people on this forum could help me diagnose my motor problems. I have contacted the seller but they haven't gotten back to me yet after sending my story and photos.

About a month ago I bought and installed a TSDZ2 on my bike. I flashed it with OSF after a few test runs and it worked great up till a few days ago, when it shut down mid journey. I stopped and tried to turn on the display again to see if it was my battery shutting down (I had ridden the bike to below cutoff voltage on a previous occasion but I charged it the night before so figured it might be a bad battery or w/e). The display turned back on without me having to reset the battery bms however, it did immediately give me a torque sensing error on the display for a second.

On closer inspection the casing was scorching hot, probably hot enough to burn my fingers if pressed long enough. So I figured it was overheated. So I turned off the assist and rode back the rest of the way, with the display turned on but without assist. When I got back home the casing was cool to the touch again. So I tried turning back on the assist to see if everything was fine. But alas everything was not fine and I haven't been able to get it running again ever since, no assist no throttle.

I did have one ride in heavy wind where it shut off mid ride, but the casing didnt feel hot. It did give me a torque sensing error. After resetting it gave me very little assist, no matter what assist level chosen for the rest of the ride to work that day. It also seemed less powerful the ride back home later that night. After that I had the suspicion it felt less powerful. But I figured it might just be me getting too used to the assist. After scouring the interwebs searching for similar problems from other people I think I the motor might have demagnetized from overheating.

TLDR my TSDZ2 motor has overheated and is perhaps demagnetized to the point it doesn't run at all anymore. It seems it doesn't engage/power the motor at all so more things or maybe different things are broken. The display works fine and the speed sensor works aswel. The plastic on top of the motor at the wire connections looks a bit molten and the black silicon seems to have softened again. Theres more (molten?) gunk that maybe came from the controller on/in the casing (see pics).

I generally never used any assist level above a 100% assist (level 2 in my case) for more than a couple minutes at time. Hardly ever used the throttle and when I did it was also never for prolonged periods. I generally rode at speeds around 32-35km/h on average. I kept the motor settings pretty standard I think (see screenshot of the configurator settings), so hopefully it wasn't my own fault it died . Any thoughts/comments would be greatly appreciated.

Some pics :

 

[Elinx]

Seeing the brown laminated core and melted plastic, your motor looked overheated and demagnetized.
Maybe your controller is also (partly) ruined.
You said you had battery cutoff before. I think this could be the reason for overheating.

On the picture I see you have installed v0.20.1C for stock displays.
mbrusa had replaced this release with v.0.20.1C1 with safety updates for possible problems with power shutdown.

Problably one of the fets on the controller has send a uncontrolled current through the motor.
Normally the controller burnt first, but in your case it looks it didn't.
I advice you to have a close look at the controller too.

 

[chocomel2]

Thanks for the reply Elinx. That does look pretty similar to my issue. Though it is implied his still ran somewhat, from what I can tell from his post. On the software version, that looks like the issue is about version v0.20.1C.3 specifically, hopefully not v0.20.1C. At the time of my installation that version was disabled and the version I am using was the recommendation. I really hope that wasn't the issue.

 

[Elinx]

... the issue is about version v0.20.1C.3 specifically, hopefully not v0.20.1C. At the time of my installation that version was disabled and the version I am using was the recommendation. I really hope that wasn't the issue.

On v0.20.1C.3, three people had a burned controller, done by a FET.
After this mbrusa take this version immediately down.
mbrusa was convinced that v 0.20.1C was safe, although there where allready some complains about overheating for espacially 48V motors.
But the cause was in the first place directed to the "always enabled" Field Weakening.

When mbrusa found the error in v0.20.1C.3, he found out that this risk also existed with v 0.20.1C, maybe not in such a destructive way, but it was there too. which he subsequently repaired and replaced it for v 0.20.1C.1
Nobody has mentioned burning problems with v 0.20.1C before, so maybe you are the first, just because the battery cutoff.

It is possible to replace the bare motor only.
Because of the overheating I advice also to improve the heat dissipation with heat conductive material between motor and case and if possible a temperature sensor on the motor.
For the sensor you can use a integrated sensor, which cut down the power if motor heated up or a separated simple digital temperature meter.

 

[raylo32]

Should we not be running with field weakening enabled?

But the cause was in the first place directed to the "always enabled" Field Weakening.

 

[Elinx]

Should we not be running with field weakening enabled?
.....

Maybe. It depends of the user if he want to keep some assistence with a higher cadence, which could happen with short MTB hills. Imho on flat roads you don't need it.

Field Weakening is not free. It is less efficient, uses more power of battery and can (if used too long) heatup the motor.
That is why users, in the first place, directed the heatproblem to, with lower cadences "too active", Field Weakening.
With v.0.20.1C and v0.20.1C.1 builds there is no choice to disable Field Weakening, if you don't need it and users did ask how to disable it.

The extreme heating problem was only noticed with 48V motor and v.0.20.1C of mbrusa, not with the base build of mspider65.
mbrusa did found the reason and has updated to v0.20.1C.1 (same options, but with safety updates).
The extreme heating problem was not caused by "always enabled" Field Weakening. So there was no need to change this.

With the latest builds of mbrusa (and mspider65) you can choose to enable/disable Field Weakening if the user wants.

See also the changelog of mbrusa builds.

 

[raylo32]

I know FW is not free but I like to ide higher cadences on fast road rides so I have been running with it enabled. But I don't want to keep doing that if if it will risk the motor.

Should we not be running with field weakening enabled?
.....

Maybe. It depends of the user if he want to keep some assistence with a higher cadence, which could happen with short MTB hills. Imho on flat roads you don't need it.

Field Weakening is not free. It is less efficient, uses more power of battery and can (if used too long) heatup the motor.
That is why users, in the first place, directed the heatproblem to, with lower cadences "too active", Field Weakening.
With v.0.20.1C and v0.20.1C.1 builds there is no choice to disable Field Weakening, if you don't need it and users did ask how to disable it.

The heating problem was only noticed with 48V motor and v.0.20.1C of mbrusa, not with the base build of mspider65.
mbrusa did found the reason and has updated to v0.20.1C.1 (same options, but with safety updates).
That heating problem was not caused by "always enabled" Field Weakening. So there was no need to change this.

With the latest builds of mbrusa (and mspider65) you can choose to enable/disable Field Weakening if the user wants.

See also the changelog

 

[Elinx]

I know FW is not free but I like to ide higher cadences on fast road rides so I have been running with it enabled. But I don't want to keep doing that if if it will risk the motor...

You don't risk the motor if you use Field Weakening with care.
If I remember right it comes only in with higher cadences of 90.
If you have a temperature sensor/meter and cooling additions there is for sure no risk.
With a build in temperature-sensor the power goes down if temperature rises too high.

imho the best way to get higher cadence is overvoltage, than you don't need Field Weakening.
This is possible with another battery or same battery with (1500W) boostconverter.

 

[raylo32]

If the risk is strictly heat then I have no problems since I have effective cooling mods. However if the risk were some electronic glitch and failure due to FW that would be another matter. I have done 6 rides or so with it and experienced no problems.

I know FW is not free but I like to ide higher cadences on fast road rides so I have been running with it enabled. But I don't want to keep doing that if if it will risk the motor...

You don't risk the motor if you use Field Weakening with care.
If I remember right it comes only in with higher cadences of 90.
If you have a temperature sensor/meter and cooling additions there is for sure no risk.
With a build in temperature-sensor the power goes down if temperature rises too high.

imho the best way to get higher cadence is overvoltage, than you don't need Field Weakening.
This is possible with another battery or same battery with (1500W) boostconverter.

 

[Elinx]

If the risk is strictly heat then I have no problems since I have effective cooling mods. ...

Only heat and battery drain, because you use about 25% more power for the same movement.
No risk for electronics on the controller, if you limit your max power and current settings at the safe side.

Some field weakening was already there from 2018 (experimental high cadence), but more effective from may 2020.
The heating with v.0.20.1C and burning controller with v.0.20.1C3, for stock display only (java configurator), was a problem of the code, which is repaired now.
For other display versions (hex code) there was never such a problem.

 

[raylo32]

I haven't limited power. Set to 800W max... but I don't often actually go there. But I will have to remember to disable field weakening when I do a slow ride with my friends on pedal bikes. If I can get anything close to 25% more efficiency that will support longer rides.

If the risk is strictly heat then I have no problems since I have effective cooling mods. ...

Only heat and battery drain, because you use about 25% more power for the same movement.
No risk for electronics on the controller, if you limit your max power and current settings on the safe side.
Some field weakening was already there from 2018 (experimental high cadence), but more effective from may 2020.

 

[Elinx]

.. Set to 800W max... but I don't often actually go there. ... If I can get anything close to 25% more efficiency that will support longer rides.

If you never go there it is a sort of limitation too If you need permanent 800W your motor will heat up for sure.
You use only 25% more power, with high cadence.
If you use lower cadence (<90) there is no field weakening anymore and no extra power used.
That is why field weakening was, at the first place, always enabled.

But becasue of the calculation of it, field weakening can, if the battery is more discharged, come up with lower cadence too. In that case the drain will be faster too.
If you never pedal with high cadence this is a waist of energy, that is why disabling is usefull.

See here too

 

[raylo32]

On my ride today I disabled field weakening and I didn't like it. It's fine just cruising around but I run out of assist in my highest gear when I really get spinning on a flat stretch of road. I have the speed limit set at 35mph and with FW I can pedal up pretty close to that. But with FW disabled the motor assist starts dwindling long before my legs are done and short of 30 mph. It is almost like riding a PAS bike without shifting assist levels where you get to a point where the only way to go faster is more human power... but at ~30 mph on an upright bike this human can't make much more progress. So... I believe I will enable FW when I feel like doing a faster ride and disabling it otherwise.


If you never go there it is a sort of limitation too If you need permanent 800W your motor will heat up for sure.
You use only 25% more power, with high cadence.
If you use lower cadence (<90) there is no field weakening anymore and no extra power used.
That is why field weakening was, at the first place, always enabled.

But becasue of the calculation of it, field weakening can, if the battery is more discharged, come up with lower cadence too. In that case the drain will be faster too.
If you never pedal with high cadence this is a waist of energy, that is why disabling is usefull.

See here too
[/quote]

 

[Elinx]

On my ride today I disabled field weakening and I didn't like it.....

If you are used to pedaling with high cadence, I can imagine you are not happy with the cadence limit of the tsdz2
In the past Jbalat found too the cadence of tsdz2 is too low, with stock FW a little above 70-80
He did a lot of experiments and by changing some values he could get 90-100, but this "experimental high cadence" option was not very battery friendly, because it did work with lower cadences too.
The Field weakening as it is now inside the code is a lot better configured and I think if you want/need higher cadence when needed gives a satisfactory result.
But if the Field weakening is activated you get a higher cadence which uses more power, more current and more heat.

As said, a higher voltage on the motor is another option to get a higher cadence too.
The advantage of this is with the same power as with lower cadence, a lower current and less heat.
imho if you need higher cadence most of the time, this is a better way

 

[raylo32]

So is the field weakening only engaged when a rider pedals to higher cadences so that below some threshold it does not create the issues you listed? IOW, only really activated when called on by the pedaling cadence? If so that seems pretty ideal to me.

But I did have a look at the DC/DC boost transformers. Interesting idea... but another box to mount and I don't really understand how to use them. Does one just connect them between the battery and the motor controller? Any changes needed to the OSF installation?


The Field weakening as it is now inside the code is a lot better configured and I think if you want/need higher cadence when needed gives a satisfactory result.
But if the Field weakening is activated you get a higher cadence which uses more power, more current and more heat.

As said, a higher voltage on the motor is another option to get a higher cadence too.
The advantage of this is with the same power as with lower cadence, a lower current and less heat.
imho if you need higher cadence most of the time, this is a better way
[/quote]

 

[raylo32]

So is the field weakening only engaged when a rider pedals to higher cadences so that below some threshold it does not create the issues you listed? IOW, only really activated when called on by the pedaling cadence? If so that seems pretty ideal to me.

But I did have a look at the DC/DC boost transformers. Interesting idea... but another box to mount and I don't really understand how to use them. Does one just connect them between the battery and the motor controller? Any changes needed to the OSF installation? I will assume that higher voltage to some point won't damage the motor, but what voltage is that?


The Field weakening as it is now inside the code is a lot better configured and I think if you want/need higher cadence when needed gives a satisfactory result.
But if the Field weakening is activated you get a higher cadence which uses more power, more current and more heat.

As said, a higher voltage on the motor is another option to get a higher cadence too.
The advantage of this is with the same power as with lower cadence, a lower current and less heat.
imho if you need higher cadence most of the time, this is a better way

[/quote]

 

[Elinx]

..field weakening only engaged when a rider pedals to higher cadences .....

Yes and No, it depends of the battery Voltage, when Fieldweakening comes in. Mostly with higher cadence above 90, but with lower battery it could be with 70 too. Please read this how FW works for tsdz2.

... DC/DC boost transformers. .........

I think this subject is again a new one.
We are in the troubleshoot topic and are already far of the original question and answer of chocomel2.

For the best you can visit the battery topic for that.
You find a lot more information about it for example here and there
Beemac has used a DC booster with Tsdz2, see his short review

 

[raylo32]

No worries, if I decide to go down the rabbit hole for voltage boosters I'll start a new topic in "batteries" later... but after reading some of your links I don't think I'll mess with it. The bike works OK as is using field weakening... or without it for more relaxed rides.

..field weakening only engaged when a rider pedals to higher cadences .....

Yes and No, it depends of the battery Voltage, when Fieldweakening comes in. Mostly with higher cadence above 90, but with lower battery it could be with 70 too. Please read this how FW works for tsdz2.

... DC/DC boost transformers. .........

I think this subject is again a new one.
We are in the troubleshoot topic and are already far of the original question and answer of chocomel2.

For the best you can visit the battery topic for that.
You find a lot more information about it for example here and there
Beemac has used a DC booster with Tsdz2, see his short review

 

[chocomel2]

After taking apart the unit further and inspecting the controller more in detail. It looks like the controller also has several burn marks. So it looks like I'm going to have to replace both the controller and inner motor.

Buying a replacement controller seems pretty straightforward, though the motor not so much. It is sold by vendors just plainly as 48v ( or 36v) but then there's also vendors that have different rated versions? Is there actually a difference or are they all just the same 48v (or 36v) motor?

 

[Elinx]

.... I'm going to have to replace both the controller and inner motor.

Buying a replacement controller seems pretty straightforward, though the motor not so much. ....

The controller is for all tsdz2 exactly the same for electronical parts, only stock firmware differs.
A small difference is the connector 8pin vs 6pin (throttle vs no-trotthle)
If you have the throttle version, you can use the wires for a temperature sensor with OSF.
As a reminder, from 2020 there is a new type that can't be flashed with OSF (till now only sold by Enerprof).
If you buy at an international seller you get the flashable version.

There are only two types of bare motors, 36V-4000RPM and 48V-4000RPM

So if you flash OSF, you can go for the cheapest options you can find