TSDZ2 cuts out on load

[Clumsy6608]

This is a new problem, I've been riding this setup for about three years with no problems (TSDZ2, 36V, OS firmware). When the battery is about 60% I start to get complete power loss if I apply more than minimal load. I don't have this problem with a freshly charged battery. The screen goes off and the only way to turn it back on is to power cycle the battery with its built-in switch. Does this mean the BMS is cutting power?

What can I do to isolate the problem? I have a multimeter (but not for DC current), plenty of tools, happy to open up the battery, can run the bike with resistance on a trainer in the workshop using pedals or throttle.

 

[docw009]

It's probably an unbalanced battery, The BMS is cutting power.

You have to open the battery, identify its construction, and find either the individual ends of the cells group or locate the BMS harness that connects to them. You also have to understand how a battery is built,

Personally, I don't think unbalanced batteries are repairable,

 

[Clumsy6608]

It's probably an unbalanced battery, The BMS is cutting power.

You have to open the battery, identify its construction, and find either the individual ends of the cells group or locate the BMS harness that connects to them. You also have to understand how a battery is built,

Personally, I don't think unbalanced batteries are repairable,

Ah, so even though the voltage of the whole battery is only 0.5v lower, an individual cell can bring the whole lot down. Seems a shame not to replace that cell, but indeed I do not have the equipment or experience to spot weld a replacement. I'd like to confirm it though, so I think I'll try to access the BMS and measure the voltages. I've attached a picture of hte battery, any comment, advice specific to this unit?

 

[docw009]

What is an unbalanced battery? You may know that a 36V ebike batteru consists of ten series cell groups. Each group operates between 4.2V and 3.0V. In a healthy battey, the group voltages charge and discharge tiogether,



If one or more cell groups happen to lose capcity or develop a fault like high discharge, they can drop to something like this, THe battery will read 36V at the outside, but one cell friuo is close to being shut off by the BMS. Unless the BMS has balance capability, it can't be equalized. EVen if the BMS could equalize, it might take several days to a week to do so,

In my opinion, bad cell groups indicate the cells are near the end of life, or there is a quality issue. In either case, if the battery were repaired, it would happen again wioth another yroup,

 

[docw009]

I've attached a picture of hte battery, any comment, advice specific to this unit?

View attachment 360496

Bottle batteries usually have a round circuit board containing the BMS. Somethimes it's all encased in shrinkwrap, Other times only the cells are wrapped.

If you can identify the BMS harness, and unplug it, it;s usually an 11 pin connector that eaches all 11 connections to the cells,. Yiu can probe that, Sometines, on these round cards, they split the connector into two or more smaller ones, Most voltmeter probes won;t fit inside the BMS connectors, so you have to improvise, Take care not to short circuit any of the balance pins,

 

[Clumsy6608]

It was split, took me a while to map the adjacent connections. You were correct:

The differences are: 4.1, 4.1, 3.6, 4.1, 4.1, 4.02, 4.08, 4.08

Given it's a few years old and the cells and chinesium, time for a new battery. Think I'll get a rear rack one, I hate the look of frame-hung ones. Pswpower have one they say uses Samsung cells