Low Voltage cutoff from BMS or Controller

[Afmu]

Hi fellas,

Almost one year ago I had an injury in my knee so I decided to "electrify" my bike to help me riding uphills.
I made my own battery with Samsung cells (2800mA), a chinese BMS and a HL case. I also bought a ebike kit from Amazon, Prystel 26T, for rear wheel.
From the beginning, the thing hadn't work as intended. Obviously it works but when the voltage came below 37 V the LCD turned off and I had to reset the battery (disconnect and connect again) in order to continue riding.
I use a manual throttle, I haven't installed the PAS because I only use the bike to go to work (and rehab).
Recently I bought another controller because I thought the problem was it. Specific, I thought is was 48V and that would explain the cutoffs al 37V. I also bought another LCD, with much more information in screen.

The thing is that still the same. I've tested with different P1 values but I can't guess the correct value since there's no information in the wheel. When I press the throttle and the voltage comes below 32V-31V, the battery BMS or the controller itself cuts the current.

I'm adding a video I recorded today with the bike behaviour.
https://drive.google.com/file/d/1H97BEfe_LBmwbURnHbRIv-O4fzK3AFUn/view?usp=sharing

I don't know that to do, I've being trying things for almost one year, but none worked. The only solution was to recharge the battery everyday (to 41.5 to 42 V).

Can you help me?

 

[wheeliepete]

It sounds like you are suffering voltage sag from your battery. What exact model cells did you use, how many and how did you join them together? Because you are only using a throttle, you are asking the battery to supply max. current most of the time. If you can use PAS you can limit the current draw on the battery by using a low setting and the voltage sag will be less.

 

[Afmu]

Thanks for your answer, Wheeliepete.
The cells were bought in gearbest and they are Samsung ICR18650 - 28A 3.7V 18650 2800mAh. I've welded them in 10S, so they're 40 cells. I did it following youtube/internet tutorials from ebikeschool.com among others.
I don't have a spot-welding machine so I use my standard welder and niquel plates. The BMS should works with over 40A currents, as its specification.
The components are the ones in the attachment.

 

[KirstinS]

Wheelie Pete is right, sounds like bad sag

But this could be wrong cell type (can't provide enough amps- the c rating is too low)

or a bad cell or cell string is pulling the pack down

Or the motor and BMS aremismatched.

All of this may be tested but before then would you please post the exact cell type

Also the BMS specification

And finally the motor specification

Edit- crossed posts !!


Sent from my iPhone using Tapatalk

 

[wheeliepete]

I'm afraid you have made a bad choice of cells with 28A. They are high capacity, but very low discharge, They cannot give out enough amps to power your motor. They are quite likely to be at the end of their useful life.

 

[Afmu]

Thanks for the prompt answer KirstinS.
More specifications:
The cells are this ones, I don't know what else info can I add:

The BMS spec:

The controller spec:

Regards.

 

[wheeliepete]

As you are running a 20amp controller, you are maxing out the discharge ability of you cells everytime you give your bike full throttle which is stressing them out (man!) You could reduce the amp draw in the settings on your LCD as low as they will go, prob.about 15 Amps and this will help a little with the voltage sag, but better cells is the only real cure I'm sorry to say.

 

[Afmu]

Hehe
Thanks for the information wheeliepete, it's what happen when you are a noob (me!). No problem, I can use the cells for other porpouses like external batteries for sports camera, smartphones and so.
Two questions I may ask:
1. Which setting should I set to reduce the amp draw meanwhile I get a new battery. Maybe, should I set P1 and C12?
2. Which batteries do you recommend to do a new pack?

Greetings and thanks again for your help!

 

[wheeliepete]

Hey Dani, don't worry, we all started knowing nothing and we have to make mistakes to learn from them
I've only recently finished my first build with one of the KT controllers that you have, so I'm only just learning about the settings myself, but there are plenty on here that are more famiilar with them, hopefully someone will reply.
As far as cells goes, look on Nkon, lots of us buy from them, genuine cells and cheap postage to the EU. You want a cell with 10 amp min. discharge, so if you build a 4P10S battery pack, as you have and ask it to give 20 amps, it is only suppling half the amps it is able to and thus stressing the cells less, which means minimal voltage sag and longer life for your battery.
I'm interested to know when you said in your earlier post that you don't have a spot welder and are using an "standard one", do you mean a MIG or TIG?

 

[Afmu]

Hehehe None of them, I'm using a gas powered tin solderin iron.

 

[wheeliepete]

TFFT, had to ask, wondered why there weren't any youtube videos for that method Found this vid for quick run though controller settings, C5 looks like the one to try.

 

[Afmu]

The problem with the tin solderin iron is overheating the cells, so they can lose capacity. I took the risk.
Thanks for the vid, I will try it.

Greetings.

 

[Nealh]

Your components are totally mis-matched .
You are asking 4P x 5.6a max rated cells to have a theoretical 30 - 45a drawn out of them by the controller, it's never going to be good. The 28A is 4.3v li- hv type so you will be losing capacity any way as they are being under charged.
The batteries are being highly stressed and shagged as the controllers ceiling is way above the cells capability. The issue is simple sag is to severe for these batteries to cope.

For your controller you need a minimum 15a rated cell to give 60a rating in 4p configuration.
30Q, HG2, VTC5A , HE2 or 25R cells.

 

[Nealh]

Battery spec for 28A/4.3v cell.
The 5a sag test is terrible imagine them at the 7.5 - 10a your controller is asking of them.
https://lygte-info.dk/review/batteries2012/Samsung ICR18650-28A 2800mAh (Purple) UK.html

A batteries discharge in amps not ah needs to exceed the controllers max amp output.
All is dependant on the cell specs in your case your battery max discharge is 22.4a for a short time the max rated controller for this would be 15/17a.

 

[Afmu]

Thanks for your explanations, Nealh, very usefull.
I will check some new cells with better discharge.

Cheers.

 

[wheeliepete]

Dani, you could also try refitting the original controller you got with the Prystel kit, it looks like a standard KT 15/17 amp one and play with the C5 settings. Might give your battery an easier time.

 

[Afmu]

OK, that sounds good to me Easy and cheaper, for the moment.

Thanks, wheeliepete.

 

[Afmu]

Following your advice, I set the C5 to a ÷2 current value, so theorically now it shouldn't be higher than 10 amps with the current controller.
The results are the expected, the wattage never comes over 300W and the voltaje falls from 42v (idle) to 37v (running) when I pull the throttle.
I've being researching the cells you recomend and I think I will try the Sony VTC6, but some time in the future.
Thanks for the advices.

 

[Afmu]

Hello again, fellas.
I haha a doubt on what cells to pick.
In nkon I can choose the Sony vtc5a or the Sony vtc6. Thay have the same price. What should I buy with my config?

Greetings.

 

[Nealh]

VTC6 for the increased capacity and a solid 15a continuous rating more then enough for 4P config.

Review of the two cell from 2 different independent testers.

https://www.e-cigarette-forum.com/threads/bench-retest-results-sony-vtc6-15a-3000mah-18650…a-fantastic-15a-20a-battery.846895/

https://www.e-cigarette-forum.com/threads/sony-vtc5a-2500mah-18650-bench-test-results-a-fantastic-25a-battery.746719/

https://lygte-info.dk/review/batteries2012/Sony US18650VTC6 3000mAh (Green) UK.html

https://lygte-info.dk/review/batteries2012/Sony US18650VTC5A 2600mAh (Green) UK.html