Is my controller 'bad'? Severe loss of power at low SoC?

[bob2.0]

I've decided that this drop in power at low SoC is 'severe'.

I mean, it has always done this and I first assumed it was the controller or BMS protecting the battery.

You guys said it was just a drop in voltage.

But, I get a much bigger drop in power than the drop in voltage, they do not go down together.

Fully charged at 84V I get strong, 'nice' performance.

At around 74V it drops off so much that it's barely adequate.

At below 72V it's dangerous, and only a mad person would ride it. Which, I do all the time!

I do a regular 75 mile trip on it, taking back roads because in that lower half of the battery, it literally cannot pull off on a hill. There is a point in my trip where I have to stop at lights pointing up hill. If I mis-time the lights, I literally have to peddle and push the bike forward to get it going, it can't do it on its own.

One day, maybe someone in a car will not be as kind to me as they usually are, and will just drive over me. I have to be so careful.

So, here is my question; how low SoC do you guys take your bikes, and do you find the loss of power is 'severe'?

I bought a spare controller, if you recall, and I am wondering if there is any sort of fault with the controllers that might cause this, and I should swap it out? I have resisted doing it because I did not think it would help, but, I don't know.

[bob2.0]

I did a long ride out this weekend and it seems to be worse.

it might be my perception, but I was at around 40%, still 5~6 bars, and it would barely accelerate up a gentle gradient.

It is basically in 'tortoise' mode from 6~7 bars, as if the battery is totally flat, but it is still >40%.

So, I think it is the controller, and that it is an error mistake measuring the battery voltage.

'chatgpt' suggested some component failure in a resistor divider circuit. Perhaps a signal smoothing capacitor that has cracked and leaking, pulling the voltage down.

This makes sense.

I need to get that computer reading feature going. I have a hard wire lead julet connector but I can't find the thread explaining the steps connecting it up.

I have a replacement controller with bluetooth cable connection. But I worry replacing it all in one go, making too many changes.

Does this failure mode make sense? If the controller software reports a lower voltage than actual, then I think that could explain everything.

[jeff-jordan]

--- sorry, double post ---

[jeff-jordan]

Please connect* your controller via bluetooth & fardriver app.

In the "Graph" screen, you can monitor the voltage (the one that the controller "thinks" the battery has).
Compare this value with the voltage you measure with a precise multimeter at the charging port.
The values should be roughly identical (+/- 1%).

Maybe you need to adjust the "-Protect" parameters: LowVolProtect (LowVolRestore will be adjusted automatically to LowVolProtect + 2V).
The LowVol... values define when the bike gets into turtle mode.
You can even switch off the turtle mode when you set the LowVolProtect value to 60V or below, but I won't recommend that.

I recommend to drive with a smartphone holder and an (old) android phone running the fardriver app, connected to the controller.
Then you can monitor what happens if you loose power.

But maybe the problem is the controller temperature?
I had this last year, because the cooling of my ND72530 was not adequate (before I installed a fan above my ND72530).

_____________________________


*) To figure out the correct pinout of the small julet connector (which comes with random colors!) and make a connection to the bluetooth dongle with the 4x 2.8mm automotive connector, just figure out between which pins of the julet connector you can measure about +5V, using your multimeter.
If you then connect the +5V to the purple wire of the dongle and GND to the black wire of the dongle, the red LED at the dongle should start to blink slowly (approx. once every second).
You then should be able to "see" the dongle with a bluetooth monitor (as described here).
To communicate now with the controller/dongle, you can connect the two remaining pins to the two remaining wires of the dongle (red and blue).
There are two possibilities to do so.
You can try out both without damaging anything.
At one of the possibilities you should be able to connect to the controller via fardriver app and bob's your uncle .

[bob2.0]

Thank you nicely for your reply. I appreciate the direction towards bluetooth, I was hoping to avoid rewiring things. I have a julet to usb connection.

I should have done this and sorted it before and last year. Funny, that the machine rolling at 45mph max speed was not a problem for me, I can at least say it makes the range better!

But this lack of acceleration is a big problem now.

It remains poor on acceleration even after the controller has cooled down. So, maybe not that. But perhaps that has 'also' gone bad now, too. I often seem to suffer dual, simultaneous failures that make problems a total mystery. It is meant to be rare. I am now convinced The Gods of machines play with me, just for their amusement. They like seeing me shouting at inanimate machines that refuse to be fixed.

[bob2.0]

The USB lead I had seemed to work fine with my PC so I just plugged it in to the motorcycle and it all worked fine.

I cannot see any issues, apart from the motor temp is reading zero, which I can't see why that would cause a restriction as the battery voltage drops.

Battery voltage is slightly under what I think it is, but by fractions of a volt.

Controller temp seems fine.

These settings, this current battery charge, 'couldn't pull skin of custard'. Well, it can, just.

So, I am none the wiser.

At least now I have all the settings, so I can go ahead and swap with the new controller I have. Then, we'll see if it is some strange MOSFET failure, and not opening the gate fully.

If you can take anything out of these screen shots, I can give you any from the other menus.

'Mystery'.

[bob2.0]

Progress!

Of sorts.

OK, thanks for that input regarding dropping the protection voltage.

That is 'what' the answer is, but I do not know 'why' it is the answer.

I set it to 50V/52V (+2V) as a test, and there is no power restriction. Battery supply was at 71V, maybe 20% (?) and I zapped it up the road, it pulled normally and reached 100kph easily. This would normally only be able to do 60~70kph with it set at 63V/65V.

After ~5km, it had dropped a volt or two, ~69V, showing up as just 12% SoC on the charger readout. So, obviously, one needs to be careful at lower SoC as the battery drops faster with power as the cells get to a knee curve.

I would, please, like to ask your input on what I should do right now, today. This is because while it is at 12% I could do some tests to see where the protection voltage should be set so that I get the active restriction happening at this level, but not earlier.

Otherwise, I would charge it up tonight and it will stay at this (dangerous?) protection level of 50V.

I can only assume there is a gradient of power reduction built into the hardware, because the power available dies away gradually. I see no settings in software?

But overall, I assume the battery actually takes care of itself. Of course, we do not want to routinely get the BMS to the point it shuts off, that does not sound right. But I have done it once already (when I discovered it cuts at 3 bars) and it did not seem to cause any lasting harm.

So, I presume, just relying on the battery BMS to cut the power is 'OK', perhaps not ideal.

Therefore, the issue in this thread is 'resolved', it is;-
A) The low voltage protection seems to be very/too aggressive, could the sensors on the bus voltage be faulty?
B) it can be worked around by artificially setting the protection voltage lower. But I am not sure that is the best answer.

My questions for you guys;-
1) Should I do some testing now (today), the sooner the better while the battery is low to find a voltage setting that gives me 'fair' power restrictions at ~10%/68V?
2) Or leave it set at 50V, see how it goes, but what do I need to keep in mind, and could I damage anything if I just left it, and let the battery cut out if it needed to?
3) Maybe I fit a voltage gauge to show me a continuous instantaneous readout, and I just go carefully as the battery voltage drops, keeping an eye on it like that and avoid <3.3V/cell?

Thanks for the input.

[bob2.0]

Does anyone know what those coefficients are at the far right?

"Cap0Coeff" and "Cap100Coeff", "LimitStartSOC"?

What do yours say on your controller?

I've no issue with power restrictions at the right voltage, I feel I just need to change the derating slope.

According to chatgpt (I'll take its output as guidance). it seems those are indeed the coefficients that control the ramp down.

I got it to arrive at this prediction. I can go test it, now.

[dominik]

The capcoeff are for the battery SOH Meter and the upper one can also stop regen before your BMS needs to cut.

The ramp is made with low vol protect and low vol restore and low vol way.

There is no current reduction until low vol + low vol way is reached.
I still have 1400A phase current and 400A battery current, when the voltage is above 65V. My low vol protect is set to 63V and low vol way ist 2V. I have 21S Li-NMC battery.
So if your battery is good you can set your low vol protect to maybe 3V/cell or even a little lower. But 50V is far to low. Your BMS will probably cut out during driving when your battery gets under 60V.

You can check the values out when you set your low vol protect to 2V under you actual battery voltage, than the controller limits current as soon as the voltage drops.
So you will see reduced power directly from the start.

[dominik]

At around 74V it drops off so much that it's barely adequate.

At below 72V it's dangerous, and only a mad person would ride it.

If your battery drops 7V during acceleration then it is either too cold or you have got a problem with your battery or the wiring or the circuit braker.