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

[bob2.0]

Thanks for the links, I'll consider them.

But I wouldn't expect to see connectors if it was waterproof. I have doubts on the weather robustness of those.

I don't like the idea of live 72V lines being added. If they can power a readout then they can't be fed with protection resistors (else it'd change the reading).

There are too many lines at the front of the bike that remain live already (IMHO). I did not want to add yet more.

What I wanted to do was to make a plug that fits between the harness plugs of the alarm, under the seat. It'd just be a plug-in piece (I have that connector type) and I would string a 300k Ohm and 15k Ohm [the only standard resistors to make the right ratio] across the 72V to make a 1:20 (per cell) voltage, to be read by a 12V powered off the number plate light. Those 3 wires can then be safely routed to a 12V powered readout at the front.

That was my plan. I just needed a waterproof 12V powered 3 wire.

Can I please ask this to check something; in this system, is the 'negative' voltage of the 12V and 72V common? Are those already connected, or are they floating wrt each other?

I'm trying to minimise any 'further' 72V risks.

[jeff-jordan]

...
Can I please ask this to check something; in this system, is the 'negative' voltage of the 12V and 72V common? Are those already connected, or are they floating wrt each other?
...

The 12 V and 72 V circuits share the same common negative terminal—due to the DC/DC converter (with a common negative).
Please note, however: The motorcycle chassis is not connected to the common negative terminal (and should not be).

[bob2.0]

I might have an answer to this, some more analysis pending.

I have installed a direct voltage readout (that is mis-calibrated, but still I can see from it what is going on at battery voltage) and also added and testing out an extra 15Ah or so of extra batteries.

The behaviour I've described here remains the same, even though I have set the low voltage protect to 55V.

I have added battery, I have got the readout and things still the same, I can see the voltage dip is smaller as this issue appears, which makes me think it is the actual controller.

But I will do some more tests because I think I might have the answer.

Maybe, the controller is overheating? That's it? When I get this problem I am usually doing a long run, so the bike is at high power for a long period of time.

I can see that full power demands is pulling the voltage down by near 200mV per cell when things are 'normal', but when this 'derating' occurs it's more like 70 to 80mV dips. So, clearly the battery can deliver more, it's just not being asked to. So, I am now thinking 'controller temperature'?

[dominik]

Then the App should show a higher mos temperature.

[bob2.0]

Maybe, but I don't have the bluetooth dongle set up atm. Even if I had, would it report if it is derating when the temperature increases?

I'd see the temperature go up, and then it derates. I don't think that proves anything.

I have a new controller I bought when the motor went bad and I figured it must be the controller (when has 'rusty magnets' ever been a known ailure mode in a motor?!?). Anyway, it has the higher current rating so presumably won't heat up as quick, and bigger heat sink, so I will probably just move on to installing that.

It's just .... I can't find it! I tidied up the garage and put it somewhere safe.

[dominik]

With the App and Bluetooth you could have a look what are the settings for mosfet temperature and the app will show a temperature.

I think it is 60C start reduction 80C cut off.


The meter in the middle is mostfet temperature, on the right the motor temperature, when you have a temp sensor in your motor. I have one, but it was not set when I made the Screenshot, i was testing different firmware versions.

In the app it is called Controller temperature

[bob2.0]

I'm loath to install the new harness with the bluetooth dongle on it. Maybe I will make up a patch lead for the julet cable, one day.

I think there are a few ways forward;
i) now the problem is understood, live with it. I've always just ploughed on to get my journey done. Stopping for half an hour before steep bits would solve this, if it could cool down by then.
ii) replace the controller with the higher current controller, maybe not get so hot with higher current capable transistors
iii) ask you guys 'is there a way to reset the temperature derate on the controller', assuming it is a bit conservative and then the question whether I should push it a bit?

I might still have to do ii) if I do iii) and if either (iii) or (ii)+(iii) don't quash the issue, then I'll have to rely in (i), or carry on ignoring it like I have been doing.

It is reassuring to know that it is not a battery limitation. I'm now certain of this, with the new voltmeter display. Yesterday I came home from a long fast ride and it would pull the volts down 80mV from a standing start in '3'. Rested overnight, no additional charge added to battery, I can pull 220mV. Clearly, the battery is capable, the controller is not happy.

At the end of my 10 mile test ride today, I could see this same voltage dip was now around 150mV. So, it seems to be a gradual decline, and I guess I have hit 'max temp' before now, quite often.

Is there a way to adjust this controller temperature derate? Or is that a route to just blowing it up?

[dominik]

Is there a way to adjust this controller temperature derate? Or is that a route to just blowing it up?

I was not able to change it on my 72680 and my 721800.
To never get there I installed a fan with a temperature switch.

Controller and motor will overheat faster at low rpms with high load

I don´t know if the limit is konservative or not.

Here are some test , the basis is the other controller line from fardriver ND72450/72680/72100
You can see how fast they heat up with full phase current applied

https://dq-drives.de/en-eu/blogs/test-results

[bob2.0]

Controller and motor will overheat faster at low rpms with high load

Thanks. Useful. I was just about to ask about that, because we know the motor is basically a 'short' at low speed but I have also noticed that it will fly along at full speed (100kph) for longer if I keep up the speed than if I take a slower route and dawdle through towns and average 50kph.

In fact, I was wondering if I should actually INCREASE the speed settings so I can go even faster, at the expense of reducing the range, so as to delay the controller overheating?

..... I'm set on taking everything out of the garage and finding this new controller. I think there is no point discussing much more, until I have fitted the new controller.

Fitting a fan is another option. I hate noise though! One thing to say for the hub motor on these is that they do not whine like other ebikes with belts and sprockets.

[bob2.0]

To never get there I installed a fan with a temperature switch.

+1 idea


Fitting a fan is another option. I hate noise though! One thing to say for the hub motor on these is that they do not whine like other ebikes with belts and sprockets. I do appreciate that.

OK, so, did you fit the fan above the metal plate, like, under the fake fuel tank box?

Or pointed at the connector side of the controller, from underneath?

I have some larger sized 24V fans, maybe I get a separate DC-DC converter, and set it on a timer, so that it comes on after 20 mins riding, or a manual switch?

Might I please ask about your temperature controlled solution? What did you use and mounted where? I guess a simple mechanical switch type might be adequate?