[Yukon_Shane]

Originally Posted by Tesseract

Ah... good point. I'm kind of embarrassed I didn't notice that myself! We'll address that in the next code revision, then.

It's a pretty easy thing to miss and I think there's good reasons for choosing 40C as the cooling system setpoint: for one thing the soliton1 doesn't start to derate until 55C so 40C seems pretty reasonable but also the steady state temperature for my controller seems to settle in around 30-35C so having the cooling turn on at 40C seems logical.

It's just the unique derating curve of the soliton jr that makes a lower cooling setpoint possibly a better choice. You could always increase the temperature at which the JR starts to derate power

[GerhardRP]

Originally Posted by Yukon_Shane

It's a pretty easy thing to miss and I think there's good reasons for choosing 40C as the cooling system setpoint: for one thing the soliton1 doesn't start to derate until 55C so 40C seems pretty reasonable but also the steady state temperature for my controller seems to settle in around 30-35C so having the cooling turn on at 40C seems logical.

It's just the unique derating curve of the soliton jr that makes a lower cooling setpoint possibly a better choice. You could always increase the temperature at which the JR starts to derate power

Perhaps the liquid cooling should turn on at 40C or motor current > 300A, say.
Gerhard

[dougingraham]

Almost a month has passed since I started using my EV as my daily driver and I came up with one more item for my wish list. And it is a good one. It is traction control. Yep traction control. And it can be done with just software as everything you need to know to detect a loss of traction is already there. If you have traction the current will go up as the voltage increases. If the current ever suddenly drops as the voltage goes up you have lost traction and to regain it all you need to do is lower the voltage until the current goes back up. You could double check this by watching the RPM. RPM will spike if traction is lost. When you reach the RPM where the back emf starts to kick in the current will drop but at this point the controller is already going to be at 100% duty cycle.

Of course this would require an input so it could be disabled by a switch. The racers would probably love this. An output that indicated traction control was operating would be a good addition as well although I would expect you would know this because the tires should be chirping kind of like they do when ABS kicks in during hard braking.

[Qer]

"just software" he says...

[dougingraham]

Originally Posted by Qer

"just software" he says...

For a lot of my professional career I wrote code in assembly and I didn't mean this in a disparaging way. What I meant was that existing hardware could (probably) take advantage of this "wish" without modification. I don't know what the actual internals of the Soliton 1 look like. When mine showed up I wanted to look inside but I managed to resist the urge. It is entirely possible that there isn't enough performance in the cpu to support the additional calculations necessary in the inner control loop. But you know better than anyone else what that part of the code looks like and how changes there can affect other things.

[Qer]

Originally Posted by dougingraham

What I meant was that existing hardware could (probably) take advantage of this "wish" without modification.

Mmmm. I've toyed with ideas like this back and forth for a while, but the tricky part is detecting things like that in a reliable way. Ordinary you detect it by comparing speeds between the tires, if they offset too much one tire is spinning. Simple. Detecting it on only one set of data... Not so simple.

I really don't want to add something that screws up more than it works correctly so if I add something like that it has to be so reliable that it doesn't generate tons of support questions.

[Otmar]

Originally Posted by Qer

"just software" he says...

I call that a SMOP, a Simple Matter Of Programming. It's what the hardware guys say to tease the software guys. What's the problem? It's just a SMOP!

Says the guy who is in the middle of implementing those SMOPs since I do all the hardware and the software for the Zilla. I have a lot of arguments with myself sometimes...