OleyT said:
Now I think I understood. Sorry, it's just because my technical english (and my english in general) is not really high-end as you might have noticed
Actually, it's better than a large portion of people I know and talk with that grew up here in the USA.
Ok, if Grin doesn't answer then I will get an Voltmeter in the hardware store and test the different connector sites (with your help hopefully :wink: ). Just asking myself what consequences that would have if I get different signals at the connector sites.
Basically, if the tests turn out ok on the hardware, then Grin would probably have to help with why it doesnt' work at that point, because it *should* work if all the signals are present.
I have two last questions for you:
Do you see any issue with driving the motor sensorless temporarily? Is it harmful for the motor itself or the other parts?
No, it's a completely normal operating mode. I'd guess that a large fraction of low to medium power bikes are using generic sensorless controllers (without good position detection routines, and so noisy/juddery/hard to startup from a stop) just because it is cheaper and simpler (no hall sensors needed in motor, no hall wiring, no phase/hall combination worries, etc).
And secondly: If I have understood correctly, you are driving your vehicle sensorless as well?! Do you have any good advice how to keep the stutters, jerky phases (especially while accelerating) on a low level(avoid them. I was trying to start rolling without any assist and as soon as I reach a certain speed I will start the assistance from the motor. That works okayish but not perfectly...
The rightside motor on the trike is using a generic sensorless controller and a big MXUS 450x motor. Given the 500lb+ weight of the trike with me on it, it does pretty good.
The grinfineon is running sensored, and makes startups much better. But it was sensorless at first until I found and fixed the hall connector problem in the motor cable, I forget what ti was but probably something I didnt' crimp well enough, or a pin I didnt seat well enough that backed out. When it was sensorless it was probably fairly typical of such stuff; I don't recall exactly (I'm sure my SB Cruiser thread has that info in there somewhere between last summer and winter).
Some sensorless controllers are better at it with some motors under some conditions, and worse under others.
I also tested the Incememed SFOC5 sensorless FOC controller, which usually worked pretty well (for a "beta test"), but it was not very good at starting up from a stop. Supposedly the firmware changes since then would fix that, but he's also added the option to have hall sensors. I discontinued the testing before then for a number of reasons, so I don't know if any of that works.
The new controllers I will be using once completed are a very DIY thing, over here:
https://endless-sphere.com/forums/viewtopic.php?f=30&t=105711
and they are capable of sensorless only, or a mode using sensored-start / sensorless-run so that they use the best features of both modes. But you can't buy any version of the Lebowski controller off the shelf; you have to build at least some part of it; the simplest is something like that thread (mating the brainboard from one of the people that builds them to an existing powerstage). I will probably be using the sensored-start / sensorless-run mode.