I would like to run 0.5 seconds bursts of 130 A (maybe even 150A) at 48V. Is this possible/feasible with VESC 6?
I see there is a limit of 100A in VESC Tool, but sensor range and FETs should be capable of 150A intermittently, right?
Do you have any advice on how to increase the max current?
The application is a rowing device using 8-pole motors with lambda=8mWb, Rs=6mOhm, L=60uH. High current is needed in order to get max utilization of my motors and mechanics and the most powerful rowing strokes. I can add more cooling if needed -- there is plenty of water.
When you say 150A, you mean 150A RMS AC current. This means the peaks extend above 150A... about 210A. And that is a value that can no longer be measured using the shunts.
Similarly, when the fet-manufacturer specifies some current "peak", they mean very, very short periods. So 0.5S is more or less "continuous" for the FET. I don't think 0.5s of 150A is achievable.
Do you perhaps know the measurable current range of the VESC6.4? I would love to know it, I currently building a water cooling setup to see how far I could push it ^^.
What I remember is that the scaling was the same as for vesc 4.x But I'm not sure. I'd have to look in the source. ... checked. Yes.
So in theory you can measure +/- 165A. However analog components are usually not that linear near VCC or GND. So for example, the ACS758-200 specifies a 4.0V voltage swing away from the 2.5V mid-voltage (when powered with 5V). So the bottom and top 500mV are not usable. How much of the theoretical range is "not usable" I don't know, But a common sensor from Allegro tells us to not use the top and bottom 10% of the range.
It sounds like you are multiplying by sqrt(2) as you usually to to get the peak value from the RMS value of a sine wave. This does not really work for the currents in the VESC running FOC. Here is a code snippet from my SVM simulator:
With only Q-axis current (which is the commanded current on the VESC) you can try to follow how iq maps back to ia (the current in phase1, which has the same peak values as for the other phases). As you can see it maps to i_alpha directly, which itself never is higher than iq (when id is 0) as the sine value of the rotor angle (s = sine, c = cosine) never has a magnitude greater than 1. This means that the highest value a shunt will see is exactly the commanded current. This is not very intuitive at first, but if you do the math and look at the plots it is what you will get.
As for the question in the thread, the theoretical maximum current that the VESC can measure is +- 165A, and from what I have tested so far it is nice and linear up to at least 150A. The problem if you want to push the limits is that the firmware does not accept values higher than 100A since I haven't done any testing with more than 100A sustained. I did some testing on my RC car with 120A which worked fine, but those accelerations never lasted for more than a second or so. If you want to push the limits, there is a firmware included called VESC_default_no_hw_limits.bin that permits setting all limits to any value (you have to check the box "Show non-default firmwares" to see it). You could give it a try at your own risk, but I cannot guarantee that the VESC will survive. I think this is an interesting test, so if you want to give it a try anyway and the VESC fails you can send it to me and I can try to repair it if possible. For this test I would recommend starting by setting the absolute maximum to something like 160A, the current to 140A and increase the DRV8301 OC Adjustment to something like 20 (otherwise the DRV will cut the current). Do not disable the DRV OC mode as this will most likely kill the VESC in a test like this.
Thank you for answering. I will try it later (probably in August), after testing other parts of the system.
RWolf: I intended to ask for Q-axis current 130 or 150 A, i.e. DC or amplitude (not RMS).
Hi, did you finish your test? I'm also curious how far we can push.