Hello,
i'am planning to build my own VESC. But i don't want to use Shunt Resistors in the Phases.
I know that FOC will not work, but BLDC (trapeziodal) should work.
Is it possible to flash the VESC Firmware on a VESC Hardware without Shunts?
What would happen if the current measurement Pins are connected to a single shunt in Bat+ line? With INA169 or something else. Just for Current logging/Protection funktions.
I'am having an ESC build by my own with a Power PCB an a logic PCB. And i don't want to change the power PCB. I just want to have a better logic with better firmware.
best regards
John
As far as I know you need at least two low side shunts to use the VESC firmware also in BLDC mode.
I am not sure if running it without the phase shunts is possible. At least with the TI DRV8301 chip. Take a look at the documentation with that chip, there are lots of resources and kits from TI under the name InstaSPIN.
Cheers!
You could probably wire up some allegro current sensors to the wires going to the motor or use Lem sensors that go around the phase wires.
You could of course run without shunts or current measurement. However, you would be forced to run in duty cycle mode. So, for example, my 50V electric skateboard class motor has a resistance of 20mOhm. So when you suddenly go to half throttle, the VESC would normally limit the current to whatever you've configured but 50A max. But with 42V battery voltage, no current feedback and 50% PWM we'd get 21V / 0.02 ohms. = about 1000A. This would kill your motor controller in a few milliseconds.
In reality because you don't go from 0 to half power in microseconds, it will take a few seconds before you blow up your motor controller.
Maybe i should have posted the question in the Firmware forum.
I have build 2 Versions of my Brushless Controller. One with SimonK Firmware an one with my own Firmware (STM32). But SimonK is to old, BlHeli_32 is closed source and my own firmware is ok, but without configuration Software.
But all these firmwares are based on BLDC Mode (trapeziodal commutation) in Duty Cycle Mode.
So my question is: Would the VESC Firmware work without Shunts/Currentfeedback? (BLDC Mode/Duty Cycle Mode)
The other Question is: I have a Shunt in my Controller. 1 Shunt in the Lipo+ trace. So the VESC could get a feedback about the current.
I dont want to build an new hardware, just to test and see that its not working.
I have a VESC 4.7. Maybe i could remove the Shunts. But VESC Firmware for the 4.7 ist not the newest. An the VESC 6 has 3 shunts and maybe another logic behind the current sensing.
I hope you guys can follow my thoughts.
Roger, motors act like big inductors so the current won't be that high. All the hobby escs with no current sense manage duty cycle mode just fine without blowing up. Although most will blow in our applications because of that lack of current limiting and high motor loads.
The "no current sense" configuration works in a few cases. For example, when the load is predictable and low at low dutycycles. Like in a model airplane with a prop connected to the motor. But ask the model airplane pilots what happens when you crash with a little throttle applied.....
The ESC is used in a RC Racing Boat. I plan around 200A @ max. 10s. The actual design is tested to 100A. I have not more Load for testing.
So, i think i will have to test the "no sense current" with the VESC 4.7
Hi...i am a new user here. As per my observation you would be forced to run in duty cycle mode. So, for example, my 50V electric skateboard class motor has a resistance of twenty milliohm. So when you suddenly go to half throttle, the VESC would normally limit the current to whatever you've configured but 50A max. But with 42V battery voltage.
quick-turn pcb assembly
So, i did some tests without the shunts.
i removed the 100Ohm resistors between the shunt and the DRV OPAMP input. The first problem was, the the VESC measured 150A all the time. So i shorted the OPAMP inputs to ground.
And it worked. The VESC measure 0A (with some noise) and a could run the BLDC parameter detection.
Ok, the Parameters were wrong and i have to tune them. But after that, the Motor spin. Next Test will be under Load with ppm input.
But i got a dejavue. Some years ago i decided to build my own ESC with VESC 4.7 Hard and Software. I ordered one VESC 4.7 and made some test.
Problems: The detection didnt worked at all, or had wrong values. The VESC got to hot....And it is much to complicated. I want to swap the ESC in my Boats. Burning ESC's are a normal thing, so i want to build in another esc without using a Laptop.
Now at the weekend a saw that the Problems with the VESC are still the same. So i think i will switch to BLHeli.
The only thing to use use VESC is FOC. But, in my Layout there is no space for Phaseshunts. Just Shunts in the Lowside are possible.
So, did the VESC work with 3 Lowside Shunts?
Why is there room for low side by not phase shunts? I think you are just not wanting to take the time to change your design? It's actually a lot easier to add phase shunts from my experience as you no longer care about capacitance on the output side of the fets. They can be anywhere between the esc and motor, makes no difference.
I tested some possible changes in my design. But the PCB would get too big, or i have to remove one or two Fets in Highside or Lowside.
The main problem of my Design is: Highside and Lowside are not connected on the PCB.
The PCB is 40mm x 80mm with 12 DPAK Mosfets per Phase (6 Highside and 6 Lowside)....so the are just Fets....nothing else.
And the Copper in the Lowside (Ground Plane) have some Space for the Shunts
I still think your original question is of some relevance: can the VESC run BLDC mode with a single shunt in the supply or ground line? What is necessary to change?
In BLDC mode only a pair of phases leads the current, so there is no need for more than one shunt if you do not need or even want FOC. E.g. for all traction applications, where you drive a fan, prop and even wheels (maybe use position sensors). And there is no need to limit the duty cycle to 95% anymore, so the switching losses at 100% are eliminated. Additionally you could run a self optimizing timing algorithm, which minimizes the current through the single shunt in the supply line by try and error a certain phase advance timing. Some PWM action is still wanted for optimization of overall powerlosses.
Maybe there are other open source projects with this approach?
Hello andreas,
there are some open source projects for BLDC Mode drive. But they are not up to date and the VESC is the most powerful ESC.
Examples:
BLHeli_s, ST's 6 Step Library, SimonK
If i would have more time, i would do more test's on VESC without shunt's.