Hello,
I'm currently trying to develop a 60kW motor drive for Emrax 188 motors. I'm in the very early stages of the design process, building a basic block diagram and researching specific part numbers. I've taken a majority of my inspiration from Marco's 150kW VESC design.
While I have a solid understanding of the underlying electronics behind any ESC, I am unsure of how to correctly interface the STM32F4 with all of the associated I/O. The pinout for the STM32 is obviously very important and with the variety of changes I'm making to the schematic its hard for me to be confident that the pinout will be correct. I've suffered through enough bad pinouts for a lifetime.
Also as far as I understand, resolver feedback negates the usefulness of any hall effect motor position feedback, but why does Marco's design still use both?
So, in the end my question is how to ensure that the STM32F4 firmware is set up for the exact pinout of my schematic and secondly how to implement resolver feedback.
Attached in the planned circuit's block diagram.
Thank you!
Hey,
Surely you can almost keep all the pins identical to the original VESC6? SPI on hte same pins, PWM outs on the same pins, voltage and current monitoring on the same pins etc..
Provided you scale the voltages & currents appropriately the VESC won't be able to tell the difference.
Resolvers can be quite tricky to work with and the VESC isn't commonly used with them so unless your resolver chip operates identically to the AS5047P's that have code written for them you're going to have to make a few firmware changes. Marco may have kept both as a surefire way of getting his 150kW motor up and running then worked on the resolver later once he's sure the rest of the ESC operates correctly.
I'm not too clear on the exact details of setting a variant definitions file, Benjamin's still lagging on that promised explanation but that would be where you need to do it. As for resolver feedback I think you'd have to add that to the firmware yourself, try checking out the SPI absolute encoder code and seeing if you can copy much of that. I'm guessing you'll just want to be reading slightly different registers for an absolute ~12bit angle value.
Hi Acable,
in my control board, HALL position sensor inputs are there only to avoid explaining why I removed them. I'm strongly against using hall sensors in high power drives, and in my boards its only available as testpoints witout proper ESD protection and they are not routed to connectors.
About your resolver chip, I provided firmware support for it and a while ago Benjamin merged it in the codebase, so its ready for you to try out. In VESC Tool you just select General->Sensor port mode->AD2S1210 Resolver. Let me know how it works.
To make sure you don't screw up with the pin assignments, try to stick to my design. If you want to get creative with pin assignments you need to read the very long reference manual and datasheet of the F4 mcu.
Um, now that I see the schematic, your microcontroller GND must be isolated from the battery (-). Also that gate driver without short circuit protection will cost you a lot of $ and fires when you start tuning that kind of parallel powerstage. If you PM me I'll send you information about our incoming board, you might want to test drive it, you can avoid lots of headaches.