Interest in a "Power Board" devkit for homebuilt controller?

[rhitee05]

The development board makes it easy to access all of the PWM signals and as such I would like to drive a more capable power stage board(100V/100A), perhaps using a gate drive optocoupler (e.g. HCPL-3180). The DirectFET parts you mentioned are attractive yet seem hard to work with for the do-it-yourselfer. Anyway, a modular power board that includes the fets, gate drivers,caps, current shunts, and an abundance of heat dissipation is what's really needed. I know, easier said than done...

Very nice little board you have there, and nice work using it.

What you described is more or less what I'm trying to accomplish. The board would include all of the high-power and analog circuitry required. Obviously that means FETs, gate drivers, caps, and one or more current shunts (only one req'd for normal control, two req'd for more advanced control schemes). I also plan to include signal conditioning circuitry and a couple of ADCs for logic boards without analog inputs. Most dev boards include a lot of functionality and allow the user to choose what and how they want to use it. I plan to do the same thing. It's very easy to include a lot of features in the layout but only populate what's actually needed.

 

[rhitee05]

That would be cool if we could come up with a controller interface standard between the board you come up with
and the board that I come up with!?!?

A standardized interface would be a useful thing. Let me take a shot at what signals would be needed.

- 6 gate drive signals for each FET/bank of FETs. You could also do this with only 3 signals (one per half-bridge) if you use tri-state signaling. That is, low = lower FET on, high = upper FET on, high-Z = neither FET on. This scheme would require a little bit of logic on the gate driver side, but inherently prevents shoot-thru.
- Battery voltage, although what you probably want here is a scaled version for measurement
- Supply rails, which would be 12-15V gate drive supply and 3.3-5V logic supply. Most logic boards will probably want 3.3V as their supply, some may want 5V.
- Current sense. You could just provide the +/- directly from the shunt, but it would be much more useful to put a differential amp at the shunt and provide a scaled voltage suitable for direct input to an analog input. Trying to put the differential amp off-board would make it harder to keep a clean signal. I'd suggest making room for two current signals to support systems that measure two phase currents rather than just the battery current.
- Phase voltages (for sensorless control). As with the battery voltage, you could provide a scaled version of each signal. You don't actually need to measure the exact voltages, though, for sensorless control all you need to do it know the zero crossing when the undriven phase is equal to half the battery voltage. You could put comparators on-board and just provide flags, which would save 3 analog inputs on your logic board.
- Including pins for a digital bus would allow room for lots of expansion, onboard ADCs, etc. I2C is a popular choice and only requires 2 signal lines.

So, 6 drive signals, 2 power rails + ground, 2 current sense pins, 4 voltage sense pins (battery + 3 phases), and 2 pins for I2C. That's 17 pins required, so make it a standard 20-pin ribbon cable to leave a little extra room.

 

[sjacome]

Rhitee05 when i read the post title i imagined the e-bike valhalla - custom controller - how? - Open-source hardware + custom adapted to ur specs open-source software -Imagine timming, power,thermal management, u name it.

allowme some brainstorm webclipping im gathering about some time.


Developing Brushless DC (BLDC) motor control using PIC18Fxx31 - Part 1 -- http://www.youtube.com/watch?v=j1mlnYLtPxE

Developing Brushless DC (BLDC) motor control using PIC18Fxx31 - Part 2 -- http://www.youtube.com/watch?v=e7dCqSffcKA

Developing Brushless DC (BLDC) motor control using PIC18Fxx31 - Part 3 -- http://www.youtube.com/watch?v=QC8RzDmDZfM

Developing Brushless DC (BLDC) motor control using PIC18Fxx31 - Part 4 -- http://www.youtube.com/watch?v=_suQfKVaSk4

http://www.mikrokopter.de/ucwiki/en/BrushlessCtrl

 

[rhitee05]

That's pretty much the idea I'm hoping for. If you have a FET bridge which makes all the requisite signals available, you can do pretty much anything you want with the control. It would only be limited by programming skill and the limits of the hardware.

 

[sjacome]

i am wondering for a while in the idea of 8 - bit MCU like Atmega 368 (arduino open source platform) and design a sensored controller in a Arduino shield. The shield hould be suitable for other platforms like ur Microchip MCU or whatever other developers are confortable. Just have to pack the MCU basics in a pcb shapped like a arduino platform.

the FET stuff and all the power electronics are the obstacle since im an electronics newbie with very few resources, but buyed some electronics magazines and maybe i will copy/past a suitable FET bridge/circuit that suits the R&D needs.

Read also somewhere about high power, high voltage optocouplers but surely high cost and hard to find also.

Of course hope to get some hints here too. keep the good work.

 

[rhitee05]

Yes, a generic bridge circuit could be driven by any controller architecture of the user's choice. I'm also including on-board ADCs in my design, so it will be compatible with FPGAs and other controllers without built-in analog input. If the controller does have analog input, you can leave out the ADC chips and connect directly to the analog signal. Flexible for many purposes. Put everything on the board now, then assemble only the parts that you need.