Paul and Sabrina's Cheap 3 Phase Inverter (AC Controller) with Field Oriented Control

[MPaulHolmes]

Good idea Jack! That would be bad. The micro-controller would never normally allow it, so it would have had to be something really bad (or coding mistake hehe). The microcontroller has it's own fault pin that does to the pwm outputs whatever you want it to do under a fault condition. Maybe if we assume the microcontroller craps out and the High and low are both stuck high for some reason, maybe a fast blow fuse? I would hope it would be a rare enough situation. Do they make fuses that could blow before the destruction of the igbts?

[mrbigh]

Originally Posted by MPaulHolmes

Do they make fuses that could blow before the destruction of the igbts?

Paul, PM sent..........

[jackbauer]

There are so called semiconductor protection fuses but i wouldn't bother. I only suggested it as back in the day i used to design hf ballasts for fluorescent lamps that used a push - pull mosfet setup and the most common failure was both devices turning on at the same time. I designed one with additional circuits that prevented this failure but the bean counters deemed it too expensive and made me remove the parts as it was cheaper to let a percentage of the ballasts blow then solve the problem! As the igbts are the most expensive parts it makes sense to protect them.

[mrbigh]

Being that lately I have enough spare time, I did some research on the basics of the AC motor controller and found on the PIC website something very interesting, a complete block diagram and some schematic of the AC controller final stage, not very complex and as you can see, the IGTBs are paired for each AC leg with out hardware (fuse) protection. I imagine that if it will be an inrush load above the software parameters, the PFC based power supply will cease operation protecting the final stage devices. Also note the implementation of the "BRAKE Chopper" for recapture of Regen back to the main raw supply.

[dcb]

that power factor controller section is exactly the boost converter diagram on the prius as well.

[mrbigh]

Originally Posted by dcb

that power factor controller section is exactly the boost converter diagram on the prius as well.

well, the basic of designs are almost the same, these are simple circuits though that need to be scalled up.
I have the same concept in my ManzanitaMicro PFC40 HM AC-DC/DC charger

[jackbauer]

Thats fine for an ac based supply. Now try that with 12 series connected odyssey agm batteries capable of sustaining 5000amps for 10 seconds

[mrbigh]

Originally Posted by jackbauer

Thats fine for an ac based supply. Now try that with 12 series connected odyssey agm batteries capable of sustaining 5000amps for 10 seconds

Now you are pushing the limits........

[jackbauer]

Didn't mean any offense with my last comment just wanted to point out that an ev inverter is a slightly different beast to an off the shelf vfd

what i'd like to see with this project is that it stays within the limits of what a person with a soldering iron and a few hand tools can build given a pre made pcb. I'd also like to see some smart design in there to prevent or at least minimise the probability of major failures. thats what i like about the dc controller. Its small , smart and easy to build. No crazy magnetics or stupid qfp smd parts etc. I built an atmega programmer in 30 minutes from some junk box parts.

[MPaulHolmes]

Believe me, I'm not going to use surface mount components! I definitely want smart safety stuff built in too. At some point we should brainstorm all the horrible things that could happen, and try to figure out solutions that are as simple as possible that protect against those things. That will be fun!