Forum

Hybrid Drivers  

  RSS
wlichti
(@wlichti)
Member

Hi Josh

I have tried both sets of drivers now and checked out the IGBT's for functioning.   I am still getting a fault after 500 uS.  I have tried the ceiling time from 95 which was in the software all the way down to 600 which is 21khz and the system still faults out at the same time.   Have you tried one of the new drivers in your system?  I have sent everything to Powerex they are going to look and see if they see anything.  The system is hooked to 110v thru the filter to the diode bridge which feeds the IGBT's.  There is no load on the system currently.  With the driver boards connected to the system and not to the IGBT's it does the same thing.  With the driver boards removed the system runs from ceiling to floor and then errors out on a floor error which it should with no lock. 

 

Wayne

 

Wayne

Quote
Posted : January 15, 2019 2:33 pm
Topic Tags
edgar reyna
(@edgar-reyna)
Member

the capacity   in your fillter have to be less than 10uf for experience, i will buy those gate driver too, 

which software are you using to simulate this ?

ReplyQuote
Posted : January 17, 2019 5:21 pm
wlichti
(@wlichti)
Member

Hey Josh,

Have been working the problem I purchased real Pwrx driver modules and that didnt change anything, I changed to 18V input to drivers and that didn't make any difference.  After studying doc from Mits and powerex I found a section in both that talked about false faults for larger IGBT's so in the Powerex BG2B design manual they suggested a 200pf cap between pin 2 and pin 4.  Tried that and didn't seem to work so the max value in the Mits doc was 10000pf so i put those in and it immediately ran.  Now I am only looking at input and fault line but it running for 350uS and faulting out is now not happening.   Probably need to go back and test again from the 200pf  and work up till I find min values that work and then make the decision on how much safety to put in.

Wayne

ReplyQuote
Posted : January 29, 2019 12:20 pm
wlichti
(@wlichti)
Member

I should be noted above that there was no load on the drivers IGBT's were not hooked up to IGBT's in both cases where it faulted out and when it ran.

 

 

Wayne

ReplyQuote
Posted : January 29, 2019 12:23 pm
wlichti
(@wlichti)
Member

Josh,

 

After testing caps from 200pf to 10000pf I found that 4700 worked and 3300 didnt.  I also when I got the drivers working it burned up the 12v trace on the core board.  I will need to cut the power from the core board and connect to the 12v directly in the power supply.

Can you verify what I have found. 

 

Wayne

ReplyQuote
Posted : January 29, 2019 3:10 pm
wlichti
(@wlichti)
Member

Josh,

 

I hope I have not patched over the problem but when I dropped the ceiling to 80khz the unit failed out again.  I then went back to 10000pf on the modules and it ran from 100khz down to 52khz approximately.  I am wondering if this is just masking or is a real fix.

 

Wayne

ReplyQuote
Posted : January 30, 2019 8:46 pm
Share: