Little Bit of a Massive Problem

A

Alexsgarage

New Member

• Oct 31, 2009

• #1

I am building a PWM IGBT driver and have encountered a little bit of a massive problem, when the transistor is off the voltage is still 5V or so above 0V. See the picture below, the straight line is 0V.

 

Attachments

• PA310404.JPG

  3 MB · Views: 169

B

BrownOut

Banned

• Oct 31, 2009

• #2

You're thread title made me chuckle. Can you post a schematic?

 

A

Alexsgarage

New Member

• Oct 31, 2009

• #3

I had an original schematic but I have modified the circuit so much that I don`t think it will be much use but I will draw up a quick block diagram.

 

B

BrownOut

Banned

• Oct 31, 2009

• #4

I would like to see details of the driver, and where the measurements are being made. My crystal ball doesn't work

 

A

Alexsgarage

New Member

• Oct 31, 2009

• #5

BrownOut said:

My crystal ball doesn't work

Click to expand...

LOL Funny

Here is the block diagram, the PWM is an SG3524 and it is transformer coupled to a comparator, the output which then drives the gate of a small transistor. The measurments are being made at the output and ground.
This is a high side driver.

 

Attachments

• block diagram.jpg

  30.1 KB · Views: 178

Last edited: Oct 31, 2009

B

BrownOut

Banned

• Oct 31, 2009

• #6

I'd be concerned about the voltage at the base of the transistor; make sure it's really turned off. If it is, then the driven circuit might be holding a change. There is a potential problem with the "+" input to the comparator, the diode prevents the charge from flowing away from that node during turn off. But I suspect that there is a resistor in there somewhere that isn't shown on your diagram.

 

A

Alexsgarage

New Member

• Oct 31, 2009

• #7

It can`t be the diode because the same thing happens on the low side driver and that one is just connected with a single wire. Could it possibly be the pull up resistors?

 

B

BrownOut

Banned

• Oct 31, 2009

• #8

It could be. That's why I asked you to measure the base voltage. The pullup probably needs to stay in the circuit, but it might be too low valued.

 

Last edited: Oct 31, 2009

K

kpatz

New Member

• Oct 31, 2009

• #9

What's the value of the pullup? Try the highest value needed to turn the transistor on when the comparator is "off". For an IGBT something in the 1M+ range should work.

Is the comparator an open-collector output? Depending on the application you may be able to use an op-amp and eliminate the pullup.

Or use a PNP transistor off the comparator to drive the NPN, eliminating the need for the pullup.

 

Last edited: Oct 31, 2009

A

Alexsgarage

New Member

• Oct 31, 2009

• #10

Thanks kpatz the PNP transistor works nicely.

 

Last edited: Oct 31, 2009