bjt totem pole gate driver not working with sawtooth wave

TeeKay6

Joined Apr 20, 2019
572
The totem pole works fine square waves but not sawtooth. The capacitor should have a few amps peak in it but that only happens with the sqaure wave input.
Thanks for including your .asc file; it runs on my PC. However, you have mislabeled the 18V DC power source as 25V. Confusing.

I do not understand you: "The capacitor should have a few amps peak in it but that only happens with the sqaure wave input." Please restate what you believe to be wrong.
 

crutschow

Joined Mar 14, 2008
25,382
As Alec noted, it's the fall-time of your signal.
The peak capacitor discharge current is determined by the rate of change of the signal.
If the capacitor voltage changes essentially as fast as the input signal to the driver, then the driver is doing its job.
 
Last edited:

Thread Starter

pager48

Joined Nov 25, 2018
161
As Alec noted, it's the fall-time of you signal.
The peak capacitor discharge current is determined by the rate of change of the signal.
If the capacitor voltage changes essentially as fast as the input signal to the driver, then the driver is doing its job.
This totem pole wont work for driving large igbt gates. They should be able to output large peak currents into a capacitive loads.

Whats the issue with this driver?
 
Last edited:

Thread Starter

pager48

Joined Nov 25, 2018
161
Depends upon how large an IGBT gate capacitance and how fast you need to switch it.
This totem pole is made of 5amp DC BJTs and it cant put =>2 amps peak into a small 8nf capacitor. It should be able to put much greater peak currents than its 5amp rating.
 

crutschow

Joined Mar 14, 2008
25,382
This totem pole is made of 5amp DC BJTs and it cant put =>2 amps peak into a small 8nf capacitor. It should be able to put much greater peak currents than its 5amp rating.
We seem to going in circles here. :rolleyes:
It's already been stated that the peak current depends upon the signal rise/fall times.
 

Alec_t

Joined Sep 17, 2013
11,634
The totem pole is working fine. Expand the trace around the pulse falling edge and you'll see. Your 470n cap has a discharge time of about 14uS and Vbjt is closely following the cap discharge curve. It only needs a few mA to discharge the 10n cap completely in 14uS.

Edit: Crutschow pipped me at the post.
 

Thread Starter

pager48

Joined Nov 25, 2018
161
Other than a gate driver what other way can a gate be driven from this oscillator? This totem pole doesn't supply the peak current.
 

crutschow

Joined Mar 14, 2008
25,382
This totem pole doesn't supply the peak current.
You sound like a broken record (go ahead and ask what's a record), repeating the same thing as if you have no comprehension.
What is this obsession with peak current? :confused:
Did you read somewhere you need a particular value of peak current for a particular capacitance and rise-time?
A different driver won't provide any more peak current as the driver is delivering all the current the capacitor requires to charge at the given fall-time.
It's not possible to provide more peak current.
Why can you not understand that?
 

Thread Starter

pager48

Joined Nov 25, 2018
161
If you can quantify what peak current you need, and show exactly why you need that particular value, then we can answer the question.
Switching losses would be high for =>100nC IGBTs using only the ramp generator without delivering peak current into gate of IGBT.
 

crutschow

Joined Mar 14, 2008
25,382
Switching losses would be high for =>100nC IGBTs using only the ramp generator without delivering peak current into gate of IGBT.
You are still dealing in generalities.
To do a design you need to list quantitative values for the current you need and how those values were determined, and you haven't.

Tell us the highest sawtooth frequency, and the specific IGBT you will use.
 
Last edited:
Top