MOSFET gate drive problem

mosfet gate driver

Hi All

I am using LTC3810 in one of my application.
As given in the datasheet ,I used Si7852 / Si7174 MOSFETs as TOP and BOTTOM Mosfets.
My Input is 24V ( range is 21v to 62v)and Out put is 12v @ 3A and switching frequency is 250Khz. This gives us duty cycle of 50% at Swithcing node.
I got very clean switching node waveform.

My application also include 12V at higher load that is 12v @ 7A.
But since Si7852 or Si7174 used to get heated up when 7A load was connected.
So I changed the MOSFETs and used D2PAK MOSFET . That is ST micro electronics make STB160N75F3.
but in this case , I have output 12V ,not a problem. But when i looked into Switching node waveform (Vin=24v)
The wavform is getting corrupted fully. I was expecting a duty cycle of 50% here.
What could be the problem ? Is the gate charge capacitan ?drain to source capacitnce ?
why i am not getting pure duty cycle of 50% with 250khz as frequency ?

One thing i should say is , since original MOSFET was POwerPAk-8 , and now i am using D2PAK , i made some jumper and cutting on the board.
is that the problem ??

I have attached my circuit here.

thanks in advance

gate drive

The wavform is getting corrupted fully.
How does it corrupt?

si7174

One thing i should say is , since original MOSFET was POwerPAk-8 , and now i am using D2PAK , i made some jumper and cutting on the board.
is that the problem ??

Click to expand...

Yes, it might be a problem. Since the waveforms are corrupted, as you say, this might be due to ringing on the pulse front ends. The ringing can be severe. Your jumper and cutting might have brought some stray inductance. A fast pulse has a lot of harmonics. Some of those frequencies might excite the LC resonator made with the stray inductor and the MOSFET capacitance.

Try to insert some small resistors in series with the gates and see if this corrects the problem. The resistors and the gates capacitance increase the pulse rise/fall times so the ringing will disappear. Try different values for these resistors from say tens of ohms to a couple of hundreds. Be aware that, since the rise/fall time increases the switching losses increase and the transistors might get hotter. This is a trade-off between efficiency and a clean signal.

If you need to increase the output current capability I would use the same transistors with an extra pair in parallel, rather than changing the transistors. I did this several times and works very well.

mosfet gate drive

Hi AdrianN,

thanks for your suggestion. To keep MOSFETs parallel , i have got space problem. and i am not allowed to make both side placement.
So i may not be able to do it. I need to check with series resistor with Gate. DO you think , increase in Gate capacitance is the Problem ? Becasue , earlier MOSFETS , Si7174 were having 34nC of Qg and now I am using STB160N75F3 , which is having almost three times i,e 100nC. So I suspect this this may also a well reason for the problem.

thanks a lot.
Manju

waveform of mosfet gate pulse

DO you think , increase in Gate capacitance is the Problem ?

Click to expand...

Very possible, as an increase in capacitance means that the ringing will occur with lower stray inductance for the same ringing frequency.

Try the gate resistance and also try to use TO220 transistors for your prototype. That way you don't need to cut the PCB traces and the transistor connections will be cleaner. You can also mount a small TO220 heatsink on these transistors if necessary.

If this works, and you have a functional prototype, then you can create a new PCB with the same transistors but in a surface mount package.

gate drive problems mos

Thanks AdrianN,

Let me try that option and will update.

Cheers!