Mosfet driver giving strange respond

[FerdinS]

Hello everyone;

I am working on my project which is boost converter and currently and I have a question for all you smart and experienced electronic masters.
I am trying to drive my N mosfet with driver becasue IC pwm pin is obviously not capable to do so by itself.

So I am using:

• Mosfet : SQJ454EP-T1_GE3
• Driver : UCC27517AQDBVRQ1

You can see my wireing also on the picture : 1 and also responses before and after resistor Res3.

So the question: What could couse that strange respond ? ... so driver is giving me pwm output but after that resistor that is working as current limiter pwm becomes really strange.
Thank you a lot for you answers.
Best regards,
Ferdin Shotani
" alt="" class="bbc_img" />
" alt="" class="bbc_img" />

[jmelson]

R42 is I000 Ohms.  FETs have huge input capacitance, your gate is not charging up enough to turn the FET on.
About 10 Ohms would be a normal value.

Jon

[FerdinS]

So what would be best solution ? putting there smaller resistor ?

[jmelson]

So what would be best solution ? putting there smaller resistor ?

Yes, try 10 Ohms.  But, you can look up the gate capacitance, and figure you want the transistor fully on in less that 1/10th of the PWM period, calculate how long it will take to charge up.  Note, the driver has some current capability, too.

Jon

[BrianHG]

Check your datasheet on the GATE input capacitance of the 'SQJ454EP-T1_GE3'.
Then with a series of 1K from your 0-12v drive, calculate how fast the rise and fall time of the GATE on the mosfet will be.  This should answer if the 1K is a wise choice or not.  Note that some mosfets have GATE capacitances way above 5000pf.

Hint: There is a definite reason the GATE driver IC UCC27517 has a 4 AMP high and low drive.  Will using that series 1K end up using anything near that IC 4 amp drive capability?

[FerdinS]

To be honest, I saw 1k resistor on datasheet and that is why I put it there. Also tried without it and it came up with fire 
So I am kinda scaried to put smaller resistor since connections are pretty small and voltage that goes out of driver is 12 volts, so there would be 12V/10ohm = 1.2 A... isn't that too much ?

Thank you for fast respond.

[FerdinS]

Input capacitance of my transistor " SQJ454EP-T1_GE3 " is Typ : 1990 and max 2600 pf.
But still, puting there only 10 ohm would make me 1.2 A which I think is too much, don't you ?

[BrianHG]

Read my above edited post:
https://www.eevblog.com/forum/projects/mosfet-driver-giving-strange-respond/msg3182828/#msg3182828

Also, nowhere in the datasheet does it say that R1 should be a 1K resistor.  (R1 is the example resistor in the datasheet)

Also, that 1.2amp is only while the GATE of the mosfet is at 0v, and, how long will it be there?

[Benta]

So what would be best solution ? putting there smaller resistor ?

Dunno where you got 1 kohm from. Try with 10 ohms and add heavy decoupling directly on the UCC part, eg 1 uF and 10 nF in parallel.

[FerdinS]

Thank you everyone , as it seems the resistor was really cousing a lot of problems.
But now when I try to measure when and how mosfet is switching, I can see the picture below.
What do you think, is that how switch has to work when I have pwm 50% ?
Thank you again for answers.

[T3sl4co1l]

That is a normal DCM (discontinuous conduction mode) waveform.  Whether that was you intent behind the the L and Fsw values chosen, I don't know.

Tim

[waytec]

Yes I agree too that the waveform is quite normal.......

When the FET is switched off the energy from that 200uH is reversed in polarity and has nowhere to go causing that bubble effect or bubble distortion...
As an experiment try changing the Duty Cycle and the bubble will also change and at one point of DC will all blend into the squarewave and look normal...

Thats all from me...



Wayne

[FerdinS]

Thank you a lot my friend. I used a smaller resistor from driver to mosfet and it looks now there is no distortion anymore. Thank you a lot everyone. It solved my problems.

[FerdinS]

Thank you a lot. I see now my boost converter works on discontinous mode mostly. I put output resisor low ( lower resistance ) and it come to continous mode.
Is there any other way I can do that ?
Maybe by changing frequency of PWM ?

Thank you for answers

[FerdinS]

I tried with chaninging output resistance and it looks like there is also change from discontinous to continous mode of my boost converter.
Is there any other that you might know ?