• Welcome to our site! Electro Tech is an online community (with over 170,000 members) who enjoy talking about and building electronic circuits, projects and gadgets. To participate you need to register. Registration is free. Click here to register now.

Mosfet driver problem

Not open for further replies.

I have bought IR2101 mosfet driver to drive my IRFB3034 N-CH mosfets for my brushless dc motor controller.
When i soldered everything to my new PCB for that application, motor rotates, but doesnt have that momentum that it had with my old version of PCB board.

The reason for that is, in my opinion because high side of mosfet driver output doesnt work correctly.
When LOW INPUT is HIGH, LOW OUTPUT is also HIGH, as it should be.
When HIGH input is HIGH, HIGH OUTPUT is HIGH when it should be only when motor is rotating, and when you force to stop a motor with hand, it stays that way because even with high input on high side driver, output stays low, and there is no current to the motor to start rotating again.

I have to say that on schematics on bouth PCBs (old one that is working, and new one that is not) is THE SAME and i cant figure out what is wrong with it.
Could it be the IR2101 or mosfet problem? Or something else. Maybe faulty Ir2101 on this new board?
Please help me if you know :)



Well-Known Member
Most Helpful Member
Hi rascupanamuha,

Your pdf file does not open.



Active Member
The PDF opens alright for me (using evince) - don't know what the problem might be there.

How long are your "on" pulses on the high side (at "HIN A" and "HIN B")? I presume you know that you can't keep the high-side MOSFET on indefinately because the bootstrap capacators will discharge. That is to say, each phase must start by driving low before it can drive high.

That said... if the schematic is the same as a previous version of the board that worked prefectly then there are really only two possibilities:
1) You've got a faulty or incorrect part, or
2) There is a mistake sonewhere in the new board layout and the circuit is not actually the same as the working board.

Did you assemble the board yourself, or did someone else do it? If so, is it possible that they've placed a wrong component somewhere or miss-wired the connections to the motor?
Thanks for your answer!

I dont know how long my "on" pulses last. I know that high side MOSFET can't be high for a long time, but once again, schematics for that 2 boards are the same, so it should be the same.

1) I also think this might be the problem because i can't find anything else that is wrong.
2) Schematics are the same. Because when i made working BLDC controller, i c/p that part to my bigger PCB with other functions (BMS, etc). So the only thing that has really changed is routhing. Maybe some wires are shoter or longer, wider or thiner but i dont thing that it has something to do with this (not so many big changes from working PCB design)

Yep, i have assembled it all by myself. Because i thought the problem is with the soldering or some faulty components, i soldered 3 PCBs and non of it is working as previous PCB design (motor has moment, but it can be easily stopped)

I have tried IRF3205 and IRFB3034 mosfets and on my working version are IRFB3006 but i can't believe that this is the problem (if all parts are original and not copy of something else with this label on it)


Active Member
The flying capacitors (C1/C3) and those on the other FET drivers do not need to be 470uF. Is this in an application note or have you picked these values yourself? These capacitors store the charge to drive the high side FET, so they really only need to be 100x the input capacitance of the FET. if the input capacitance of the FET is, say 2nF (reasonable for a large FET) then the flying cap needs to be about 220nF. Much more than 1uF in this place could cause problems with the capacitors charging up and if the low side FET is ON for a only a short amount of time, these caps might not be charging up in time for the next high side pulse. try removing the 470uF and see what happens. Better still, remove both and fit only a 1uF in their place.

Finally, with FET drivers, you need a good decoupling cap close to the Vcc pin (to provide a low impedance path for the low side FET drive) so C5 etc has to be a ceramic cap and physically close to the Vcc and COM pin
Wow, i didnt know that those capacitors (C1,C3 etc) need to be about 100x the input capacitens of FETs. Input capacitance is 10nF so C1 should be 1uF. I replaced that caps, and removed 4.7uF from all mosfet drivers.
I tried to drive the motor and everything works like before. Motor is turning and has small amount of torque, but can be easily stopped from what point it doesnt run anymore even with throttle on full.

C5, C6, and C13 are as close as possible to VCC and GND pin..

Any other ideas? :/


Well-Known Member
Most Helpful Member
It's not clear to me how you've got your current sense system set up. But that is a potential problem. If it's spinning at about the right speed, but has no power, it could be starved for power. Which would happen if the current sense function was being tripped.

Time to get the o'scope out and start probing points, and looking at waveforms.
- I'd start on pin 10 of the controller to make sure that it's oscillating at the intended frequency.
- Then check the controller outputs. Use two channels, or four if you have them, to see A to B and B to C timing.
- Look at A low and A high, to learn what they're doing. Withone probe on the controller output pin, use the second probe to successively look at each link in each phase all the way out to the motor terminals.
- Look at timing between HALL A inputs, and each of A low and A high outputs. Then do the same to B and C

If' it's not running right, there's got to be a reason why, You'll just need to find where the problem is. And a two or 4 channel scope is the right tool for that job.

CAUTION: High voltages may be present in the motor control and power circuitry. Use extreme care, and follow good high voltage test principles to prevent injury.
Not open for further replies.

Latest threads

EE World Online Articles