I have a problem with a self oscillating IRS20957 circuit.
It will not startup and I'm getting an use DC voltage at the output.
The DC voltage is coming from the high-side OCP (from the VDR/VCC voltage) so is seems like that the IRS20957 is not starting up properly.
Unfortunately I'm not allowed to post the schematic here, but I is partly similar to IRS2092 2kW dev board.
(or at least the OCP etc)
At first I tried the values that I calculated for the protection circuits, after that I switched to the 1-2kW IRS2092 version (which is very similar)
The strange thing is that the amplifier was working, but was constantly switching on and off (oscillating frequency was working)
After removing the self reset capacitor the amplifier was staying on, but I got a very strange PWM signal.
When I applied a sine wave the circuit was switching off suddenly.
I already set the dead-time to highest time value (otherwise my amplifier is getting really hot).
The voltage supply is around ±70Vdc (at the moment for testing, will be ±85Vdc), I use four IRFP4227 MOSFETs and FMMT718/716 as a buffer
It will not startup and I'm getting an use DC voltage at the output.
The DC voltage is coming from the high-side OCP (from the VDR/VCC voltage) so is seems like that the IRS20957 is not starting up properly.
Unfortunately I'm not allowed to post the schematic here, but I is partly similar to IRS2092 2kW dev board.
(or at least the OCP etc)
At first I tried the values that I calculated for the protection circuits, after that I switched to the 1-2kW IRS2092 version (which is very similar)
The strange thing is that the amplifier was working, but was constantly switching on and off (oscillating frequency was working)
After removing the self reset capacitor the amplifier was staying on, but I got a very strange PWM signal.
When I applied a sine wave the circuit was switching off suddenly.
I already set the dead-time to highest time value (otherwise my amplifier is getting really hot).
The voltage supply is around ±70Vdc (at the moment for testing, will be ±85Vdc), I use four IRFP4227 MOSFETs and FMMT718/716 as a buffer
Well, I think it is just soldering issues (you were talking about).
Today I re soldered a new PCB, and guess what, I'm getting the right oscillation frequency and signal. 🙂😎
Unfortunately with a lot of switching noise, but maybe that's because I only connected the essential parts. 😛
Take care, it could be a tolerance issue.
Usually proper starting depends on having the right preconditions at the comparator input ==> means ensuring that the amp wants to turn ON the low side switch first.
Or if this is not appreciated for audisoteric reasons one has to ensure that also the high side gate drive supply is available without having turned ON the low side before.
Another reason could be this (posted by jlester87 in the 25-1200W thread):
http://www.irf.com/technical-info/designtp/dt99-7.pdf
Usually proper starting depends on having the right preconditions at the comparator input ==> means ensuring that the amp wants to turn ON the low side switch first.
Or if this is not appreciated for audisoteric reasons one has to ensure that also the high side gate drive supply is available without having turned ON the low side before.
Another reason could be this (posted by jlester87 in the 25-1200W thread):
http://www.irf.com/technical-info/designtp/dt99-7.pdf
Mmm, that's very interesting.
I never read any problems that people had with a Class-D amplifier.
I will read the paper more closely next week, it's not really clear to me at the moment
Ok, this morning the amplifier was working fine.
I had some problems with switching noise.
After adding some 100nF caps to ground/earth (because I was working on an almost empty PCB board), the sine wave was nice and clear
Unfortunately, after power down (had to solder some extra caps) en firing it up again, the amplifier keeps in some kind of reset mode.
When I measure the CSD pin, I get some sort of triangle wave of 2Vp-p (charge-discharge).
All tests were without load, so it seems like the problem they talk about in Design Tip 99-7 ??
I use a BJT buffer before the MOSFETs, so maybe they broke down and therefore the IRS keeps resetting?
I had some problems with switching noise.
After adding some 100nF caps to ground/earth (because I was working on an almost empty PCB board), the sine wave was nice and clear
Unfortunately, after power down (had to solder some extra caps) en firing it up again, the amplifier keeps in some kind of reset mode.
When I measure the CSD pin, I get some sort of triangle wave of 2Vp-p (charge-discharge).
All tests were without load, so it seems like the problem they talk about in Design Tip 99-7 ??
I use a BJT buffer before the MOSFETs, so maybe they broke down and therefore the IRS keeps resetting?
- Status
- Not open for further replies.