Amp had all four 75339 power supply FETs blown
Replaced with 3205's.Changed gate resistors to 47R and amp starts and produces rail voltages.
Noticed the power FETs getting hot very quickly.Amp had 69R gate resistors so I installed 68R to see what would happen.No change.
The drive signal seems okay compared to photos of other MTX amps.Swinging about 30v peak to peak.
Is there changes that need to be made to get the 3205s to run cooler?Or perhaps I'm missing something being the first time working on an mtx amp.
Replaced with 3205's.Changed gate resistors to 47R and amp starts and produces rail voltages.
Noticed the power FETs getting hot very quickly.Amp had 69R gate resistors so I installed 68R to see what would happen.No change.
The drive signal seems okay compared to photos of other MTX amps.Swinging about 30v peak to peak.
Is there changes that need to be made to get the 3205s to run cooler?Or perhaps I'm missing something being the first time working on an mtx amp.
It's very odd that the FETs are heating up. If you look at the waveforms, you can see that the drive signal for one bank is at -3v when the other bank is just switching on at about +3v. What is the frequency on the gates of the FETs?
How much current is the amp drawing at idle?
With the FETs clamped to the heatsink, does the heatsink neat the FETs get hot or just barely warm?
Is the gate signal swinging at least 4v above the source leg?
With the FETs clamped to the heatsink, does the heatsink neat the FETs get hot or just barely warm?
Is the gate signal swinging at least 4v above the source leg?
About 1.5 amps
There not clamped to sink.Besides I always check power supply FETs outside of sink and they never get hot.I read an old thread with you stating,if they can go 30 seconds without getting hot outside of sink there probably okay.
They get very hot inside of 10 seconds.I could give it a test run and see what happens but I've never seen power FETs get this hot in any other amp in good working condition with no load.
Yes,there at least 3 volts above source.I'll post source photo.Same scope settings.
There not clamped to sink.Besides I always check power supply FETs outside of sink and they never get hot.I read an old thread with you stating,if they can go 30 seconds without getting hot outside of sink there probably okay.
They get very hot inside of 10 seconds.I could give it a test run and see what happens but I've never seen power FETs get this hot in any other amp in good working condition with no load.
Yes,there at least 3 volts above source.I'll post source photo.Same scope settings.
Well,clamped it to the sink and let it idle for 20 minutes.The sink got pretty hot from just idling.
Went ahead and ran it into a dummy load at full power for 20 minutes and it didn't die.
Perhaps it'll be fine.
Went ahead and ran it into a dummy load at full power for 20 minutes and it didn't die.
Perhaps it'll be fine.
I've probably never replaced more than a few sets of FETs in these amps so I've never really had a reason to monitor their temperature. I have a junker 4501 that I powered up and the original FETs get hot.
The drive circuit works much like a pulse amplitude modulation drive circuit (like Orion amps use). This is why the FETs are getting hot at low power. The amplitude of the drive signal drops to the threshold of the FETs so the FETs are not switching efficiently. The efficiency is low because the FETs are not fully on or fully off like they would be in most switching circuits.
When the 12v DC supply voltage is relatively high and the target rail voltage can be easily maintained, the amplitude of the drive signal is the lowest (approximately 2v with the original HUF75339 - this minimal amplitude may be different with other FETs). This causes the FETs to heat quickly (~0.1C/second at ~40C).
When the 12v DC supply voltage is lower (~10v) and the amp has to increase the drive to maintain the target rail voltage, the drive signal increases to about 5v (6v peak amplitude oddly shaped waveform, not sine or square waveform). With this drive amplitude, the FETs were still heating up but only about 0.1C every 5 seconds at ~40C.
It appears that the heating of the power supply FETs is normal for these amps.
During testing, the scope was connected in differential mode so that all I was seeing was the voltage Vgs and not the large swing seen on the gate and source terminals.
The drive circuit works much like a pulse amplitude modulation drive circuit (like Orion amps use). This is why the FETs are getting hot at low power. The amplitude of the drive signal drops to the threshold of the FETs so the FETs are not switching efficiently. The efficiency is low because the FETs are not fully on or fully off like they would be in most switching circuits.
When the 12v DC supply voltage is relatively high and the target rail voltage can be easily maintained, the amplitude of the drive signal is the lowest (approximately 2v with the original HUF75339 - this minimal amplitude may be different with other FETs). This causes the FETs to heat quickly (~0.1C/second at ~40C).
When the 12v DC supply voltage is lower (~10v) and the amp has to increase the drive to maintain the target rail voltage, the drive signal increases to about 5v (6v peak amplitude oddly shaped waveform, not sine or square waveform). With this drive amplitude, the FETs were still heating up but only about 0.1C every 5 seconds at ~40C.
It appears that the heating of the power supply FETs is normal for these amps.
During testing, the scope was connected in differential mode so that all I was seeing was the voltage Vgs and not the large swing seen on the gate and source terminals.
Thanks for the lengthy reply.After testing the amp at full power and had no issues I assumed that was normal.Thanks for confirming it.
- Status
- Not open for further replies.