Hello everyone,
Recently, I received a HammerTech 600.1 class D amplifier with a dead output section. Upon further inspection, I found that two of the buffer transistors were dead as well, so I replaced them with new working parts. Two marked resistors also appeared to have been drawing a lot of current, so I replaced them too.
Everything now checks out okay in and out of the circuit, but there is no oscillation on the outputs, and the amplifier turns on normally without going into protection.
Also removing those two marked zeners will result in some oscillation but it's not fully forming the pwm signal and sounds very distorted.
Can anyone be kind enough to help me find out what the problem is?
Thanks in advance.
Recently, I received a HammerTech 600.1 class D amplifier with a dead output section. Upon further inspection, I found that two of the buffer transistors were dead as well, so I replaced them with new working parts. Two marked resistors also appeared to have been drawing a lot of current, so I replaced them too.
Everything now checks out okay in and out of the circuit, but there is no oscillation on the outputs, and the amplifier turns on normally without going into protection.
Also removing those two marked zeners will result in some oscillation but it's not fully forming the pwm signal and sounds very distorted.
Can anyone be kind enough to help me find out what the problem is?
Thanks in advance.
Dead? Open, shorted, leaking?
Did you use the EXACT same part number replacement transistor?
Resistor drawing current? Explain.
What are the part numbers on the Zeners that you remove and get partial oscillation?
What is the IC in the socket?
Did you replace it and install the socket or was it like that when you received the amp?
What are the output transistors in the amp?
Did you use the EXACT same part number replacement transistor?
Resistor drawing current? Explain.
What are the part numbers on the Zeners that you remove and get partial oscillation?
What is the IC in the socket?
Did you replace it and install the socket or was it like that when you received the amp?
What are the output transistors in the amp?
Thanks for replying
Sorry for the delay 🙏
Yes I used exact parts.
Those resistors were burnt as it seems they were running so hot but the value of them was normal.
The zeners were 1N4148
The op-amp is NE5532.
I was trying to experiment with different op-amps so to avoid damaging the board I installed the socket temporarily.
The transistors used here are tagged in the following attached picture.
So far I replaced the tagged parts on the board and now it's oscillating at low frequency with alot of spikes. Some of the resistor were burnt with no signs of burning and replacing them caused the output to oscillate again. Replacing the 500ohm with 560 also didn't affect the output oscillation.
The output of the op-amp to the driver circuit was pretty noisy so I replaced the capacitor on the feedback with lower value and the noise was gone but didn't changed anything on the outputs.
The capacitors in the feedback circuit often determine the frequency. I'd go back tot he original value.
If you have the original op-amp and it's not definitively defective, I'd reinstall it.
1N4148s are not Zeners. They are switching diodes.
What's the DCV across R63 (1.2k?)?
What's the DCV across D9?
If you have the original op-amp and it's not definitively defective, I'd reinstall it.
1N4148s are not Zeners. They are switching diodes.
What's the DCV across R63 (1.2k?)?
What's the DCV across D9?
I reinstalled the original Op-Amp and now its oscillating at around 70 kHz but still with a lot of spikes presence.
Across R63 I can read 7.55 and for the D9 is 8.16 DCV
Before me, someone had worked on this board, and the D9 was replaced with a 1N4738. I thought it was faulty (leaking current) and replaced it with the same part number. Maybe it's the wrong part.
The R63 reads fine and is within the specs.
They were IRF640Ns and I used the same part.
Last edited:
It's possible that it's the original part but I've only seen in one other amp. You can leave it for now.
What were the original output transistors?
What output transistors did you use here?
Are you limiting the current to the amp or are you letting it have the current it needs to function?
Are the output transistors heating up?
What were the original output transistors?
What output transistors did you use here?
Are you limiting the current to the amp or are you letting it have the current it needs to function?
Are the output transistors heating up?
The original output transistors were IRF640Ns, and I used the same part number for them.
No, I'm not limiting the current.
Yes, they are heating up too much, and without the heatsink, they will die in under one minute while sitting idle.
Do you have a 6.2v Zener? 6v? 5.6v?
If so, try those in place of D9.
Do the outputs heat up with those?
Does the amplifier produce clean audio?
If so, try those in place of D9.
Do the outputs heat up with those?
Does the amplifier produce clean audio?
Currently, I don't have them, but I will try them soon and report back.
Thank you.
What is the rail voltage measured across the rectifiers with a multimeter?
You can try increasing the value of R63 instead of replacing the diode.
You can try increasing the value of R63 instead of replacing the diode.
Last edited:
Go up to the next value you have 1.5k? It's not too critical at this point. You're just doing this to see if the current draw drops. You will fine-tune later.
There may be better options for the outputs if the total voltage across the rails is less than 100v. No need to change them now.
There may be better options for the outputs if the total voltage across the rails is less than 100v. No need to change them now.
I tried 1.5, 2.2, 3.3 and 3.8(3.3+0.5)
Anything more than 3.5 it won't oscillate
With 2.2 the wave form is better but the spike at the top is more pronounced, with 3.3 that spike is minimum but at the bottom a step will appear.
So I did use a 2.2k resistor and put it back together on the heatsink. Yes, it can produce acceptable audio driving a subwoofer.
What would be a better choice instead of the 640s?
The first thing you need to do is to make sure that the rail voltage cannot go above ±50v. Adjust the DC output of your power supply to at least 14v to see how high the rail voltage goes. If nothing over that, something like the IRF3710Z may be better. If the rail voltage goes above ±50v, something like the IRF3415 may be better.
By better, I mean more rugged. If the amp won't be abused, the 640s may be OK. The 640Ns may be somewhat better.
For the 'right' resistor, when you find the FETs that you will leave in the amp, you would use a potentiometer to adjust the resistance and adjust for the value that just makes the amp start to draw current. Start with the pot reading approximately and adjust carefully from there.
This is much like setting the bias on a class AB amplifier. Bear in mind that the amp must be oscillating when finding the value. If it doesn't want to oscillate, drive a low-level signal into the amp (no speaker/dummy load) to encourage it to oscillate.
By better, I mean more rugged. If the amp won't be abused, the 640s may be OK. The 640Ns may be somewhat better.
For the 'right' resistor, when you find the FETs that you will leave in the amp, you would use a potentiometer to adjust the resistance and adjust for the value that just makes the amp start to draw current. Start with the pot reading approximately and adjust carefully from there.
This is much like setting the bias on a class AB amplifier. Bear in mind that the amp must be oscillating when finding the value. If it doesn't want to oscillate, drive a low-level signal into the amp (no speaker/dummy load) to encourage it to oscillate.