Hello folks,
My amp has intermittent 120Hz buzz, it happens on both channels but not at the same time, and it happens very often when the amp is powered on when cold; it usually subsides as the amp gets to its normal temperature. It was worse a couple years ago, I changed power supply caps for the regulator section that were closest to the power transistors, suspecting heat-induced reduced lifetime. It fixed the problem for a while but it slowly returned. The amp is almost 35 years old so I'm pretty sure all electrolytics will have to be changed.
I live in a rural area that has pretty crappy utility service, probably old transformers now undersized for their load, and there's a lot of trees so we often get very short losses of power (less than a second), probably due to tree branches shorting the wires momentarily - It happens mostly when it's windy. Sometimes there are two of them within a couple minutes. Yes, the power company is pretty lousy on network maintenance... 😡
These can't be good on my amp that has no "zero-crossing" switching like many modern equipment. After the last one of these there was a louder than usual buzz on the left channel, a bit on the right, and it doesn't subsides - So on the bench the amp went. Tracking the buzz I find that there seems to be a problem on the negative rail of the regulated supply. Positive rail is fine, so the shared voltage reference is ok. The ripple on the negative rail is higher than the ripple on the corresponding input caps, so I directed my attention to the final caps. The very lower buzz on the right channel probably suffers from the resulting noisy ground.
I was preparing to add a new cap in parallel temporarily to test my theory but then I thought I should set up a "dim bulb" current limiter. I used a 100W bulb to test the setup first without my cap mod, there was an expected bulb glow due to inrush current, but after the main caps charged the amp stabilized and the buzz was gone! Hooked the O-scope and everything seems normal, the voltages are what they're supposed to be, played music and sounds ok... However can't crank the power up because the bulb gradually kick in.
How can added resistance on the mains input cure a buzz? The hot line goes from IEC jack to a 6A fuse, power switch and splits to the two toroid transformers - That's it. I can only think of an oscillation occurring, it may be still be related to caps aging with higher ESR, but I'm at a loss. What do you make of this? Should I try insert a NTC thermistor at the mains input?
Thanks,
Joris
My amp has intermittent 120Hz buzz, it happens on both channels but not at the same time, and it happens very often when the amp is powered on when cold; it usually subsides as the amp gets to its normal temperature. It was worse a couple years ago, I changed power supply caps for the regulator section that were closest to the power transistors, suspecting heat-induced reduced lifetime. It fixed the problem for a while but it slowly returned. The amp is almost 35 years old so I'm pretty sure all electrolytics will have to be changed.
I live in a rural area that has pretty crappy utility service, probably old transformers now undersized for their load, and there's a lot of trees so we often get very short losses of power (less than a second), probably due to tree branches shorting the wires momentarily - It happens mostly when it's windy. Sometimes there are two of them within a couple minutes. Yes, the power company is pretty lousy on network maintenance... 😡
These can't be good on my amp that has no "zero-crossing" switching like many modern equipment. After the last one of these there was a louder than usual buzz on the left channel, a bit on the right, and it doesn't subsides - So on the bench the amp went. Tracking the buzz I find that there seems to be a problem on the negative rail of the regulated supply. Positive rail is fine, so the shared voltage reference is ok. The ripple on the negative rail is higher than the ripple on the corresponding input caps, so I directed my attention to the final caps. The very lower buzz on the right channel probably suffers from the resulting noisy ground.
I was preparing to add a new cap in parallel temporarily to test my theory but then I thought I should set up a "dim bulb" current limiter. I used a 100W bulb to test the setup first without my cap mod, there was an expected bulb glow due to inrush current, but after the main caps charged the amp stabilized and the buzz was gone! Hooked the O-scope and everything seems normal, the voltages are what they're supposed to be, played music and sounds ok... However can't crank the power up because the bulb gradually kick in.
How can added resistance on the mains input cure a buzz? The hot line goes from IEC jack to a 6A fuse, power switch and splits to the two toroid transformers - That's it. I can only think of an oscillation occurring, it may be still be related to caps aging with higher ESR, but I'm at a loss. What do you make of this? Should I try insert a NTC thermistor at the mains input?
Thanks,
Joris
@rayma: In his case inserting the dim bulb tester, thereby, reducing the AC mains voltage to the amp eliminates the hub. Had the bulb made the hum worse I'd agreed with your theory.
Are there any electrolytic decoupling capacitors that could have gone soft?
Tom
I would repair rather than modify.
Are there any electrolytic decoupling capacitors that could have gone soft?
Tom
Do you have a Variac? If lowering the AC line voltage eliminates the hum, you can use that to check further.
Maybe a component is breaking down at higher AC line voltages. Is your nominal AC line too high?
Often there can be rather high AC line voltage overshoot after a dropout, which could do damage.
Maybe a component is breaking down at higher AC line voltages. Is your nominal AC line too high?
Often there can be rather high AC line voltage overshoot after a dropout, which could do damage.
This probably is a factor, because the utility delivery norms are horrible here... 120V minus 10% PLUS 15% ! ! That's why I replace the caps with a higher voltage spec where I can. But in this case the caps are after a series pass regulator, they should not have been exposed to a higher than normal voltage. But I do have a variac, that was next test.
Thanks!
Last edited:
If the AC is too high, the transformer might saturate and buzz. A bucking transformer would solve this
Running on the variac, the buzz starts over 110V, so yes the dim bulb "cured" the buzz because of the potential divider it forms with the amp... That's well below the rated 120V voltage, I don't think transformer saturation is the cause here... Brings me back to the failing caps theory. They certainly have it rough with the AC line voltage overshoot and dropouts,
Just a thought. Look for negative rail noise injection on the series regs. Maybe caps/ components are getting over V in that section.
Interesting. Perhaps check the reverse protection flywheel diode on the regulator.
Indeed I was heading in that direction... Thanks for your reply
If you get a UPS, it should be a true sine wave type. But most UPS will pass AC line noise through to their output.
If the noise is simply due to low or high AC line, the UPS should work.
But if the noise is due to an external source, the UPS may not work at all to eliminate the noise.
But the UPS may still be necessary, to better regulate the input voltage.
If the noise is simply due to low or high AC line, the UPS should work.
But if the noise is due to an external source, the UPS may not work at all to eliminate the noise.
But the UPS may still be necessary, to better regulate the input voltage.
Sine wave I agree, but his issue seems to be twofold.
1. Series regs should do a reasonable job of cleaning up the noise.
2. His line V is unstable
A quality sine UPS should cure No 2 and No 1 should be resolved if the regs are working properly
1. Series regs should do a reasonable job of cleaning up the noise.
2. His line V is unstable
A quality sine UPS should cure No 2 and No 1 should be resolved if the regs are working properly
Would a Tripplite Product be applicable here? I used to live a similar environment in the mountains of AZ..... Or even an older PS Audio P300? These both of course are different than a UPS.
Burned out a lot of light bulbs. Then found 130 volt bulbs and that helped some.
Burned out a lot of light bulbs. Then found 130 volt bulbs and that helped some.
He should regulate the AC line first, and then do further testing on the noise, which may or may not be improved.
Or get audio equipment with universal AC line input, 95VAC - 240VAC.
I should have clarified. The Tripplite that I am referring to involves a transformer with multiple taps, that, when a lower or higher voltage is sensed, a new tap of the transformer is connected. FWIW, the one that I owned seemed to be constantly sensing voltage differences! Clickety click click went the relays.
A tap switcher is meant for long term line sag or rise, and won't help with short term line fluctuations
like he has. An actual AC line regulator is needed.
like he has. An actual AC line regulator is needed.
