I purchased a Son of Ampzilla recently off eBay. It was functional with one blown meter and a hum when first powered that would come and go somewhat as the amp warmed up.
I recapped the amp using Panasonic FM caps where possible, and used power supply filter caps of larger value than the originals as I wanted something close to the same physical size. I set the bias to 30 mv across the power resistors of the output stage as described in some other articles here. As the voltage varied a bit across the 4 resistors, I picked the one with the highest voltage to do the setting. I set the offset trimmers to 0 volts on the speaker outputs as well. I rechecked this a couple of times as the amp warmed.
The weird hum that would change as the amp warmed is now gone, replaced by a much quieter hum that does not change as the amp warms. Sonically, the amp is much improved, in fact I am struck by how good it sounds compared to the Audiosource Amp One I was using in its stead.
One very weird problem remains: when I turned off my preamp (a GAS Thalia) without powering down the Son, a pulse occurred which caused one of the meters to stay up. It stayed up until I unplugged the Son from AC power. When I plugged it back in it was fine. I noticed, however, that I could create the same problem by playing the Son very loud and turning up the bass, at which point one channel will "clamp": the VU meter will go up and stay steady and the volume on that channel will drop way down and get distorted. It remains this way until I unplug AC power again. When I repower it is ok again.
Any idea of what might case this? Is it the VI limiter circuit or something else going bad?
Any help would be greatly appreciated.
--Paul R.
I recapped the amp using Panasonic FM caps where possible, and used power supply filter caps of larger value than the originals as I wanted something close to the same physical size. I set the bias to 30 mv across the power resistors of the output stage as described in some other articles here. As the voltage varied a bit across the 4 resistors, I picked the one with the highest voltage to do the setting. I set the offset trimmers to 0 volts on the speaker outputs as well. I rechecked this a couple of times as the amp warmed.
The weird hum that would change as the amp warmed is now gone, replaced by a much quieter hum that does not change as the amp warms. Sonically, the amp is much improved, in fact I am struck by how good it sounds compared to the Audiosource Amp One I was using in its stead.
One very weird problem remains: when I turned off my preamp (a GAS Thalia) without powering down the Son, a pulse occurred which caused one of the meters to stay up. It stayed up until I unplugged the Son from AC power. When I plugged it back in it was fine. I noticed, however, that I could create the same problem by playing the Son very loud and turning up the bass, at which point one channel will "clamp": the VU meter will go up and stay steady and the volume on that channel will drop way down and get distorted. It remains this way until I unplug AC power again. When I repower it is ok again.
Any idea of what might case this? Is it the VI limiter circuit or something else going bad?
Any help would be greatly appreciated.
--Paul R.
For one the circuit that drives the VU is extremely generic plus that has absolutely no protection against any type of failure or overdrive
Add to that the visual result that is not accurate and motion is too quick ...been modifying some circuits in the past with better results but i didnt keep a note about it
Obviously your VU has a mechanical damage and this will not be a circuit issue .
behavior of thalia is expected to be like that still excessive DC or thumb might indicate capacitor failure inside the Thalia which is also very common to equipment of this age
Kind regards
Sakis
Add to that the visual result that is not accurate and motion is too quick ...been modifying some circuits in the past with better results but i didnt keep a note about it
Obviously your VU has a mechanical damage and this will not be a circuit issue .
behavior of thalia is expected to be like that still excessive DC or thumb might indicate capacitor failure inside the Thalia which is also very common to equipment of this age
Kind regards
Sakis
Thanks for the reply, however I believe you have misunderstood the issue. The Thalia has been used to drive other amplifiers (both an Audiosource Amp One and a Tigersaurus) with absolutely no issue.
The VU meter hanging on the Son of Ampzilla is also not the main issue. There is no mechanical problem with the VU meters. The issue is that when I play the Son very loudly, one channel suddenly goes almost completely out; you can still hear sound through it but very distorted. The amp remains this way even if I turn the volume back down. The only way to "reset" it and get clear undistorted full gain back is to unplug it from the AC outlet and plug it back in again.
The VU meter hanging on the Son of Ampzilla is also not the main issue. There is no mechanical problem with the VU meters. The issue is that when I play the Son very loudly, one channel suddenly goes almost completely out; you can still hear sound through it but very distorted. The amp remains this way even if I turn the volume back down. The only way to "reset" it and get clear undistorted full gain back is to unplug it from the AC outlet and plug it back in again.
Yes, the one channel is definitely latching up to one of the rails. I turned the four VI limiter pots to maximum resistance and there is no difference in behavior.
All I have to do to make this happen is to play the amp loud enough, especially with the bass up. It will start to distort, the channel B meter will pin, and Channel B will drop volume drastically and hum. If I unplug the amp from AC, turn the volume down on the preamp, and plug the amp back in it will play normally.
I disconnected the wires that feed Channel B from the power supply just so I could really crank Channel A. I can turn the volume up as far as I can stand it with just Channel A connected and there is no weird behavior. With this in mind, I will check Channel B closely to make sure I didn't install any capacitors backward as I do not recall the amp acting like this before the recap job.
With just Channel A connected I also do not get a bad turn off pulse if the amp is powered and I turn the preamp off. With Channel B connected I get a big speaker pop from Channel B when doing this, and it will latch to the rail requiring me to unplug the amp from AC.
I did notice one other weird thing (unrelated) - my meter circuit board has no ground. The diagram shows a ground connection from the meter diode bridges to the power supply board but I do not recall one during disassembly and reassembly. The meters don't move without the ground - if I connect a jumper they work. Anyone know how the ground is supposed to connect?
All I have to do to make this happen is to play the amp loud enough, especially with the bass up. It will start to distort, the channel B meter will pin, and Channel B will drop volume drastically and hum. If I unplug the amp from AC, turn the volume down on the preamp, and plug the amp back in it will play normally.
I disconnected the wires that feed Channel B from the power supply just so I could really crank Channel A. I can turn the volume up as far as I can stand it with just Channel A connected and there is no weird behavior. With this in mind, I will check Channel B closely to make sure I didn't install any capacitors backward as I do not recall the amp acting like this before the recap job.
With just Channel A connected I also do not get a bad turn off pulse if the amp is powered and I turn the preamp off. With Channel B connected I get a big speaker pop from Channel B when doing this, and it will latch to the rail requiring me to unplug the amp from AC.
I did notice one other weird thing (unrelated) - my meter circuit board has no ground. The diagram shows a ground connection from the meter diode bridges to the power supply board but I do not recall one during disassembly and reassembly. The meters don't move without the ground - if I connect a jumper they work. Anyone know how the ground is supposed to connect?
Missing mounting screws = missing ground. If you had your power supply board out you noticed some nylon spacers at each of the four corners those same size spacers go between the chassis and the meter PCB. You'll need two each #6-32 x 3/4" panhead screws, 4 each #6 internal tooth lockwashers, and two each #6 x 1/4" nylon spacers. That should fix that problem.
Q103 and Q104, right above the connector for the output stage, are the VI transistors. They might be MPSU06,07 and MPSU56,57.
Craig
Q103 and Q104, right above the connector for the output stage, are the VI transistors. They might be MPSU06,07 and MPSU56,57.
Craig
Someone apparently altered the way my VU meter board mounts as it is just stuck to the back of the meters with black RTV. There are no screws or spacers. I will change that.
Removed Q103 and Q104 and this made no difference. I can still get Channel B to latch to one of the rails by cranking the bass or by turning off the preamp with the Son still powered. I can avoid doing the turnoff but can't avoid cranking the bass.
Removed Q103 and Q104 and this made no difference. I can still get Channel B to latch to one of the rails by cranking the bass or by turning off the preamp with the Son still powered. I can avoid doing the turnoff but can't avoid cranking the bass.
Q101, 102, 105 and 106 are respectively GAS116, F104A, GAS117 and F103A.
I did hook a signal generator up to the input with a scope and AC voltmeter on the speaker output. Test frequency was 500 Hz. As I cranked up the input, I could get about 12 volts output, then the signal suddenly disappeared on the output. I checked aging measuring the DC on the output and all is well until it goes crazy and suddenly latches to +20 volts. As I am doing this, the amount of input necessary to make it latch is getting smaller - now I cannot run it for even 30 seconds at a low input signal level before it bonks out. Nothing is getting hot btw.
I did crank the scope timing down after it goes awry and note there is sinusoidal signal on the output of about 200khz, so the thing is oscillating when this happens, but the entire output signal is well above 0 volts. The 200khz signal slowly dies in amplitude until it disappears.
Changing the bias adjustment has not effect on this behavior.
I did hook a signal generator up to the input with a scope and AC voltmeter on the speaker output. Test frequency was 500 Hz. As I cranked up the input, I could get about 12 volts output, then the signal suddenly disappeared on the output. I checked aging measuring the DC on the output and all is well until it goes crazy and suddenly latches to +20 volts. As I am doing this, the amount of input necessary to make it latch is getting smaller - now I cannot run it for even 30 seconds at a low input signal level before it bonks out. Nothing is getting hot btw.
I did crank the scope timing down after it goes awry and note there is sinusoidal signal on the output of about 200khz, so the thing is oscillating when this happens, but the entire output signal is well above 0 volts. The 200khz signal slowly dies in amplitude until it disappears.
Changing the bias adjustment has not effect on this behavior.
I wonder why only 20 Volts? Are your 28 Volt power supplies OK? Should have + and - 28VDC at resistors R104 and R125, 52VDC on one side and 28VDC on the other side. Offset adjustment working?
Do the GAS116/117 have an "F" on them? IIRC those are the Fairchilds that are the ticking time bomb xsistors. Do you have a transistor analyzer? You could have a leaky one. You said you were going to check for backwards capacitors, all OK? If nothing else shows up remove both boards and put them side by side and compare part for part.
In all of my GAS pieces that had the Fairchilds at least one channel had been repaired before and always transistors where the Fairchilds would have been. I have not seen the symptoms of the bad ones so I'm just guessing at this point. I know my SOA had a problem/repair in that area when I bought it years ago. My latest Ampzilla IIa same thing. Now when I get a piece I just strip it and start over, never have any problems after that.
Craig
Do the GAS116/117 have an "F" on them? IIRC those are the Fairchilds that are the ticking time bomb xsistors. Do you have a transistor analyzer? You could have a leaky one. You said you were going to check for backwards capacitors, all OK? If nothing else shows up remove both boards and put them side by side and compare part for part.
In all of my GAS pieces that had the Fairchilds at least one channel had been repaired before and always transistors where the Fairchilds would have been. I have not seen the symptoms of the bad ones so I'm just guessing at this point. I know my SOA had a problem/repair in that area when I bought it years ago. My latest Ampzilla IIa same thing. Now when I get a piece I just strip it and start over, never have any problems after that.
Craig
Not sure why only 20 volts - I will check the power supply, but like I said the other channel works flawlessly. All capacitors are installed correctly.
The guy who sold it to me said it had never been worked on which is obviously not true given the meter board placement. Also, the right meter was blown suggesting this problem existed for the previous owner as well.
The guy who sold it to me said it had never been worked on which is obviously not true given the meter board placement. Also, the right meter was blown suggesting this problem existed for the previous owner as well.
Each board has it's own +/- 28VDC power supply, in the upper left corner as you look at the board. Some have two zener diodes, TO92 case and other version has only one zener diode I think. Are there two 220uf capacitors in the upper left side just to the right of the VI limiter pot? If so that's the two zener version. One channel could still be good and the other bad, it's NOT the main supply.
Craig
Craig
The +/- 28 volt power supply on the board tests good. This board has two zeners. I did remove the four input and four TO220 transistors and checked with a transistor tester. I get no leakage current on any but a wide variety of betas, ranging from 28 to 125 on the TO220s and from 85 to 250 on the input transistors. Is this important at all?
Also, I did not realize that the ground for the input stage came from the chassis ground point of the RCA input jack. I had disconnected the twister pair during testing and so had no real ground on the input stage. This did make the unit much more sensitive to going into its lockup mode. Once I grounded it I was able to take the gain up a lot more before the problem occurred - it is still there and it is an "all of a sudden" thing, very non-linear. I looked around closely for bad solder connections and component heat damage etc. and that all looks good. I would like to pinpoint exactly what component is giving out as my gut feeling is I will replace a bunch of transistors and still have the problem.
Also, I did not realize that the ground for the input stage came from the chassis ground point of the RCA input jack. I had disconnected the twister pair during testing and so had no real ground on the input stage. This did make the unit much more sensitive to going into its lockup mode. Once I grounded it I was able to take the gain up a lot more before the problem occurred - it is still there and it is an "all of a sudden" thing, very non-linear. I looked around closely for bad solder connections and component heat damage etc. and that all looks good. I would like to pinpoint exactly what component is giving out as my gut feeling is I will replace a bunch of transistors and still have the problem.
Last edited:
- Status
- This old topic is closed. If you want to reopen this topic, contact a moderator using the "Report Post" button.