Re: Hiwatt CP103 build
Posted: Fri Jan 11, 2019 11:02 am
I think I may be on the right track now...
It looks as though the preamp is good to go after my previous troubleshooting. The output from the PI looks balanced and decent. The amp is also loud enough, but sounds like crap, so the signal driving the output tube grids seems to be sufficient.
Moving on to the power amp then. As there seemed to be some kind of imbalance in the output, I decided to simplify Things and try to run the amp with just two power tubes, one per side. I set the impedance selector at 8 ohm to adjust reflected primary impedance accordingly. I adjusted the bias so the dissipation reading via the 1 ohm bias reading resistors came to about 40 mA each (plate voltage 480ish with just two power tubes). Put the two tubes in V6 and V9 respectively. Turn amp one with the chassis upside down, so I couldn't see the power tubes well. Amp still sounds like crap. But after just 20-30 seconds of playing, a loud pop and a flash on one of the power tube sockets (V9, on the side that seemed to dissipate a lot). Sound disappears.
Turn amp off. Check for burn marks on sockets and tube, none to be seen. Check fuses, none of them blew. Then I go checking on the grid stop, screen grid and bias reading resistors again. Turns out the 1 ohm bias reading resistor was the one that popped and blew. But strangely enough, one of the resistors on the OTHER side (V7) is now also bad, but not the one where I had the tube on the other side. I had checked them before and they seemed fine.
I check the DC resistance between CT and each primary wire again. Same result as before, about 15-16 ohm on each side.
The power tube on the "hot side" must have been dissipating a lot since the 1 ohm resistor blew. Perhaps we are on to something now - if there's a gross imbalance between power tube dissipation, I would get a lot of crossover distortion, right? And if the bias reading resistors are badly imbalanced or on the brink of going out, that could probably account for it. These represent the ground connection for each tube, and if contact is intermittent or badly imbalanced, that could probably account for some of my problems.
So I yank the resistors out, and replace with a regular ground wire. Check that I have good continuity to ground on pin 8 and 1 on all power tubes. Then I try to test it again with four tubes this time. I don't have a tube tester so I don't know if the power tube in the socket that blew a resistor survived, but I try to put all four tubes from the quad back in. Then I just turned the bias adjustment pots down to maximum negative bias voltage on both sides just for precaution.
Power up and turn amp on, but before I get to play anything, I see a power tube redplating, this time in V8. Turn the amp to standby, and check bias voltage. -42 on V6 and V7, -7 on V8 and V9.
WTF?? This is where I ran out of time before I had to go to work, but now I think we are getting closer to the root of the problem. Maybe some kind of intermittent problem in the bias circuitry? Remember, I have dual bias pots where I set the range for each bias pot with some balancing resistors, so there's actually a lot of potential for things to go wrong. However, at idle I have, up until now, never seen any problems either with reading bias voltage or power dissipation. But perhaps when it begins playing/conducting/vibrating that's when problems show up.
I need to really go over the bias circuitry now. But this has got me thinking about those 1 ohm resistors I have put in all my amps. Are they such a good idea after all? If they go bad, what could happen with the sound? If they are imbalanced, what does that do with the sound. On the other hand, as it looks like in this case, they may also represent a sort of safety valve/fuse function, and burn up in the case of over-dissipation.
Any thoughts from you guys who know much more than I do?
It looks as though the preamp is good to go after my previous troubleshooting. The output from the PI looks balanced and decent. The amp is also loud enough, but sounds like crap, so the signal driving the output tube grids seems to be sufficient.
Moving on to the power amp then. As there seemed to be some kind of imbalance in the output, I decided to simplify Things and try to run the amp with just two power tubes, one per side. I set the impedance selector at 8 ohm to adjust reflected primary impedance accordingly. I adjusted the bias so the dissipation reading via the 1 ohm bias reading resistors came to about 40 mA each (plate voltage 480ish with just two power tubes). Put the two tubes in V6 and V9 respectively. Turn amp one with the chassis upside down, so I couldn't see the power tubes well. Amp still sounds like crap. But after just 20-30 seconds of playing, a loud pop and a flash on one of the power tube sockets (V9, on the side that seemed to dissipate a lot). Sound disappears.
Turn amp off. Check for burn marks on sockets and tube, none to be seen. Check fuses, none of them blew. Then I go checking on the grid stop, screen grid and bias reading resistors again. Turns out the 1 ohm bias reading resistor was the one that popped and blew. But strangely enough, one of the resistors on the OTHER side (V7) is now also bad, but not the one where I had the tube on the other side. I had checked them before and they seemed fine.
I check the DC resistance between CT and each primary wire again. Same result as before, about 15-16 ohm on each side.
The power tube on the "hot side" must have been dissipating a lot since the 1 ohm resistor blew. Perhaps we are on to something now - if there's a gross imbalance between power tube dissipation, I would get a lot of crossover distortion, right? And if the bias reading resistors are badly imbalanced or on the brink of going out, that could probably account for it. These represent the ground connection for each tube, and if contact is intermittent or badly imbalanced, that could probably account for some of my problems.
So I yank the resistors out, and replace with a regular ground wire. Check that I have good continuity to ground on pin 8 and 1 on all power tubes. Then I try to test it again with four tubes this time. I don't have a tube tester so I don't know if the power tube in the socket that blew a resistor survived, but I try to put all four tubes from the quad back in. Then I just turned the bias adjustment pots down to maximum negative bias voltage on both sides just for precaution.
Power up and turn amp on, but before I get to play anything, I see a power tube redplating, this time in V8. Turn the amp to standby, and check bias voltage. -42 on V6 and V7, -7 on V8 and V9.
WTF?? This is where I ran out of time before I had to go to work, but now I think we are getting closer to the root of the problem. Maybe some kind of intermittent problem in the bias circuitry? Remember, I have dual bias pots where I set the range for each bias pot with some balancing resistors, so there's actually a lot of potential for things to go wrong. However, at idle I have, up until now, never seen any problems either with reading bias voltage or power dissipation. But perhaps when it begins playing/conducting/vibrating that's when problems show up.
I need to really go over the bias circuitry now. But this has got me thinking about those 1 ohm resistors I have put in all my amps. Are they such a good idea after all? If they go bad, what could happen with the sound? If they are imbalanced, what does that do with the sound. On the other hand, as it looks like in this case, they may also represent a sort of safety valve/fuse function, and burn up in the case of over-dissipation.
Any thoughts from you guys who know much more than I do?