Thanks Bas,
Yes that makes sense to me now.
Although I haven't found the exact issue, given my voltages are ok following replacement of a number of components, should I be relatively confident?
In the event it still isn't solved, do you think I would then be down to a defective cathode bypass or coupling cap (even one of the new ones I have installed) or an issue at the output transformer?
Many thanks.
Yes that makes sense to me now.
Although I haven't found the exact issue, given my voltages are ok following replacement of a number of components, should I be relatively confident?
In the event it still isn't solved, do you think I would then be down to a defective cathode bypass or coupling cap (even one of the new ones I have installed) or an issue at the output transformer?
Many thanks.
Yes and no. Everything seems in order...that always makes me nervous.
Good one. You'd hate it to be the cause of your problems. But it is very possible. Are the secondaries wired up consistently wired for instance? Perhaps one 8 ohm tap is wired to the 4 ohm tap. If all is wired up correctly. You could measure the resistance of the primaries and secondaries of each channel and compare.
They could indeed be the culprits. Seeing that the voltages are equal on both sides..it seems more likely that a coupling cap is the culprit.
Last edited:
Thanks again, this is helpful as it gives me more things to test.
I hadn't thought about checking the primaries and secondaries.
So in terms of checking the caps (without removing them first)
I should be Checking the voltage either side which should be the same?
Are there any other tests that can be done on the caps while in circuit other than checking for a short? Thanks.
I hadn't thought about checking the primaries and secondaries.
So in terms of checking the caps (without removing them first)
I should be Checking the voltage either side which should be the same?
Are there any other tests that can be done on the caps while in circuit other than checking for a short? Thanks.
New 0.47 coupling caps arrived today Jupiter Copper Foil Paper and wax so I put these into the amp in place of the old ones that I had installed to test the amp voltages yesterday. I then put the amp into my system.
After switch on and the two minute delay where the amp pre-warms the tubes to prevent thermal shock once the amp kicked in I had a horrendously loud buzzing like a motorbike engine getting louder and higher in pitch coming from the speaker and quickly hit the off switch, it was certainly in danger of damaging the speaker I am sure. The volume pot I am fairly sure was at zero shorted to ground.
I took the amp back to my desk and after a short while checked everything over then powered it up again and the same thing happened without the speakers attached, but with a wirewound resistor load at the outputs. I could hear a transformer vibrating violently getting louder and higher in picth so again I killed the power to the amp.
I cannot understand what is going on or how it has happened from simply changing the caps. Nothing else in the amp has changed and all the four joins that were made to put these 0.47 coupling caps in look and test fine.
If anybody could chime in with a possible suggestion I would be very grateful.
Many thanks in advance!
After switch on and the two minute delay where the amp pre-warms the tubes to prevent thermal shock once the amp kicked in I had a horrendously loud buzzing like a motorbike engine getting louder and higher in pitch coming from the speaker and quickly hit the off switch, it was certainly in danger of damaging the speaker I am sure. The volume pot I am fairly sure was at zero shorted to ground.
I took the amp back to my desk and after a short while checked everything over then powered it up again and the same thing happened without the speakers attached, but with a wirewound resistor load at the outputs. I could hear a transformer vibrating violently getting louder and higher in picth so again I killed the power to the amp.
I cannot understand what is going on or how it has happened from simply changing the caps. Nothing else in the amp has changed and all the four joins that were made to put these 0.47 coupling caps in look and test fine.
If anybody could chime in with a possible suggestion I would be very grateful.
Many thanks in advance!
OK so I de-soldered the two Jupiter Copper paper and wax 0.47 coupling caps and used alligator clips to put the original caps back in (trying to step back to where I was yesterday when I measured the voltages successfully at the valve pins) and the amp switches on fine with no crazy transformer buzzing sounding like it will tear itself apart.
I switch off the amp and then check the Jupiters on the capacitance meter, they both read 0.5UF no shorts. I clips them into the alligator clips turn on the amp and after warm up there is is again the horrendous loud buzz / oscillation that sounds like the amp is about to tear itself apart!?
Any advice on what is happening here would be greatly appreciated. There must be something wrong with the Jupiters surely?
Is this to do with the orientation of the capacitors?
I switch off the amp and then check the Jupiters on the capacitance meter, they both read 0.5UF no shorts. I clips them into the alligator clips turn on the amp and after warm up there is is again the horrendous loud buzz / oscillation that sounds like the amp is about to tear itself apart!?
Any advice on what is happening here would be greatly appreciated. There must be something wrong with the Jupiters surely?
Is this to do with the orientation of the capacitors?
I had extended the leads a fair amount to be able to locate these caps well away from the tube sockets, would the lead length possibly cause an unwanted change to the properties or function of the coupling caps in this circuit? They are closer to the psu currently, is there anything to consider in terms of interference from other components?
This is very weird. It sounds like something is oscillating when you have the jupiters in.
Have you perhaps have other caps lying around...other than the Jupiters..perhaps try those...
It can make a difference which way you wire them...(more susceptible to hum if not wired correctly) In a coupling cap situation the outer foil should be connected to the plate of the tube it is coming from.
Where To Connect the Outside Foil on Capacitors
Thanks for the link and the info. This morning I tried the caps in different orientations but the oscillation still occurred. I put the old original polyester caps in again on alligator clips attached to the extended leads as this worked last night but this time it oscillated with those caps also.
Its a shot in the dark but is it possible that the long leads and the placement of these is somehow inducing the oscillation? An increase in inductance?
I guess I should try the caps soldered close to the valve pins to know for sure. Very strange that last night I had no issue with the original caps and this morning there is an issue with them in. No solder joins had been changed in either case.
Its a shot in the dark but is it possible that the long leads and the placement of these is somehow inducing the oscillation? An increase in inductance?
I guess I should try the caps soldered close to the valve pins to know for sure. Very strange that last night I had no issue with the original caps and this morning there is an issue with them in. No solder joins had been changed in either case.
Just a couple of observations,
1. The Mundorf SIO caps are quite small inside a 10 times bigger plastic rubber case, it is made of silver paste spread over a tiny metalized poly film. The poly film can deform at 170 Celsius, so the damage is caused during soldering to the rubber case. Unless the Almaro heats up the cap to 170 C I would trust the Mundorf, which is the most soft sounding caps you can get.
2. Jupiter bees wax is stable up to 70 Celsius and I would never put such a cap close to power tubes or high dissipating resistors or tube legs. Many tube amps can heat up caps to 60 Celcius and 80 C in hot weather.
3. Polyester caps can start deform as low as at 50 Celcius but many are rated 125 C.
4. Check the feedback path integrity and all solder connections for contact/cracks.
1. The Mundorf SIO caps are quite small inside a 10 times bigger plastic rubber case, it is made of silver paste spread over a tiny metalized poly film. The poly film can deform at 170 Celsius, so the damage is caused during soldering to the rubber case. Unless the Almaro heats up the cap to 170 C I would trust the Mundorf, which is the most soft sounding caps you can get.
2. Jupiter bees wax is stable up to 70 Celsius and I would never put such a cap close to power tubes or high dissipating resistors or tube legs. Many tube amps can heat up caps to 60 Celcius and 80 C in hot weather.
3. Polyester caps can start deform as low as at 50 Celcius but many are rated 125 C.
4. Check the feedback path integrity and all solder connections for contact/cracks.
Thank you for your advice here. I will double check the 1.5k feedback resistors and parallel 330pf caps.
The Jupiter's are the version up from the beeswax version. I am unsure whether or not this makes any difference in terms of its temperature stability or if it uses bees wax also. The specs state 80degrees for the copper foil paper and wax whereas the mundorf Sio state 70 degrees. I have seen many threads where it is claimed however that the mundorfs are actually 55 degrees.
I have three sets of coupling caps at my disposal.
- the mundorf Sio that have been in there for over three years and look like they might be leaking
- the original stock film ones polyester I think.
- the new Jupiter copper paper and wax version.
I have been doing some reading on tube amp oscillation to try and educate myself a little and see references to "phantom circuits that can be hard to trace" plus advice as mentioned above to check the feedback circuit.
The Jupiter's are the version up from the beeswax version. I am unsure whether or not this makes any difference in terms of its temperature stability or if it uses bees wax also. The specs state 80degrees for the copper foil paper and wax whereas the mundorf Sio state 70 degrees. I have seen many threads where it is claimed however that the mundorfs are actually 55 degrees.
I have three sets of coupling caps at my disposal.
- the mundorf Sio that have been in there for over three years and look like they might be leaking
- the original stock film ones polyester I think.
- the new Jupiter copper paper and wax version.
I have been doing some reading on tube amp oscillation to try and educate myself a little and see references to "phantom circuits that can be hard to trace" plus advice as mentioned above to check the feedback circuit.
Thank you for the reply,
mundorf Sio can't leak because they contain no liquid... maybe just melted plastic from the casing.
It is easy to see if it's the temperature which is the cause of oscillation if it occurs as soon as you turn it on cold.
When I build my amps I first listen to it without feedback to make sure the circuit is sound, only then I connect feedback carefully!
mundorf Sio can't leak because they contain no liquid... maybe just melted plastic from the casing.
It is easy to see if it's the temperature which is the cause of oscillation if it occurs as soon as you turn it on cold.
When I build my amps I first listen to it without feedback to make sure the circuit is sound, only then I connect feedback carefully!
Thanks again for the advice.
I shortened the wire for the 0.47 coupling caps using the Jupiter's and the huge oscillation still kicks in every time as soon as the amps soft start has finished. This is with a dummy load of 8ohms resistance at the output.
I replaced the 1.5k feedback resistor and parallel 330pf cap and also played with the orientation of various signal wires but still every time the amp goes into a spin sounding like it will mechanically tear its self apart if on for more than a few seconds after soft start. I dont dare to leave it on and hit the power switch leaving some time before trying again.
I can't for the life of me understand what changed to bring about this issue. I took out the Jupiter's and put the stock polyester in on jump leads again but still the same problem.
If I remove the 1.5k feedback resistor and its cap and jumper this position my understanding is that the amp should function but there will be no negative feedback, is this correct?
The reading I have done on this seems to point towards interstage coupling, dressing and shielding of wires ans of course the feedback circuit. None of the values have been changed so this must be an issue that I have introduced.
I shortened the wire for the 0.47 coupling caps using the Jupiter's and the huge oscillation still kicks in every time as soon as the amps soft start has finished. This is with a dummy load of 8ohms resistance at the output.
I replaced the 1.5k feedback resistor and parallel 330pf cap and also played with the orientation of various signal wires but still every time the amp goes into a spin sounding like it will mechanically tear its self apart if on for more than a few seconds after soft start. I dont dare to leave it on and hit the power switch leaving some time before trying again.
I can't for the life of me understand what changed to bring about this issue. I took out the Jupiter's and put the stock polyester in on jump leads again but still the same problem.
If I remove the 1.5k feedback resistor and its cap and jumper this position my understanding is that the amp should function but there will be no negative feedback, is this correct?
The reading I have done on this seems to point towards interstage coupling, dressing and shielding of wires ans of course the feedback circuit. None of the values have been changed so this must be an issue that I have introduced.
If you remove the feedback and the amp still oscillate you have something like a damaged winding, damaged rectifier, damaged power supply connection.
Measure all resistors values.
The only other things I could think (now that you eliminated the caps) is that the new volume pot acts like an antenna and the amp amplifies the parasitic. The feedback re-entry point is also critical, the wires should be placed exactly like they were originally.
This amp looks like it is designed to work with healthy amount of Fb. But you can still try it without and hope the buzzing will stop. It is important to isolate which transformer is buzzing.
Try 500K to 1M resistors (one/channel) between the input of volume control and ground of input. (can stay permanent)
Place 500K to 1M resistors between the socket pin#1 of input tube and ground. (can stay permanent)
Place a 2K resistor in series after the 0.22uf cap to the 6c33b input pin #5. (can stay permanent)
Measure all resistors values.
The only other things I could think (now that you eliminated the caps) is that the new volume pot acts like an antenna and the amp amplifies the parasitic. The feedback re-entry point is also critical, the wires should be placed exactly like they were originally.
This amp looks like it is designed to work with healthy amount of Fb. But you can still try it without and hope the buzzing will stop. It is important to isolate which transformer is buzzing.
Try 500K to 1M resistors (one/channel) between the input of volume control and ground of input. (can stay permanent)
Place 500K to 1M resistors between the socket pin#1 of input tube and ground. (can stay permanent)
Place a 2K resistor in series after the 0.22uf cap to the 6c33b input pin #5. (can stay permanent)
Last edited:
Thank you very much for your help with this. I will try these suggestions.
The crazy thing is that it was working (except for the imbalance) without oscillation with all of the new components and then suddenly this issue appears upon swapping caps on the end of jump leads except that I cannot now get back to the non oscillating state of the amplifier.
- I will have a look at the pot, perhaps some of the wires are a little longer than they should be.
- I could also try taking the pot out of the circuit all-together temporarilty turning it into a straight power amp.
- I think that perhaps the wiring and placement of the feedback resistors and 330pf caps needs to be revisited. Its tricky to get this as tidy and close as it was from stock. I regret removing these parts now but one of the resistors was suspected as shorted.
- I have checked all of the other resistors again and am getting correct readings.
- I will also try your kind suggestions for the added resistors. I take it these may tame the oscillation. Would you be able to elaborate on what the addition of these resistors will do? Is it a case of changing the impedance of the circuit and should this work are there any negative consequences of these additions?
The crazy thing is that it was working (except for the imbalance) without oscillation with all of the new components and then suddenly this issue appears upon swapping caps on the end of jump leads except that I cannot now get back to the non oscillating state of the amplifier.
- I will have a look at the pot, perhaps some of the wires are a little longer than they should be.
- I could also try taking the pot out of the circuit all-together temporarilty turning it into a straight power amp.
- I think that perhaps the wiring and placement of the feedback resistors and 330pf caps needs to be revisited. Its tricky to get this as tidy and close as it was from stock. I regret removing these parts now but one of the resistors was suspected as shorted.
- I have checked all of the other resistors again and am getting correct readings.
- I will also try your kind suggestions for the added resistors. I take it these may tame the oscillation. Would you be able to elaborate on what the addition of these resistors will do? Is it a case of changing the impedance of the circuit and should this work are there any negative consequences of these additions?
the first res. is to lower the resistance of the pot back to something close to original 100K, no drawbacks or effects on sound
second res. is a grid leak resistor, no drawbacks or effects on sound
third it a grid stopper to limit bandwidth of the input pin, can prevent oscillation, no drawbacks or effects on sound
I haven't tried adding the resistors that you have mentioned yet but I have shortened and tidied up the input wires to and from the 100k stepped shunt pot as before I had a long section that was un-shielded to and from the pot. This is now been reduced to a 10mm or so. Unfortunately this didn't fix it but I figured it was worth a try.
I then cut the wire which carries the feedback via 1.5k and parallel 330pf to the channel which I believe is oscillating and turned the amp on and the oscillation is gone.
I am going to try for a third time to replace the feedback line with the resistor and cap on the defective side that I have identified, taking care to dress and locate the wires as they were originally. Unfortunately I have lost length so will need to extend with a different shielded cable. I will also trace the signal on that side and re-check solder joins between the stages.
I then cut the wire which carries the feedback via 1.5k and parallel 330pf to the channel which I believe is oscillating and turned the amp on and the oscillation is gone.
I am going to try for a third time to replace the feedback line with the resistor and cap on the defective side that I have identified, taking care to dress and locate the wires as they were originally. Unfortunately I have lost length so will need to extend with a different shielded cable. I will also trace the signal on that side and re-check solder joins between the stages.
OK an update :
After identifying the channel that is oscillating I then traced the signals through the stages looking closely at these in line with the valve pin-out diagrams at each stage. I noticed that plate 2 on the first valve was running to grid 1 on the first valve. Obviously this isn't an issue but I thought I should swap the output jump leads on the two 0.47 coupling caps so that section 1 on the first valve runs to section 1 on the second valve for ease of tracing. I then had the strange thought that I should put the feedback circuit back in "just in case" this minor change of the signal paths through the various wires has an effect. Turned the amp on and no oscillation with the polyester stock caps on jump leads.
I swapped the polyester for the Jupiters and still no oscillation. To try and repeat the oscillation to be sure, I switched the output leads from the 0.47uf pair around and back it came.
Of course this makes little sense to me, is this a case of the amp being in such a delicate balance with negative feedback, that the routing of the cables channel to channel is the difference?
After identifying the channel that is oscillating I then traced the signals through the stages looking closely at these in line with the valve pin-out diagrams at each stage. I noticed that plate 2 on the first valve was running to grid 1 on the first valve. Obviously this isn't an issue but I thought I should swap the output jump leads on the two 0.47 coupling caps so that section 1 on the first valve runs to section 1 on the second valve for ease of tracing. I then had the strange thought that I should put the feedback circuit back in "just in case" this minor change of the signal paths through the various wires has an effect. Turned the amp on and no oscillation with the polyester stock caps on jump leads.
I swapped the polyester for the Jupiters and still no oscillation. To try and repeat the oscillation to be sure, I switched the output leads from the 0.47uf pair around and back it came.
Of course this makes little sense to me, is this a case of the amp being in such a delicate balance with negative feedback, that the routing of the cables channel to channel is the difference?
Last edited:
I am happy for you the issue was only the feedback line picking up interference.
Mundorf caps somehow catch more parasitic from nearby components and are more delicate for placement.
To me it makes sense. Yes: Routing of the feedback is super important, tube amps especially with the lack of regulators, shunt stages. Solid state have more issues with power supply ground point loops.
Mundorf caps somehow catch more parasitic from nearby components and are more delicate for placement.
To me it makes sense. Yes: Routing of the feedback is super important, tube amps especially with the lack of regulators, shunt stages. Solid state have more issues with power supply ground point loops.
All components soldered back in, Jupiter caps now properly installed.
The oscillation issue has gone and I can finally get the amp on and leave it on.
I have just begun checking voltages again before going into the system with it and I noticed there is 120mv of dc at the speaker outputs on both sides at the 8ohms taps and 86mv at the 4ohm taps each side.
This isn't something I have checked before so I have no reference point prior to the work I have done on the amp. I know this is certainly not great and appears quite high but is this a big red flag of an issue? Looking around on this forum there are a few views on what is acceptable and what could be a problem.
The oscillation issue has gone and I can finally get the amp on and leave it on.
I have just begun checking voltages again before going into the system with it and I noticed there is 120mv of dc at the speaker outputs on both sides at the 8ohms taps and 86mv at the 4ohm taps each side.
This isn't something I have checked before so I have no reference point prior to the work I have done on the amp. I know this is certainly not great and appears quite high but is this a big red flag of an issue? Looking around on this forum there are a few views on what is acceptable and what could be a problem.
- Status
- This old topic is closed. If you want to reopen this topic, contact a moderator using the "Report Post" button.