I'm stumped once again so am reaching out to all ye merry builders.
Brought latest build to a 'jam' last week for preliminary testing. It was built as a low wattage bass amp for studio & home use but should (and did) work well for louder clean guitar sounds. It's 40-50 watts, HV secondary is on the high side, 375-0-375, so I'm using a 5R4 rectifier for the voltage drop to keep B+ reasonable.
Fixed bias output stage. Last week, the B+ at the output tube plates was around 470VDC biased around 38mA or so. Brought it to the space, hooked it up, sounded huge! I was very pleased.
I drilled the chassis for external bias points and an external bias pot (using an Alpha 50K Linear pot which has worked beautifully on a few other builds) but I didn't have one on hand last week so I temporarily wired in a smaller "on board" style Piher bias pot of the same value (50K).
As mentioned, the amp tested beautifully and was used close to max volume for about 2 hours.
Waited all week for the 50K Alpha pot. Finally arrives. I take out the small Piher pot and wire up the Alpha. Everything's wrong - plate voltage is around 505VDC with output tubes biased to 35mA. THEN there is lightning inside the amp - on one of the output tube sockets I see arcing between pin 3 (OT primary) and the lead of the screen resistor right at pin 4 (screen). I shut the amp down immediately and clean the socket tabs. Calmly fire it up again and amp sounds shitty, like blocking distortion. Hit a chord and it sounds like someone is crumpling tinfoil inside the speaker cab. Did I just fry the OT? I saw this same type arcing happen once on a JTM45 clone I owned a while back. Shut the amp down the same way and fired it back up and it was fine. Maybe I was lucky that time?
Tested output tube/phase inverter coupling caps, both output tubes show -63V on the grids so caps should be fine. Wasn't sure about coupling caps from first tube or to PI from first stage so I replaced both of those coupling caps with no change or results.
Amp has a parallel 6SL7 going into another 6SL7 as LTP, straight into the output section. Volume & tone controls are between the input tube and the phase inverter. Simple amp that sounded heavenly when tested last week. Now it sounds like a garbage can.
Any suggestions before I bite the bullet and buy a fresh OT?
Weird Science
Moderators: pompeiisneaks, Colossal
Re: Weird Science
With an arc, I'd replace the tube socket.
Tube junkie that aspires to become a tri-state bidirectional buss driver.
Re: Weird Science
Thank you, TUBEDUDE, that seems unavoidable. Luckily I have a spare!
I'm wondering what could have caused the arc. Research indicates possibly an impedance mismatch? OT is a MC 40-18039, their JTM45 style. Though primary on these is 8K which is technically a mismatch to 6L6GC output tubes, it seems accepted knowledge that while maybe not operating optimally, tubes will function one step either way in mismatch.
And this seems true. I used this same model OT in another build and both 'obeyed' it's designation by using it's secondaries as intended and 'disobeyed' by intentionally mismatching say an 8ohm cab into the 16 ohm secondary to halve the primary impedance to 4K which is supposedly more along the lines of what these tubes want to see. Both ways worked flawlessly. So why wouldn't it work in this amp?
Maybe due to the abnormally high B+, which is the weird science part (despite the fancy light show inside the amp) because one week ago, with no components changed aside from the bias pot (for one of the same value) it measured a whopping 30 volts lower than today. Maybe the higher voltage was 'just too much' to contain and arcing was the result. But why the higher B+? Primary/wall voltage was stable/similar last week around 120.6 - 121 VAC
note - in OP I listed plate voltage to be 505VDC and it eventually settled down to around 501VDC. Still pretty high.
I'm wondering what could have caused the arc. Research indicates possibly an impedance mismatch? OT is a MC 40-18039, their JTM45 style. Though primary on these is 8K which is technically a mismatch to 6L6GC output tubes, it seems accepted knowledge that while maybe not operating optimally, tubes will function one step either way in mismatch.
And this seems true. I used this same model OT in another build and both 'obeyed' it's designation by using it's secondaries as intended and 'disobeyed' by intentionally mismatching say an 8ohm cab into the 16 ohm secondary to halve the primary impedance to 4K which is supposedly more along the lines of what these tubes want to see. Both ways worked flawlessly. So why wouldn't it work in this amp?
Maybe due to the abnormally high B+, which is the weird science part (despite the fancy light show inside the amp) because one week ago, with no components changed aside from the bias pot (for one of the same value) it measured a whopping 30 volts lower than today. Maybe the higher voltage was 'just too much' to contain and arcing was the result. But why the higher B+? Primary/wall voltage was stable/similar last week around 120.6 - 121 VAC
note - in OP I listed plate voltage to be 505VDC and it eventually settled down to around 501VDC. Still pretty high.
Re: Weird Science
*also should have added:
-at time of arcing amp was hooked up to 8ohm cab via 8ohm secondary (green) for 8K reflected primary impedance
- I have also built an amp using the same PT - Stancor PC-8411 - and output tube plate voltage on that amp w/5R4 rectifier is only 455VDC
-5V and 6.3V windings on PT check out ok
-output tubes are Russian made 6P3S-E type, same exact set was used in amp that uses same OT. No issues, and tubes are not old.
-at time of arcing amp was hooked up to 8ohm cab via 8ohm secondary (green) for 8K reflected primary impedance
- I have also built an amp using the same PT - Stancor PC-8411 - and output tube plate voltage on that amp w/5R4 rectifier is only 455VDC
-5V and 6.3V windings on PT check out ok
-output tubes are Russian made 6P3S-E type, same exact set was used in amp that uses same OT. No issues, and tubes are not old.
Re: Weird Science
Sorry, I have no.experience with those output tubes. With 500 on the plates, and 8k, I can imagine a high kickback voltage peak though. It's odd to have a suddenly high voltage if the tubes are biased up at 35 ma. Very strange. Even with some OT shorted coils the DC shouldn't climb that much I'd think. The socket is a cheap place to start.
Tube junkie that aspires to become a tri-state bidirectional buss driver.
Re: Weird Science
Thanks again, TUBEDUDE! I will definitely replace the socket next.
I guess I'm curious as to why the B+ would be so much higher. It is of note that the B+ was in excess of 500VDC BEFORE the arcing happened, and the only difference was the Alpha bias pot (vs the Piher that was in there). I feel like if I could find the cause of the higher B+ I could solve the issue. I tried a different set of power tubes - actual 6L6GCs - and the plate voltage was the same. I tried a different 5R4 rectifier tube, still the same. I swapped all preamp tubes, still the same.
I guess I'm curious as to why the B+ would be so much higher. It is of note that the B+ was in excess of 500VDC BEFORE the arcing happened, and the only difference was the Alpha bias pot (vs the Piher that was in there). I feel like if I could find the cause of the higher B+ I could solve the issue. I tried a different set of power tubes - actual 6L6GCs - and the plate voltage was the same. I tried a different 5R4 rectifier tube, still the same. I swapped all preamp tubes, still the same.
Re: Weird Science
When a meter gives an unexpected reading, it's always a good idea to check its battery, check that the mains voltage is in the expected range etc.
https://www.justgiving.com/page/5-in-5-for-charlie This is my step son and his family. He is running 5 marathons in 5 days to support the research into STXBP1, the genetic condition my grandson Charlie has. Please consider supporting him!
Re: Weird Science
Hi Pete, thanks for your suggestions!
The DMM used was brand new w/fresh battery installed. I sent my other one overseas with some gear last week so I bought a new one for home. The old one is great and a little easier to use but new one seems well made/good quality as well.
Mains voltage was around 120.1 when I read plate voltage to be in the 500 zone.
Could it be the bias pot? It sounds/seems unlikely but is it possible?
The DMM used was brand new w/fresh battery installed. I sent my other one overseas with some gear last week so I bought a new one for home. The old one is great and a little easier to use but new one seems well made/good quality as well.
Mains voltage was around 120.1 when I read plate voltage to be in the 500 zone.
Could it be the bias pot? It sounds/seems unlikely but is it possible?
Re: Weird Science
OK, it's always worth verifying the info you're working with.
Another thing to check are the 1 ohm current sensing resistors, as if the current draw is lower than you think it is, then the HT will tend to rise.
Yes, I could envisage that a dodgy bias pot could send the plate current crazy, which may result in arcing.
Another thing to check are the 1 ohm current sensing resistors, as if the current draw is lower than you think it is, then the HT will tend to rise.
Yes, I could envisage that a dodgy bias pot could send the plate current crazy, which may result in arcing.
https://www.justgiving.com/page/5-in-5-for-charlie This is my step son and his family. He is running 5 marathons in 5 days to support the research into STXBP1, the genetic condition my grandson Charlie has. Please consider supporting him!
Re: Weird Science
+1 on the Bias pot and replacing the socket.
What screen resistor values are you using?
What screen resistor values are you using?
Re: Weird Science
Right now I have a 1K 5W 'flying resistor' (?) connected to each output tube's own 3K 5W screen resistor. Screen voltage is nice & safe, about 20-30 volts below plate voltage.
I just left the country for a 4 week tour but when I get back I'll change the socket as well as the bias pot.
The sound was so sh*tty though, even at low volume, that I'm stillafraid the OT may be toast.
Will report back in a few weeks.
THANKS!!
I just left the country for a 4 week tour but when I get back I'll change the socket as well as the bias pot.
The sound was so sh*tty though, even at low volume, that I'm stillafraid the OT may be toast.
Will report back in a few weeks.
THANKS!!
Re: Weird Science
Finally got back into this - turns out the OT was indeed fried. Swapped the tube socket & checked the bias pot, same problem. I had another amp with the same OT so I pulled the old OT and wired in the known working one and voila, success.
Thankfully I have grown a little stockpile of transformers over the past few years so I had a similar OT I was able to wire into the amp which I pulled the replacement from. As for the bad OT (it's a Classictone 40-18039) I feel like it'd be more expensive to send it back to MC and get rewound than to just buy another and since I don't really need it at the moment I think it just might hit the garbage can :/ oh well, cost of doing business I guess.
Once again, thanks for all your help!
Thankfully I have grown a little stockpile of transformers over the past few years so I had a similar OT I was able to wire into the amp which I pulled the replacement from. As for the bad OT (it's a Classictone 40-18039) I feel like it'd be more expensive to send it back to MC and get rewound than to just buy another and since I don't really need it at the moment I think it just might hit the garbage can :/ oh well, cost of doing business I guess.
Once again, thanks for all your help!
Re: Weird Science
labb, I think perhaps for some reason the amp didn't like having it's bias pot swapped out? Either way, it was arcing on one of the power tubes between pins 3 & 4 that caused the meltdown as mentioned in the original post. Turns out I really was lucky when I saw the same thing happen before with no destructive results, because this time it was bad bad bad.
I was being impatient. I wanted to test the amp and I had a golden opportunity to do so, but I was still waiting on a part. So I subbed in a different part of equal value to hold it's place for testing. The test took place an hour away from home during a friend's band rehearsal which my schedule rarely allows for, so I didn't want to miss it. But, as the kid says to the rabbit in the Trix commercial - patience is a virtue.
The lesson, for me at least, is not to test an amp until it's properly complete. I learnt this back in the days of owning and working on old motorcycles. I would have a bike running nicely, everything in order, and for whatever reason there would be something I wanted to change or improve. Or I'd let a friend talk me into trying a modification. I'd remove a known working part, swap in a brand new part in it's place and all hell would break loose, all sorts of wacky problems would arise many having nothing to do with the part I swapped out. Those old engines get used to having the same parts stuck together for 20 or 30 years and they don't take kindly to new pieces regardless of their certification on paper. It's a little different when it comes to a brand new amp made with all new parts, but I think it's more that if I decide I want to build an amp using specific parts and values I should see to it that it gets done that way before attempting to conduct any conclusive testing.
I was being impatient. I wanted to test the amp and I had a golden opportunity to do so, but I was still waiting on a part. So I subbed in a different part of equal value to hold it's place for testing. The test took place an hour away from home during a friend's band rehearsal which my schedule rarely allows for, so I didn't want to miss it. But, as the kid says to the rabbit in the Trix commercial - patience is a virtue.
The lesson, for me at least, is not to test an amp until it's properly complete. I learnt this back in the days of owning and working on old motorcycles. I would have a bike running nicely, everything in order, and for whatever reason there would be something I wanted to change or improve. Or I'd let a friend talk me into trying a modification. I'd remove a known working part, swap in a brand new part in it's place and all hell would break loose, all sorts of wacky problems would arise many having nothing to do with the part I swapped out. Those old engines get used to having the same parts stuck together for 20 or 30 years and they don't take kindly to new pieces regardless of their certification on paper. It's a little different when it comes to a brand new amp made with all new parts, but I think it's more that if I decide I want to build an amp using specific parts and values I should see to it that it gets done that way before attempting to conduct any conclusive testing.
Re: Weird Science
It's quite possible the damage was done on the first occurrence. The arcing is likely to have caused breakdown of the insulating varnish - but maybe the turns didn't short. That left the OT as an accident waiting to happen. On the 2nd arcing, with inadequate insulation, those surly electrons just jumped willy-nilly across the turns and made it toast. IMHO, that OT was already dead, but didn't know it.