Entire Assembly Glowing in 20 Seconds

[MellowTone]

Hi all, I recently put together a 6V6 amp kit that I ordered, had a few problems (was the first tube amp I had tried at, have since completed a Simple SE) mostly due to stupid mistakes (pretty much my forte, really) but I've finally got it to a point where it makes noise.
I first tried the Shuguang 6V6 tubes that came with it, they quickly buzzed and arced (nice blue zapping inside). I thought it might be faulty tubes (I'm only running one channel now to test) so I put in some Russian 6L6G/B equivalents that I have (6P3S) and they did much the same. In between I also inspected the circuit and checked it over, I couldn't find any stuff ups. As a last ditch effort I put in a set of the Russian Military spec 6P3S-e tubes. These are what I run in my Simple SE and I can throw just about anything at them and they'll take it. This time the tube did not arc, but the entire centre assembly glowed within twenty seconds. I'm not entirely sure of the name of this section, but I think it's the second grid.
Does anyone know what could cause this whole section to glow so quickly or what could cause the arcing? If I have an idea of what causes it then I think I'll have a better chance of tracking down the component/s or wiring that is suspect.
Nic.

 

[rknize]

Grids and screens swapped? Is this a PCB kit? Schematic would be nice. Measure voltages at the various pins on the tube socket without a tube in there.

 

[jechentau]

indeed, the grids 1 glow red hot, while the plates do not get as hot

 

[richwalters]

rknize prob correct.
Similiar problem I recently dealt with here.. beginner Sonnyboy wired the sockets the wrong way. 1=8 ;
so the anode pin 6 on an Ioctal would become o/c, g1/2 swapped but heaters still work. Pins 1,8 would work.
Make sure if there is negative bias this appears on pin5.


richy

PS. There's no excuse to get this base wrong.

 

[Yvesm]

To do so with a PCB, you have to mount socket on the wrong side !
Check for that.

Yves.

 

[Arnulf]

Your trully forgot to change heater supply the other day when swapping series heater version of a tube with a parallel one. End result was even more glow (but no arcing inside) and the tube stabilized itself with ungodly high emission and consequently high current due to PTC nature of the filament before I realized that something was wrong and turned it off.

Could it be that your heater supply is waaaaay too high ? This could be due to unsuitable transformer, above avregare line voltage, incorrect wiring, etc.

 

[Geek]

If it's wired OK, it could be oscillating at RF.

Cheers!

 

[DigitalJunkie]

Or if it uses global NFB,maybe it's out of phase.

 

[MellowTone]

The kit is a point to point wired amp, which in hindsight was a bad choice for a first amp, as I tend to make many simple mistakes, especially with wiring. Just a few more bits that might help; after power on (once I've turned it off again) the capacitor in parallel with the cathode resistors becomes very hot, doesn't smoke but is too hot to touch; exactly the same thing happens on both channels, so if I have buggered up wiring it is a mirror on both channels; if powered on with the (6SL7) preamp tubes in as well, it WILL make a noise through the speaker, starts with nothing then comes in with a sort of feedback or loud hum, which then fades out once more and comes in even louder.

OK, powered up with no tubes:

P3 = 381v
P4 = 0v
P5 = 379v
P8 = 0v

On one channel, I have just reversed the Ultralinear (23%) and Output taps to check if I had messed that up. On powering up, it did exactly the same as before.

Looking closely at the pins inside, there is obviously a huge current flowing through pin 5 (grid 1) - both rods holding the grid, plus the grid itself are red hot as before.

Arnulf - the heater voltages are right on 6.3 volts + 0.1 volts.

Here are the schematics, it's all point to point wiring:

An externally hosted image should be here but it was not working when we last tested it.

 

[richwalters]

What o/p tubes are you actually using.6V6 6P6P ? For problem solving keep to known pairs that work. Don't use fresh tubes when things aren't working properly.My first instinct is to disconnect the global nfb loop,turn vol to min and allow to warm up. . As Geek mentions, if there are oscillations, a cheap AM radio tuned off station MW) put reasonably close by will make noises if circuit oscillates. If the output stage oscillates the speaker cone will usually displace and may emit 100Hz modulation hum.
As the electrolytic cap on the output stage cathodes has got hot, then at some time when all is working it will need replacing. I'm a bit suprised there is no fuse in the B+. A 500mA will do fine in the centre tap o/p tranny.
>Check that phase compensation components R8 C2 are in circuit.

 

[Arnulf]

First of all, I'm sure you made a mistake up there when you said you measured almost full anode voltage on the grid of both output tubes ? Pin 5 is grid 1 (where reading without tube in place should be 0V) and pin 4 is grid 2 (screen) where you should measure full anode voltage with no tube in place).

the capacitor in parallel with the cathode resistors becomes very hot, doesn't smoke but is too hot to touch

C6 ? Replace it with a new one and make sure you have the polarity right (schematic is correct). If the overheated (and by now probably defective) one blows it will spit its guts all over your amplifier and clenaing up the foily residue is a major PITA even if you don't mind the smell.

On one channel, I have just reversed the Ultralinear (23%) and Output taps to check if I had messed that up. On powering up, it did exactly the same as before.

Your schematic looks fine. Make sure you get the taps right, don't just randomly reverse connections on a whim

 

[MellowTone]

Good news! Upon finally figuring out that the Screen was in fact Grid 2 (Haha nice work on my part there) it is working. I did indeed have the grid and screen connections back to front. Now I just have an annoying hum (a lot of 50Hz and 100Hz buzz) to contend with. I would guess this is in the grounding connections, which I am going to completely re-do.

What o/p tubes are you actually using.6V6 6P6P ? For problem solving keep to known pairs that work. Don't use fresh tubes when things aren't working properly.My first instinct is to disconnect the global nfb loop,turn vol to min and allow to warm up. . As Geek mentions, if there are oscillations, a cheap AM radio tuned off station MW) put reasonably close by will make noises if circuit oscillates. If the output stage oscillates the speaker cone will usually displace and may emit 100Hz modulation hum.
As the electrolytic cap on the output stage cathodes has got hot, then at some time when all is working it will need replacing. I'm a bit suprised there is no fuse in the B+. A 500mA will do fine in the centre tap o/p tranny.
>Check that phase compensation components R8 C2 are in circuit.

I haven't actually got the global NFB hooked up at the moment. If I remember correctly, this will reduce distortion? (while reducing the volume a tad)
What is the purpose of the fuse on the centretap? I have seen this a few times on these forums and never really known exactly what the purpose is.

C6 ? Replace it with a new one and make sure you have the polarity right (schematic is correct). If the overheated (and by now probably defective) one blows it will spit its guts all over your amplifier and clenaing up the foily residue is a major PITA even if you don't mind the smell.:

Yes I thought I would probably have to replace them, I'll dig out some replacements ASAP. Can't say I'm a fan of exploding electolytics.

Your schematic looks fine. Make sure you get the taps right, don't just randomly reverse connections on a whim

Haha yes bad thought on my part. The documentation I received with the transformers wasn't all that clear, so I thought I may have got it wrong first time round.

 

[Arnulf]

To check where that hum is coming from:

1: Remove preamplifier/cathodyne tube and listen to the output. Is hum still present ? If yes, your power supply is probably missing filter capacitors (?!). Match the tubes and/or add a wire (high power) potentiometer in place of or in addition to cathode resistors and adjust it until the hum goes away.

2: Replace the preamplifier/cathodyne tube and listen to the output. Is hum present yet again ? If yes, filter your preamp supply better, sort out the grounding (use commongrounding point at filter capacitors if possible, if preamp has an additional RC/LC filter, use it as its grounding point).

 

[richwalters]

The mistake with the grids has strained the innards of the o/p tubes. They may appear to work and bias steady, but with other types i.e KT88's can continue to give all power but thd may rocket due to damaged heat misaligned grids. Note carefully the tubes you severely strained. Generally tubes that have been forcefully used become tempramental on the tap test and will fail sooner. While you are about it; Check the g2 screen grid resistors for correct value.
The fuse on the CT is mean't to save the expensive tranny and to some extent the tubes. The "trimming" the fuse value depends on whether MI or HiFi is used as to expected peak currents from beat notes.

richy

 

[MellowTone]

To check where that hum is coming from:

1: Remove preamplifier/cathodyne tube and listen to the output. Is hum still present ? If yes, your power supply is probably missing filter capacitors (?!). Match the tubes and/or add a wire (high power) potentiometer in place of or in addition to cathode resistors and adjust it until the hum goes away.

2: Replace the preamplifier/cathodyne tube and listen to the output. Is hum present yet again ? If yes, filter your preamp supply better, sort out the grounding (use commongrounding point at filter capacitors if possible, if preamp has an additional RC/LC filter, use it as its grounding point).

Thanks for the tips, I should have time on Monday to try to track down the buzz.

The mistake with the grids has strained the innards of the o/p tubes. They may appear to work and bias steady, but with other types i.e KT88's can continue to give all power but thd may rocket due to damaged heat misaligned grids. Note carefully the tubes you severely strained. Generally tubes that have been forcefully used become tempramental on the tap test and will fail sooner. While you are about it; Check the g2 screen grid resistors for correct value.
The fuse on the CT is mean't to save the expensive tranny and to some extent the tubes. The "trimming" the fuse value depends on whether MI or HiFi is used as to expected peak currents from beat notes.

richy

Yes I thought as much, I can't see them enjoying that. Could it be that some of the buzz/hum is being created or magnified by the damaged tubes? I currently have a set of replacements on the way (Soviet-era Russian 6SL7 and 1515 (early 'stronger' version of 6P6S, their equivalent of the 6V6) as I figure I had probably broken these tubes. Oh well, they're only Shuguag cheapeys that came with the kit. Not a huge fan of the smoked glass versions either.
Once I've rewired a few things I'll have a look into the fuse.