What do you think causes pins 2 and 3 to arc in octal sockets.

[Mark]

I was talking to someone on You Tube about the octal sockets burning between pins 2 and 3. His opinion was that it’s caused by flyback from a speaker connection going open circuit. I don’t know if I entirely believe this (of course I could be wrong) as it would be easier for the arc to go through air than a socket I assume.

I have heard people say there was dust and moisture via humidity at play, others have said the socket became hot and started to burn. Typically, I have only see this sort of thing happen to amps with a few years on the clock. Though my Boogie MK1 did a socket form this carbon track after about four or five years. Oddly enough I have never encountered this sort of thing again. I typically put ceramic sockets in the output stage of my amps to ensure this sort of thing doesn’t happen.

Thanks for reading my post. What are your thoughts?

[Stevem]

I think it's also due to the ac filament power.
Pin 3 is highly positive D.C.
Pin 2 at a rate 50 or 60 times a second goes from being positive ac to zero and then that starts the attraction of D.C.
Then pin2 then goes highly negative in relation to the D.C. Polarity of pin 3, so even a higher level of attraction is created.

[martin manning]

I was talking to someone on You Tube about the octal sockets burning between pins 2 and 3. His opinion was that it’s caused by flyback from a speaker connection going open circuit. I don’t know if I entirely believe this (of course I could be wrong) as it would be easier for the arc to go through air than a socket I assume.

I have heard people say there was dust and moisture via humidity at play, others have said the socket became hot and started to burn. Typically, I have only see this sort of thing happen to amps with a few years on the clock. Though my Boogie MK1 did a socket form this carbon track after about four or five years. Oddly enough I have never encountered this sort of thing again. I typically put ceramic sockets in the output stage of my amps to ensure this sort of thing doesn’t happen.

I agree with the notion that flyback voltage and dust with some moisture content on the socket is likely to be the fatal combination, drawing the arc along the surface. Seems like giving the power tube sockets on older amps a good cleaning with alcohol might be a good idea.

I think it's also due to the ac filament power.
Pin 3 is highly positive D.C.
Pin 2 at a rate 50 or 60 times a second goes from being positive ac to zero and then that starts the attraction of D.C.
Then pin2 then goes highly negative in relation to the D.C. Polarity of pin 3, so even a higher level of attraction is created.

I don't think +/- 10V potential is likely to mean much. In normal operation the plates swing to near 1kV (2x B+), and flyback voltage will be much higher.

[Mark]

I think it's also due to the ac filament power.
Pin 3 is highly positive D.C.
Pin 2 at a rate 50 or 60 times a second goes from being positive ac to zero and then that starts the attraction of D.C.
Then pin2 then goes highly negative in relation to the D.C. Polarity of pin 3, so even a higher level of attraction is created.

Thanks for your reply, I appreciate you taking time to reply. Do you think the ac voltage change isn’t particularly large in comparison to the 420 - 440VDC on pin 3?

I have noticed the sockets affected do tend to look darker than original when they arc over.

[Mark]

I agree with the notion that flyback voltage and dust with some moisture content on the socket is likely to be the fatal combination, drawing the arc along the surface. Seems like giving the power tube sockets on older amps a good cleaning with alcohol might be a good idea.

Thanks for your reply Martin I appreciate you taking time to answer my question. How is this flyback voltage generated. The individual claimed it was from an open circuit on the secondary. I had heard people claim an open circuit secondary leads to shorted turns because the insulation is damaged, but I haven’t heard it affecting the parts in primary. Given that the amp in question was a combo, I doubt the speakers were ever disconnected from the amp when it was working and both speakers are in parallel so it’s unlikely both speakers wouldn’t be making contact at the same time due to faulty connectors.

As I have previously stated to Steve, I noticed the valve sockets tend to darken over time with heat. I was wondering if this played a part in the problem?

[trobbins]

Photo of pin 2-3 arc damage on ceramic socket in attached article on such protection issue.
https://dalmura.com.au/static/Output%20 ... ection.pdf

[Mark]

Photo of pin 2-3 arc damage on ceramic socket in attached article on such protection issue.
https://dalmura.com.au/static/Output%20 ... ection.pdf

Thanks for the link to the article. I will read it. I thought I saw a damaged ceramic socket. That would be quite a feat.

[martin manning]

How is this flyback voltage generated. The individual claimed it was from an open circuit on the secondary. I had heard people claim an open circuit secondary leads to shorted turns because the insulation is damaged, but I haven’t heard it affecting the parts in primary. Given that the amp in question was a combo, I doubt the speakers were ever disconnected from the amp when it was working and both speakers are in parallel so it’s unlikely both speakers wouldn’t be making contact at the same time due to faulty connectors.

Flyback voltage is generated when the current flowing in an inductor is suddenly interrupted. As in auto ignition coils and switch mode power supplies, the collapsing magnetic field induces high voltage in the coils. A sudden interruption in current happens when the power tubes go into cutoff. If the secondary is open, very high voltages will appear on the primary, which greatly increases the possibility of primary insulation failure (internal arcing), or arcing from the plate leads to nearby conductors at lower potential. A poor speaker jack connection due to corrosion could possibly produce this effect.

As I have previously stated to Steve, I noticed the valve sockets tend to darken over time with heat. I was wondering if this played a part in the problem?

I suppose the insulation properties of the sockets could be deteriorating.

[pdf64]

... How is this flyback voltage generated. The individual claimed it was from an open circuit on the secondary. I had heard people claim an open circuit secondary leads to shorted turns because the insulation is damaged, but I haven’t heard it affecting the parts in primary. ..,

As Martin mentioned, the high voltages are caused by inductive flyback, back emf, the same mechanism by which 12V from a car battery can generate 20kV at the spark plugs. ie via the coil.
The speaker voicecoil is an inductor, the output transformer OT secondary is an inductor, the OT primary is a large, high value inductor.
The shorted turns will be in the primary. The instantaneous voltage spikes in an overdriven amp will have one anode several kV above 0V common, the other anode a mirror image at several kV below 0V common. So maybe >5kV between the anodes. That voltage is obviously between primary windings also.
Hence insulation in the primary circuit can tend to breakdown.

[pdf64]

... If the secondary is open ...

Just to expand, even without an open / intermittent load on the secondary, I think the inductive nature of a regular cab will tend to result in leading edge back emf spikes with a heavily overdriven amp.

[bepone]

1. internal tube structure problem, some bad assembly, then arcing internally..

2. on sockets because of flux between in 2 and 3, which with time strting to be conductive.

3. because of bad sockets, porosity in materials and humidity

[wpaulvogel]

If the case is a single ended amp that the power section is overdriven, the voltages in the output transformer and tube socket/ tube will experience induced voltages that come close to twice the B+ supply voltage. The chance of arcing is more common in single ended amps.

[pdf64]

My understanding is that whether SE or push pull, with signal, the anode voltage swings as far above the HT as it gets pulled below the HT.
eg if full load HT voltage is 400V and peak anode current (at Vg1-k=0 or whatever) pulls the anode voltage down to 50V, then as the g1 signal swings down to its lowest level and puts anode current into cut off, anode voltage will swing up to 750V.

[FUCHSAUDIO]

We use ceramic sockets for our power and preamp tubes. I usually use them for repairs (chassis permitting) as well. We do both manufacture and repair.

More often than not, we find guys with bad speaker wiring or cabinet issues will end up with the 2/3 burn. It's the most common spot (apart from a rare 4/5 burn).
It's transients from flyback voltages and/or dirt. No other real explanation for it. Ceramic is pretty tough to break through with high voltage. Usually the base of the tube shows (or smells) of a burn as well.
We change the tube and socket, screen and series grid resistors, de-flux thoroughly and rarely see the amps again. We usually try to offer a new speaker wire or a free check of the cabinet as well.

[trobbins]

The chance of arcing is more common in single ended amps.

I'd say an SE output stage is more prone too. The main reason is that there is no closely coupled half-primary winding (eg. with a PP OPT) to act as a local capacitance to soak up some of the energy that wants to instantaneously fly around.

Regular maintenance cleaning of the socket surface between pins 2-3 with IPA is well worth it to manage 'pollution' build up that can lead to reduced creepage and clearance, and then tracking/arcing. Some manufacturers were known to add spaghetti insulation tube over the pin 3 termination, and even glue that tube to the socket to form a sort of solid insulation barrier for increased clearance. That all helps avoid arcing at the valve base, but not stress on the output transformer itself or other weakpoints.

Maybe manufacturers like FUCHSAUDIO will start retrofitting a suitable MOV across each half-primary as a simpler and more direct way to mitigate such faults.