• WARNING: Tube/Valve amplifiers use potentially LETHAL HIGH VOLTAGES.
    Building, troubleshooting and testing of these amplifiers should only be
    performed by someone who is thoroughly familiar with
    the safety precautions around high voltages.

Preventing HV flashover on socket pins

Status
This old topic is closed. If you want to reopen this topic, contact a moderator using the "Report Post" button.
It occurs to me that one obvious cause of flashover on higher voltage tube amp designs is unfortunate increases in humidity in an environment and/or dust accumulation.

One way to cut down on this likelyhood (and even one flashover can be devastating to amp or tube) is to do the following:

(1) make sure that any likely expected repairs will not involve resoldering pin tabs. Mount for instance 1 ohm cathode resistors on a small turret or tab, and run a wire to socket.
Or make a mini-circuit board for resistor clusters around tubes.

(2) Carefully cleaning socket bottoms and covering pinholes, then spray High Voltage laquer / insulator on socket bottoms.
Do this lightly to prevent runs or leaks into socket holes. Bake dry with lamp to prevent dust pickup. Do a few coats.

Now in theory there will be much less opportunity for flashovers, traces, wormlines. The amp could be roaded into more difficult and harsh environments, including dusty smokey clubs and humid, drizzly outdoor concert venues.
 
Disabled Account
Joined 2010
Just for interest,

I use this idea on vero board projects..I just clean the flux off and spray with HT sealer/lacquer...Its important I think to use the sealer you can solder through and some even act as a flux..slightly different situation to above but it stops corrosion and tracking..I don't use it on the component side just the track side same on PCB's..

Regards
M. Gregg
 
I wouldn't use veroboard on HV projects.

However, I can't recall flashover being a problem with my experience of valve amps.

I think common sense prevails. Once the design is working, then any form of insulation will work. Heatshrink tubing is probably easier and less messy to use.
 
Disabled Account
Joined 2010
I wouldn't use veroboard on HV projects.

I built the oscilloscope project in the 70's from Practical wireless (vero project)..I think vero is OK as long as you know what your doing and don't put HV tracks right next to each other and cut away what your not using.. :)

But yes heat shrink is good..The problem is some lacquers once on you can't solder through them and some you can't dissolve with cleaners..then it gets messy..

Regards
M. Gregg
 
Last edited:
It's funny how now that tube amps are less common it seems like these problems have gone away.

This used to be a common problem with tube gear, most frequently PA and instrument amps, TV sets, and transmitters, where the voltage is over 500VDC.

I wonder if teflon sockets are more resistant to permanent tracking.

Circuit boards can be conformal coated. Some of the formulas are solder-through but of course you need to recoat after repairs.

You can spread the pins out to 0.2 inch spacing or stagger the pads on to-220 packages > 300VDC.
 
Disabled Account
Joined 2010
It's funny how now that tube amps are less common it seems like these problems have gone away.

This used to be a common problem with tube gear, most frequently PA and instrument amps, TV sets, and transmitters, where the voltage is over 500VDC.

I wonder if teflon sockets are more resistant to permanent tracking.

.

LOL...The brown old circuit boards around the HT doubler/tripler and the line time base... :D CICK CLICK CLICK PHUT.....white dot on the tube off line me thinks..LMAO

Regards
M. Gregg
 
At higher voltages you will get static attraction of dust.

Valve TVs, I can remember them, I was knee high to a grasshopper then. YES, Mum and Dad had an old 405 line VHF B&W set. We had to go to my Grandparents to watch a colour TV and that was transistorised (I think, an ITT back in 1970s)
 
Disabled Account
Joined 2010
I'll just solder a new socket in..

Er where is the circuit board..lump of carbon with copper tracks hanging in the air and a tube base swinging in the wind...:D

LMAO why did we go to circuit boards...LOL...the old point to point was better (saying in the late 70'S)..Huh its got cheap circuit boards,,,,It'll never last.

That was the other one...the dry joint of hell/// that gets hotter and hotter and strips the track of the board then blows like a fuse and tracks across the what used to be circuit board now carbon based life form..... :)

M. Gregg
 
Last edited:
Disabled Account
Joined 2010
As above,

Dust and HT is a major problem...mix that with damp and condensation, then green/blue copper corrosion and you have the future of the tube circuit.. one person said to me..well you can't stop the dust or damp but. If it can't come into contact with the circuit conductors then it should last..:D Untill it burns out..LOL

Regards
M. Gregg
 
However, I can't recall flashover being a problem with my experience of valve amps.

It isn't uncommon on guitar amps with the common audio tubes and 500 volts on the plates. Misload the output (wrong speaker impedance) then drive the amp beyond mere clipping into class D and there will be several KV spikes on the plates as the output tubes switch on and off. The plate is pin 3 and the heater is pin 2. The heater is usually a low impedance path to ground. If not... it will be when the arc jumps from pin 3 to pin 2, then breaks down the HK insulation in an output tube and hits the cathode bypass cap causing it to short an remove bias, resulting in meltdown! Once an arc starts due to a momentary signal transient it is fueled by the B+ supply. It won't stop until B+ is removed, usually by an open winding in the OPT.

PCBs have never been a favourite with Valve Amp projects.

All of my tube designs use PCB's. I have lived in South Florida for nearly 60 years. I did TV and consumer electronics repairs in the late 60's and early 70's. Many houses did not have AC back then so heat, humidity and dust were a big problem. Yes the early PCB's had problems. Arcs, carbonized or vaporized tracks were common and PCB's with holes burned through them weren't too uncommon. The old phenolic boards were hygroscopic and actually absorbed moisture. Moisture + copper + electricity = conductive green growth leading to fireworks!

Modern PC boards that are properly designed and laid out don't have the problems that the old stuff had. I have not seen a PCB related failure in one of my designs yet and I have been making them for 20+ years.

If you force one of my PCB based amps to arc (remove the speaker load at full power) the arc will usually be in the OPT, inside the output tube base, or inside the tube socket. Of course a dirty PCB is asking for trouble, but so is a dirty chassis.
 
Disabled Account
Joined 2010
Modern PCB's are very good,

Its just nice to remember the fun days of track repair..its just my pref to mount power tubes on flying leads to PCB's even if they are mounted mechanicaly on the board.. :)

I have not had a problem with small signal tubes..(Dry joints etc). The old PCBs were a royal pain the butt..I think that is what can make some restoration difficult with amps containing old PCB,s :)

I still like to use lacquer on bare boards..

Regards
M. Gregg
 
All of my tube designs use PCB's.

Modern PC boards that are properly designed and laid out don't have the problems that the old stuff had. I have not seen a PCB related failure in one of my designs yet and I have been making them for 20+ years.

My high voltage designs use PCBs as well.

I tend to run a prototype of a new design in my garage fab (toner transfer, no solder mask or conformal coating). Then, once the design is done, I have a board made at a PCB board house. With my prototypes, it is very easy for a small solder ball, a piece of wire strand, or even just flux to cause trouble. The same would be the case for vero board setups. This means that one has to pay attention and clean the boards thoroughly before applying high voltage. I've never had issues with the boards I've had fabricated at the board houses. The solder mask is really nice to have. But, of course, I do pay attention when I lay out the board.

I'm in the Seattle area where the humidity seems to be around 65 % year-round. Probably higher in the winter. Note, though, that flux is hydrophilic - i.e. absorbs water. So clean that stuff off (I use FluxOff).

It's very hard to beat a properly designed PCB on ease of assembly.

~Tom
 
Last edited:
Its just nice to remember the fun days of track repair

Yeah, I can remember what Coca Cola does to a live amplifier. The PCB had several large holes and was held together with epoxy when I was done with it, but it was alive.

The old Delco (General Motors) car radios had double sided PC boards with eyelets since through hole plating didn't exist yet. They were bad, but then they started screening the resistors right onto the PCB with carbon based ink. Pure Evil! Find me a GM car from the late 50's or early 60's that doesn't leak rain around the windshield. Old age test.....do you remember the Wonderbar? (probably not common outside the US)

I had far more problem with tin plated connectors in high heat and humidity than with PCB's though. Motorola Quasar TV's and most every solid state Fisher stereo had intermittent connectors.
 
The last tube based TV we had was a Setchell Carlson from around 1965 or 66. It was modular with about 6 or so small metal boxes that housed circuit boards, that plugged into a base.

I got it from my folks in 1977 and it was still going when I left college in 81. The only problems it ever had were tube related. As I remember, the PCBs were better than average.
 
It occurs to me that rust-enamel is cheap by the pint or gallon at hardware stores. And it might prevent corrosion that would contribute to tracing.

Has anyone tried something like John Deer Tractor Paint?
If its actually non-conductive and seal off contacts against 'through the air' arcing, it might be a good experiment,
especially in humid tropical environments.
Maybe it would have as high a voltage rating as that spray-on lacquer.

I used be able to buy a 30,000 volt spray enamel (clear),
but nowadays, nobody is carrying anything like this,
only overpriced contact cleaner that doesn't actually work on pots.


...
 
Has anyone tried something like John Deer Tractor Paint?

Haven't tried regular paint in years....lots of years. I found back in the 60's that the pigment in many reds is iron oxide (rust) and many blacks is carbon black. Both were conductive at high voltages. These comments were made based on observations from a big Tesla coil experiment in Florida humidity back in 1969. Voltage levels used were in the 20 KV range.

Modern paints probably use synthetic pigments, but I would test before screwing up a good circuit.
 
Status
This old topic is closed. If you want to reopen this topic, contact a moderator using the "Report Post" button.