PC Board Tube Socket Cutouts

[CJ]

taking apart an Orange amp i was surprised to find that some of the tube sockets had cutout slots to help out with HV creepage,

on the power tubes there is a single slot between heater pin 2 and plate pin 3

and someof the preamp tubes also,

 

[Matador]

Can that really make all that much of a difference?

FR4 has a volume resistivity of > 1x10^6 MOhms per centimeter (or something on the order of 100 Gohm per mm). Soldermask can be several orders higher.

So how much leakage current could flow between those two points, that an air gap solves?

I have seen arc-over when an amp was operated full bore into no load, however that damage was contained within the socket...the PCB was fine.

 

[john12ax7]

Maybe the concern is lead free whiskers?

 

[Audio1Man]

GOOD ENGINERING. Poor cleaning, dust, hand oil, tin whiskers and high voltages can cause early failures and poor/bad MTBF. I want the product not to fail during warranty period.

Duke

 

[Matador]

Maybe the concern is lead free whiskers?

Perhaps! Although tin whiskers are free to grow through air, so I'm not sure why an air gap helps. Also, I can't see a tin whisker withstanding being on both sides of a plate -> heater potential (at least not for more than a handful of nanoseconds).

GOOD ENGINERING. Poor cleaning, dust, hand oil, tin whiskers and high voltages can cause early failures and poor/bad MTBF. I want the product not to fail during warranty period.

Are any of these real problems though? I've not once seen a tube failure on a PCB due to hand oils or dust (or tin whiskers for that matter). FR4 TG130 is rated at (conservatively) 300V per mil (from a dielectric point of view).

However if it's just a hedge against some unknown/unlikely scenarios I guess that might make sense.

 

[john12ax7]

Curious if anyone has the inside scoop, was this addressing a known problem, or a let's just do it as a potentially good preventive measure.

 

[musipol]

The slots might help if beer is spilled near the power amp pcb. I have seen this happen on a couple of amps. One was a Peavey Mace. I ended up trashing the power amp pcb and hardwiring the sockets to fix it.

 

[CJ]

i have had to drill holes in pc boards after they have become conductive, there was a bad batch of Marshall amps that had this problem,

i would like to see a notch between pin 4 and 5 on the power tubes as this is where most of the arcing occurs.

but there is AC on pin 3, not much on pin 4 due to filter cap,

might be some cooling involved but hard to see that on a 12AT7,

tube amps do develop condensation when being transported in the snow after a hot gig,

 

[dai h.]

hi, (not an engineer or any sort of expert on anything, but) I thought of humidity (monsoon season), and punching "pcb, high voltage, humidity, effects on creepage" into google, gives results like this:

https://resources.altium.com/p/high-voltage-pcb-design-creepage-and-clearance-distance
https://resources.pcb.cadence.com/b...-clearance-standards-in-circuit-board-layouts
(from the aforementioned: )

• Creepage will require another solution, as putting parts on the other side of the board won’t give you the separation you need through the surface of the board. Instead, you may want to install insulating barriers or cut grooves or slots into the board. The change in the contour of the board will increase the distance over the surface that the voltage has to travel, giving you the creepage separation you need

(I guess that explains the grooves in power tube socket bases and those (sort of) ridge/barrier things between some octal tube socket pins.)

Also (though this seems more of a clearance than creepage issue), made me think of this:

https://web.archive.org/web/20221002004032/http://www.chambonino.com/work/miscguitar/misc8.html

 

[CJ]

Nice articles!

AC has peak voltage to deal with as well as git freqs up near 1 kc which creep like a nun,

 

[Youngwhisk]

Nice articles!

AC has peak voltage to deal with as well as git freqs up near 1 kc which creep like a nun,

But only on Sunday mornings....

 

[dai h.]

re: the chambonino (Champ Electronics) site, it looks like it disappeared quite recently which is too bad, since there was interesting stuff there (though thankfully some (all?) saved in the wayback machine.

re: environmental contaminants, here's an example of what can happen to a PCB in a chassis (Mesa/Boogie .22 caliber plus) with lots of holes (for venting --I guess the dust got baked on after accumulating?) :





Maybe smoke is something to think about also (cigarettes, smoke machine, spliff smoke, etc.)?

I think old electronics were "tropicalized" (had varnish coating)--you see these old boards with a golden hue over them. Also I think lightning was more frequent in the tropics (Mesas had Varistors ("V130LA20A"?--what I remember from their brochure). Pete Traynor(?) I understood he used to throw his amps off a music store roof to test their survivability. Tubes would break but the amp would still work (after they were replaced). Anyway, might be many aspects to account for to make something that can survive a world tour.

 

[CJ]

Pre CBS Fender dipped the boards in wax,

Clear coat protects but makes things nasty if you have to solder in new parts on a PC board.

 

[Matador]

Ugh I would not take any PCB technique from Mesa Boogie in the 80's or 90's as anything resembling sound PCB layout or construction practices. . Like their "ocean wave" soldermask that always ended up looking like an impressionist painting of heat waves, or PCB tracks that liked to spring up and try to run out the back of the amp.

Maybe I can refine my curiosity: given a modern standard material (like FR4-TG130), and a new polymerized solder mask, why are such cutouts needed? Even by IPC-2221 standards (which are extremely conservative), and no solder mask, voltage creep exceeds 500V per 100mils, and even with extensive annular rings on an octal socket, there are over a thousand volts of margin even between the plate and heater (pin 2 and 3), or screen (pin 4) to grid (pin 5).

 

[dai h.]

don't know enough to answer, but, do those characteristics maintain over time with the heat applied from the tubes?

and also, could there be another purpose (assuming one reason is creepage) for the slots. Maybe to help with cooling (let some of the heat escape so it isn't all trapped on the PCB underside)?