Tube creep

Thread Starter

nsaspook

Joined Aug 27, 2009
8,278
We were parting out some old equipment for the control boards when this tube was found inside (the PS was still functional). It was mounted horizontally with a bracket to hold the anode cap tight in a 5kV shunt regulator suppression supply from about 1990 with a dead fan so I think it got a little hot over that time. :eek: (leaded glass tube 6BK4C for X-ray shielding) Yes, tubes still make transistors.
http://www.radiomuseum.org/tubes/tube_6bk4c.html

 

#12

Joined Nov 30, 2010
18,217
Interesting. I've never seen a tube sag before.
I've seen them hot enough to frighten me, but I haven't seen them sag.
Seems to me, if they got hot enough to sag, the envelope would suck in.
That isn't a symptom of broken glue between the glass and the plastic base?
I've seen that.
 

Ramussons

Joined May 3, 2013
940
I totally agree with #12. If the glass envelope of a Vacuum Tube got hot enough to "creep" it would crush the insides :eek:. What the photo shows is only a failure of the glue between the Plastic Loctal Base and the Glass Enveiope.
 

Thread Starter

nsaspook

Joined Aug 27, 2009
8,278
Nope, look closely. The cathode and anode are clearly offset. The base is still almost inline with the socket but the anode is offset at an angle. It looks to be something like 'cold flow' over a long period of time with a shear force only near the cathode that would push that area of the glass envelope slightly outward from the compression forces above and below that region.
http://www.nj7p.org/Tube/SQL/Tube_query.php?Type=6BK4C

https://farm4.staticflickr.com/3798/19295701440_6d789d05ab_o_d.jpg
 
Last edited:

Kermit2

Joined Feb 5, 2010
4,162
some areas appear to have drops/splatter which look eerily similar to regular solder alloy. If it was key in the physical anchoring of the structure I can understand how it might sag at high temps when mounted horizontally.
 

alfacliff

Joined Dec 13, 2013
2,458
solder is not used inside vacuum tubes. the elements are welded together. there might have been some of the "getter" material splattered inside. tubes have been know to sag when mounted horizontally, like the 811 tube, the 811A was produced because the heater sagged into the grid when hot. and the grid sagged into the plate when heavily loaded.
Cliff
 

ian field

Joined Oct 27, 2012
6,539
Interesting. I've never seen a tube sag before.
I've seen them hot enough to frighten me, but I haven't seen them sag.
Seems to me, if they got hot enough to sag, the envelope would suck in.
That isn't a symptom of broken glue between the glass and the plastic base?
I've seen that.
During my time repairing microwave ovens, I sometimes entertained myself by nuking any spare valves laying around.

If you heat glass to red hot, it becomes electrically conductive - so once you get it started it will continue to cook in a microwave.

In a microwave, the gettering on the inside of the glass can get it up to temperature - I've had the whole glass melt and the vacuum sucks it onto the internal structure like cling film.

Glass fuses that have blown violently and metallised the inside of the tube can also be melted.

Valves can also suffer thermal runaway - occasionally I've seen published photos of hotspot damage on the glassware.
 

ian field

Joined Oct 27, 2012
6,539
We were parting out some old equipment for the control boards when this tube was found inside (the PS was still functional). It was mounted horizontally with a bracket to hold the anode cap tight in a 5kV shunt regulator suppression supply from about 1990 with a dead fan so I think it got a little hot over that time. :eek: (leaded glass tube 6BK4C for X-ray shielding) Yes, tubes still make transistors.
http://www.radiomuseum.org/tubes/tube_6bk4c.html
What sink current do you need? - the European PD500 can handle about 1/2A, and its rated 25kV - but it has a peculiar heater voltage because its designed for a 300mA series heater chain. Presumably it has a 6.3V heater ED500 counterpart - but I've never seen one.
 

Thread Starter

nsaspook

Joined Aug 27, 2009
8,278
What sink current do you need? - the European PD500 can handle about 1/2A, and its rated 25kV - but it has a peculiar heater voltage because its designed for a 300mA series heater chain. Presumably it has a 6.3V heater ED500 counterpart - but I've never seen one.
That's very similar to the Japanese version of the 6BK4C. (I'll post a picture later)
Japan version:

A short video of the creepy tube.
 
Last edited:

ian field

Joined Oct 27, 2012
6,539
That's very similar to the Japanese version of the 6BK4C. (I'll post a picture later)
Japan version:
A short video of the creepy tube.
That's very similar to the Japanese version of the 6BK4C. (I'll post a picture later)
Japan version:
You could always series connect 7,143 forward biased silicon diode junctions - but you'd probably have to derate them because of heat buildup.
 

Thread Starter

nsaspook

Joined Aug 27, 2009
8,278
solder is not used inside vacuum tubes. the elements are welded together. there might have been some of the "getter" material splattered inside. tubes have been know to sag when mounted horizontally, like the 811 tube, the 811A was produced because the heater sagged into the grid when hot. and the grid sagged into the plate when heavily loaded.
Cliff
That 811A is a great tube. I made CB band SSB RF amps from those in the early 70's. A very easy tube to neutralize and match with a simple PI output network. Lots of power if you let them glow a bit.
 

#12

Joined Nov 30, 2010
18,217
I have wondered, If you blow some air across a big tube, will that increase its power capability, or is the plate so insulated from the glass by the vacuum that it doesn't make much difference? From what I gather, the answer is, "Sometimes, depending on how the tube is built and what it's used for". Some gigantic tubes are water cooled. :eek: I'm talking mostly about octals used in audio amps.

Judging from this saggy 6BK4 tube, it does make a difference.
 

alfacliff

Joined Dec 13, 2013
2,458
air cool the seals and glass. look at the ratings for tubes in the tube manuals. Hans have been getting the most out of tubes for years, a friend used to run 1500 volts on the plates of a couple of 1625's in cw mode, a long dash would make them really light up.
 

#12

Joined Nov 30, 2010
18,217
air cool the seals and glass. look at the ratings for tubes in the tube manuals.
I own and have read the RCA Receiving Tube Manual RC-19 (published 1959) and Essential Characteristics by General Electric (can't find a publishing date), but I don't remember seeing a word about forced air cooling.
 

k7elp60

Joined Nov 4, 2008
562
FPN-42 XMTR.jpg I used to work on some high power transmitters. Output was in excess of 300KW on the peak of the pulse. The final had 8 vacuum tubes. two banks of 4 in parallel and the two banks in push-pull. The filaments were approximately 16V @ 175A. They were triodes and we changed the filament voltage to control the plate current. The plates were external finned and they were mounted on a round ceramic insulator. The HV was about 13KV and the plate current averaged about 1A. If anyone remembers Loran-c, these were some of the transmitters. The carrier frequency was 100khz and they transmitted 8 pulses seperated 1000usec apart at about every 50,000usec. The tubes are glowing, you can see the relative size of the tubes.
 
Last edited:

Thread Starter

nsaspook

Joined Aug 27, 2009
8,278
I own and have read the RCA Receiving Tube Manual RC-19 (published 1959) and Essential Characteristics by General Electric (can't find a publishing date), but I don't remember seeing a word about forced air cooling.
Just cooling the glass won't increase the avg power rating very much but it did help with the peak rating by keeping the external RF tuning components cool. Our #1 failure item wasn't the PA tubes, it was the high current RF capacitors that fried.
 

alfacliff

Joined Dec 13, 2013
2,458
the recieving tube manual should have the ratings of the usually kovar to glass seals. too hot and they crack. and a recieving tube manual probably wont tell you how to make a low power tube handle more power than the mfgr rates them for. if you look at the ratings for medium power tubes, they usually have ratings for different classes of operation, such as ICAS, intermittant commercial and amature service, a lower duty cycle than CCS or continous commercial service, icas allows more power by reducing the time the tube is heating.
 

Thread Starter

nsaspook

Joined Aug 27, 2009
8,278
I decided to rebuild the PS as a mostly compatible spare for newer equipment after finding some NOS tubes at a good price. (The interface control board needs an update)



With a new fan she powers up just fine with the control signals jumped.

Clean-room service, no dust in 20+ years.
 
Last edited:

Thread Starter

nsaspook

Joined Aug 27, 2009
8,278
Another powery supply unit with Russian tubes is having the same problem with tube creep. One machine started have problems with acceleration electron suppression. The problem was tracked to lower than normal suppression voltage. Down to 4.2 instead of 5.0 KV. Open the supply and this is what was found. :eek:


Offset tube elements again.




At least with the Russki version I can get new ones. ;)
 
Top