Vacuum Tube Conditioning, Rejuvenation, Fusing?

tvr2500m

Member
I've been having some peculiar vacuum tube adventures lately. I'd appreciate thoughts and insights on this.

As I suspect at least some other people also have, I've amassed a collection of tubes over the decades. I'm making a committed effort to sort out a life of vacuum tubeage. This requires that I go through all the vacuum tubes I've got. To aide in this, given the sizeable scope, I built a RoeTest V11 vacuum tube testing system. It looked like a better tool to manage the production level required, rather than my vintage testers (mostly a TV-7D/U and Heathkit TT-1).

I've bumped into some curiosities, mostly with the 6080s and 6AS7Gs I've acquired in support of my much loved Crack. My typical protocol when acquiring any tube is to test it with at least one tester, often two. If it doesn't show any terminal defects, after testing, I then give it a check in-circuit. And I do this as soon as possible after purchase to bring up any issues with the seller. There are some rare exceptions when I do only part of that, and rarer, none of that.

I went to sort through 20 NIB French Thomson 6080WAs, and also to enjoy rolling one or two of these into my Crack. Most I had previously double-tested (they had test results written on them in Sharpie pen), with some also checked in-circuit.

I checked each one with the RoeTest. All tested well, and test results corroborated closely enough with tests done previously with the vintage testers.

BUT, I then tried one in the Crack. Nothing. No output. Took it out, tested it. Heaters heated, but little to no output. I went through all of them. Of the 20, there are 10 good ones, a couple useable/marginal ones, and the rest are dead. ??!! Crack tests just fine. I went through resistance and voltage checks, and all's well there. The other 6080s and 6AS7Gs I have out and on-hand to use in the Crack all tested and worked just fine.

This leaves the issue apparently and squarely with Thomson 6080s. PB being his helpful, responsive self reported to a post I made about this that he knew of some issues with eBay-sourced Thomson 6080s.

I had never seen a situation like this before where so many of the same brand, apparently NIB tubes show the same kind of terminal problem. I had some Tung Sol 6550s that had a tendency to show grid current draw more often than any other 6550 or KT88 I used. So, some tube brands, vintages, construction types might show some different, notable behavior tendencies, and I have heard some tube sellers, people who see far more tubes than I do, report some behaviors that show up from certain manufacturers, tube batches, etc., but I hadn't seen this happen in my own tube population.

Chasing this further, I posted my experience to another group. Got some interesting replies. A report that vacuum tubes have internal fusing and that all vintage tubes (kinda most specifically referring to NOS) need conditioning before use, like up to a week of continuously running only the heater before any other voltages are applied.

I know something about tube production process, activation, conditioning, and related downstream to this, rejuvenating or cooking. I've heard of ways cathodes can be damaged; stripping, poisoning, contamination/oxidation, mechanical issues. There are lots of ways a tube can fail/stop working, on rare occasion this can be quite dramatic with sparks, bright orange plates, nasty noises, and more. It can be due to issues with the gear, but in the cases I've had problems it's been due to a tube defect.

I've not though seen, read, heard anywhere in my tubed audio journey that a careful post-production conditioning, or re-conditioning, is essentially necessary. Nor, that tubes have internal fuses. Some tubes have connection wiring that provides a flash of start up resistance. Sometimes some of these wiring connections can come unconnected. The Conrad-Johnson amplifiers I had had plate fuses. Tubes will change their performance over time. Sometimes tube performance improves or stabilizes with some use. Sometimes it gets worse. Ultimately, they all fade away.

The RoeTest allows you to set and measure any tube test parameter independently, and has a feature to create automated rejuvenation programs. I get how this works, but haven't worked through this in detail. Some of those near-death Thomsons might be candidates to experiment with this.

Rejuvenation? Cooking? Internal built in fuses?

Thoughts?

Thanks.
 
Last edited:
I can confirm that old tubes (and many new-production examples) need to be activated. I believe this is consistent with the physics/chemistry. The cathode coating is activated by running current through the tube, often with a higher voltage applied to the filament or heater at first. This causes some metal atoms to migrate to the surface where they can emit electrons easily. It's not surprising that long storage without applied voltage would let random thermal agitation (Brownian motion) would let the material revert to a uniform distribution of the metals.

I did some measurements of early Sovtek 300Bs in an amp, finding that they ran very low currents at first but woke up slowly over a week of continuous operation. This was 20-30 years ago and does not apply to current production. I also know of a case where grid-cathode arcing at startup led to a dead tube, which was restored by the same treatment (after resolving the arcing issue). And we have seen many cases where NOS tubes need a few hours or days of operation to work correctly.

It seems to me that, if your Thomsons are truly old-stock and haven't been even tested for 70 years, they could be restored this way. IMHO you probably don't need to elevate the heater voltage, but I would suggest grounding the grid to avoid possible arcing and/or poisoning before the emission gets high enough.

It's still a puzzle that your emission testing did not find this problem. I wonder if the plate voltage is applied momentarily, allowing the space charge to build up before the test?
 
I went back though your post and I didn't see any mention of specific plate, grid or heater voltages for these problematic tubes when plugged into the Crack. Do the working tubes measure the same as the non-working tubes in the Crack?
 
I can confirm that old tubes (and many new-production examples) need to be activated. I believe this is consistent with the physics/chemistry. The cathode coating is activated by running current through the tube, often with a higher voltage applied to the filament or heater at first. This causes some metal atoms to migrate to the surface where they can emit electrons easily. It's not surprising that long storage without applied voltage would let random thermal agitation (Brownian motion) would let the material revert to a uniform distribution of the metals.

I did some measurements of early Sovtek 300Bs in an amp, finding that they ran very low currents at first but woke up slowly over a week of continuous operation. This was 20-30 years ago and does not apply to current production. I also know of a case where grid-cathode arcing at startup led to a dead tube, which was restored by the same treatment (after resolving the arcing issue). And we have seen many cases where NOS tubes need a few hours or days of operation to work correctly.

It seems to me that, if your Thomsons are truly old-stock and haven't been even tested for 70 years, they could be restored this way. IMHO you probably don't need to elevate the heater voltage, but I would suggest grounding the grid to avoid possible arcing and/or poisoning before the emission gets high enough.

It's still a puzzle that your emission testing did not find this problem. I wonder if the plate voltage is applied momentarily, allowing the space charge to build up before the test?
I appreciate the replies! Yes, often after some use some tubes can wake up and stabilize. And this is leading me to want to try some things to reactivate the tube.
 
I went back though your post and I didn't see any mention of specific plate, grid or heater voltages for these problematic tubes when plugged into the Crack. Do the working tubes measure the same as the non-working tubes in the Crack?
I didn't test Crack with the problem tubes. The measurements I did were with tubes that were/are working. Tubes I've been using in Crack lately. I would be happy to run some checks with problem tubes. I'll do that.
 
Back
Top