Asking Valve questions showing ignorance again.

[AdrianH]

I am getting confused by a few terms in testing valves, one is 'Life test'

From what I can gather this is a drop in heater volts with a check to see what the drop in emission is, so say drop the 6.3 10% and see if the anode currents drops by some figure? Think I have it correct, but why would it be a reasonable measurement as opposed to say checking settings for anode current against published specs?

There used to be some TV CRT rejuvenators out there, I think even circuits to make your own that heated the cathode and put current between cathode and grid using a pygmy bulb, some you won some you lost, does the same get done for tubes?

Gas test, this seems to be a test done by sticking something like a 1Meg resistor in series with the g1 pin, I guess if no current draw or very little is taken by the grid then the difference in plate current readings with 1M and a low impedance grid driver should be very similar, if the grid is taking any current possibly due to the tube loosing vacuum then the anode current would change considerably.

To be honest I would have thought that a loss in vacuum in a valve would drop the anode current as it would be harder for electrons to flow, but I must be wrong in that assumption as texts seems to suggest it can increase anode current.


So is my thinking in the correct direction? And does anyone flash tubes to try and recover duff ones?

This is all because some 'good life tested' valves seem to be less than 50% down on emissions compared to specs, so I am trying to understand all the different terminology found in testers.

Adrian

[Bazz4CQJ]

Just to deal with one of your queries Adrian, I don't think I've ever heard the term "Good Life Test" before. However, the practice of looking at how a valve responds to a drop in heater voltage is something that's been around for a time, and I've even tried it myself. The argument is that a valve with a really healthy filament / cathode will still work reasonably well if the voltage is reduced a little, though a tired valve may show an appreciable drop in performance. Might be useful if you can compare a new valve with a used valve.

I think that general experience is that the performance of "new" valves (previously unused) can vary a heck of lot, some being very disappointing. You even see significant difference between the two sides of a new double triode inside the same envelope.

B

[AdrianH]

Sorry Bazz, probably me again doing posts around midnight and later, the test was named 'Life' and the stated results were Good.

But what I can not understand is that if the valve is low emission anyway, but still only drops a few percent when the heater is dropped a bit, then the Life test would still show good even on a low emission valve.

So I had a play and flashed one of my valves a Russian EL84 which at Anode and G2 of 250 Volts 'ish with a g1 of -7.3 Volts should be 48 mA anode current, this was originally down at 20mA, after a play came up to 24mA, still nowhere near, but a slight improvement.

Adrian

[GMB]

I would have thought that a loss in vacuum in a valve would drop the anode current as it would be harder for electrons to flow

But what happens is that the gas gets ionised (in a cascade process) and the ions are additional current carriers so it gets easier for current to flow. Think of a neon tube - once it strikes it effectively exhibits negative resistance since extra current generates more ions which allows more current to flow etc., hence the need to control the current with a resistor!

[David Simpson]

Adrian, there is no hard & fast rule about how much Ia drops in relation to a slight drop in Vh. But this phenomenon does occur. Even on reputable valve testers. One might select 6.3V on an AVO VCM, but with o/p valves drawing 1 or 2A or more Ih, a drop can be observed across H+ & H- of the order of .1 or .2 or .3 of a volt below 6.3V.
If folk buy job lots of valves on the internet or at junk sales, etc. there is no guarantee that they are not "Pulls", even if they are in an original colourful wee manufacturer's box. I.e. valves cast aside by radio & tv engineers, years ago, who just shoved in a new valve to keep some housewife happy so that she can watch Coronation Street that evening. But those old boys never liked dumping gash valves & shoved them in an old cardboard box under a bench somewhere.
I too have never heard of "Good Life Test". If a valve's Ia is significantly lower than the book spec., with all the correct voltages applied, then its knackered. Low Ia, means a shallower Ia/Vg slope, which means the mA/V will be well below spec.
"Good life test" sounds like some mercenary valve wheeler dealer's speil, just like "AVO Tested" in some eBay adverts.

Regards, David

[AdrianH]

Originally Posted by GMB

But what happens is that the gas gets ionised (in a cascade process) and the ions are additional current carriers so it gets easier for current to flow. Think of a neon tube - once it strikes it effectively exhibits negative resistance since extra current generates more ions which allows more current to flow etc., hence the need to control the current with a resistor!

Good grief yes, that's like my plasma cutter, small arc ionises the air suddenly lots of carriers lots of amps melts steel. Should have thought of that, thanks for the pointer.

Originally Posted by David Simpson

Adrian, there is no hard & fast rule about how much Ia drops in relation to a slight drop in Vh. But this phenomenon does occur. Even on reputable valve testers. One might select 6.3V on an AVO VCM, but with o/p valves drawing 1 or 2A or more Ih, a drop can be observed across H+ & H- of the order of .1 or .2 or .3 of a volt below 6.3V.
If folk buy job lots of valves on the internet or at junk sales, etc. there is no guarantee that they are not "Pulls", even if they are in an original colourful wee manufacturer's box. I.e. valves cast aside by radio & tv engineers, years ago, who just shoved in a new valve to keep some housewife happy so that she can watch Coronation Street that evening. But those old boys never liked dumping gash valves & shoved them in an old cardboard box under a bench somewhere.
I too have never heard of "Good Life Test". If a valve's Ia is significantly lower than the book spec., with all the correct voltages applied, then its knackered. Low Ia, means a shallower Ia/Vg slope, which means the mA/V will be well below spec.
"Good life test" sounds like some mercenary valve wheeler dealer's speil, just like "AVO Tested" in some eBay adverts.

Regards, David

I am not trying to get at any seller, and trying to keep off the Ebay subject before the thread is shut. I am just trying to understand some of the tests out there and trying to decide if they are valid to my mind or not, it may help me with my home made DC tester in deciding what options I will add.

From various Google searches such testers as the Heathkit TT1, Hickok 539, 600a and 752A plus I am sure lots of others have a life test. Text I have read on the odd forum posts seem to suggest that a Heater drop of 10% should not significantly drop the emission rate, but as one goes towards the end of a tubes life it would.

So I did a test myself this afternoon. with an old EL84, (I bought 8 to play with).

So with 250 Volts A and G2. -7.3 Volts on g1 and measured heater at 6.3 Volts, I had 23mA Ia.

A reduction of heater to 5.67 Volts dropped the Ia to 21.5mA which is 93% of first measured. So that test on it's own would suggest it is a good valve, but the valve is severely down anyway so meaningless, but yes I could see how finding a test that shows good on a tester can be used to help sell tubes ignoring all the other test functions. And if other new to the technology readers come across this post and learn with me, all is well and good.

Think I will add the 1Meg resistor to the grid input with a push to break switch, that seems to be a valid test to do.

I can test for shorts with heaters lit using a normal DVM.

Have fun.

Adrian

[Bazz4CQJ]

The question about when we should consider a valve to be knackered is an interesting one.

When I first got my AVO163, out of curiosity, I pulled out various valves from equipment that I deemed to be working well and was very surprised to find that many of those valves had Ia and Gm appreciably below the spec. Subsequently, I recall a a thread on here in which someone pointed out that good designers would create designs which were fairly resilient to fall off in performance of the valves (I think that was Wrangler).

The meter on the 163 has a scale which suggests replacement of a valve when the Gm falls by ~25% from book value...but a primary customer of those machines were the Services where lives could be at stake, and that might be a conservative standard. I don't have any hard and fast rules about the point at which I throw valves away; I think I probably cling on to them.

B

[MotorBikeLes]

Over a period, the cathode may get coated with junk. If the original vacuum was less than it should have been, and firing the getter failed to wipe everything clean, the the coating process could proceed. dropping the heater voltage would give a big drop of Ia. If however it was a well evacuated valve, with good getter action, the cathode surface would stay clean, and the same heater current reduction would have a lesser effect.
CRT rejuvenators could give a big improvement (IF correctly carried out) with the "dirty cathode" examples.
Some CRT tester manufacturers supplied "Life test assurances" which could be given to a customer after CRT rejuvenation. If it did "go down" after 6 months, an FOC repeat rejuvenation could be provided.
As I have mentioned previously, I designed my own after studying a few other testers. One manufacturer produced a reasonable tester, later "improving it" twice. The final version in my opinion was rubbish.
Les.

[AdrianH]

Les; I have broken this into 3 sections

Originally Posted by MotorBikeLes

Over a period, the cathode may get coated with junk. If the original vacuum was less than it should have been, and firing the getter failed to wipe everything clean, the the coating process could proceed. dropping the heater voltage would give a big drop of Ia. If however it was a well evacuated valve, with good getter action, the cathode surface would stay clean, and the same heater current reduction would have a lesser effect.

So from that, if I am reading correctly and jumping to my own conclusions! With a hard vacuum & working getter, loss of emission is not down to a dirty/poisoned cathode oxidisation? Possible oxide depletion or other? This could not be rejuvenated to any extent, but would show little drop in anode current with a little drop in heater, so for these valves the 'Life' test would be good although emission could be poor?

Yes/No/Other

CRT rejuvenators could give a big improvement (IF correctly carried out) with the "dirty cathode" examples.
Some CRT tester manufacturers supplied "Life test assurances" which could be given to a customer after CRT rejuvenation. If it did "go down" after 6 months, an FOC repeat rejuvenation could be provided.

So an indication that once done properly, the effects would last for some time!

As I have mentioned previously, I designed my own after studying a few other testers. One manufacturer produced a reasonable tester, later "improving it" twice. The final version in my opinion was rubbish.
Les.

OK so I do not know of your previous posts on this, but I have a couple of questions. From my brief look at the Mullard books on Picture tubes, they seem to me to be summarised as triodes with focusing and target anodes, but the overall Ia appears to be very small compared with say an EL84, or other output valves. So could a picture tube rejuvenator work with a valve, which (and this is only a guess as I have not broken them apart and compared) would suggest a valve such as the EL84 would have a much larger cathode surface area compared to a TV tube?

Can you point me to one of your posts with circuits to look at?

Can Valves be successfully rejuvenated? This last question I guess is no, otherwise we would not have a problem with old valves and everything I read suggests there will come a time when all valves will be duff.


I have a lot of understanding to catch up on, with many points that are not obvious to me. But I like valves for some reason.

Cheers

Adrian

p.s. The EL84 soviet valves I have are 6P14P generally just that although a couple have extra letters after!

[MotorBikeLes]

Questions questions. I am NOT an expert, but relying on some experience and some theory.
I think if the conditions of vacuum and getter are AOK, you have one of those valves that is still good after 80 years.
CRT rejuvenation. Let me tell you/remind you and others of a report by Les Lawry-Johns many decades ago in TV mag. He tried to rejuvenate a delta gun CRT, probably using his own design rejuvenator. Nothing happened, so he tried a different tactic. He connected the heaters to 12v rather than the correct 6.3v, and suddenly it responded to his rejuvenator and a good picture resulted.
This tells me that an elevated filament temperature gives better results.
Video Circuits made a number of testers, their V31, a V32 and (I think) a V34 (or was it V32A?). The early one allowed elevation of the heaters, the last did not. Connect that to one of the early Japanese in-line CRTs, and you would kill it quickly.
I suspect rejuvenation is only needed when there is too much "floating junk", but then a 26" delta gun CRT is a BIG thing, with plenty of volume and internal surface area to be PROPERLY cleaned up during manufacture. Throw in the adsorbtive effects of glasses at elevated temperatures, and it is a wonder any of them ever worked, but they did.
Testing is done (usually) as if a triode valve, with the first anodes used as the anode, with final anode not involved in the test.
When you test a typical triode on a valve tester, you normally supply the grid with a negative voltage. Most testers use a positive voltage. i suggest this gives erroneous results.
The tester I designed, built and used kept grid negative, but used a higher first anode voltage. I was much more confident with the results than with any other machine I used.
For rejuvenation, either a series "pygmy" bulb or a florescent lamp starter was used to control the pulsing current. Particularly for the Japanese inline tubes, this almost guaranteed a dead tube. My design used the florrie pulser but with current limited by series resistor. Heaters boosted during rejuvenation.
I recall a very expensive US made unit which bragged about switching off the heaters during rejuvenation. This horrified me.
I bought a muter tester, which involves the final anode during testing for something I was writing, but a long awaited change of fortune meant I never actually used it.
Comparisons with (say) EL84. Not sure you can make one. Forget the simple "Triode" aspect. We require an Ia of hundreds of micro-amps to light up the screen. That is at 25,000v, not 250v. I don't think comparisons are appropriate.
Can you rejuvenate a valve? I have not done so, but if you set one up for normal test, then boosted the heaters and left it to stew (don't do it on your AVO VCM, it is suggested they are unsuitable for soak tests!) then it may well pick up after a period. About 40 years ago, I bought some stuff from a guy who had just shut his TV shop. Amongst the stuff was a "Rejuvenator transformer", together with instruction leaflet. Basically,it did what was suggested in the previous sentence.
Previous posts? There was a thread, maybe search for Video circuits V31, or one on CRT testing. Try a search and see what you can find. I recall I posted a number of tester circuits including my own.
Les.

[David Simpson]

I reckon Adrian, that with many 2nd hand valves, you'll have to accept that there is no alternative to dumping them after observing a serious loss in Ia & Gm. Rejuvenation - nah nah - forget it. Soak testing - might do more damage than good. Especially, as Les says, using expensive AVO VCM's.
A year or so ago, I was given several large boxes of valves from a "silent key" clear-out. Hundreds of valves, & initial delving revealed KT66's & 88's, TT22's, & even a PX25. Wow - says me, eBay here I come ! No such luck, MK3 testing revealed about 90% were duff, including all those power valves. Obviously they were "Pulls", but never thrown away by the poor old chap who had passed away.
Some very old valves are worth keeping, even just for display purposes. Early 1920's B/E's, D/E's & Cruciform, etc.

Regards, David

[AdrianH]

Second hand valves I do not mind if they are cheap enough as I can use them for playing with and it means nothing is lost if I kill them in the process.

What I do struggle with when it comes to testing them, is the names given for all the tests and how misleading the results can be in isolation. You and many others on here will probably have gone through the things I am trying out and learning now.

I am building my own DC tester, I do not have an Avo or simple emissions diode style tester. The next major stage is building the HT side of it rather than use buck converters. Old and valuable valves I am not really after, but more the technology. At the end of it I am trying to see if a valve matches the points on the spec sheets under DC conditions, I should be able to move grid 0.5V below and above a point to measure gm in that area.

I have played at boosting/rejuvenating a valve, with very little change, in my very early 20's I bought and resold ex rental TV's that were based on the Decca Bradford chassis, I built a rejuvenator for them, basically a transformer, diode and pygmy lamp and generally had some success on bringing tubes back to life. Only very minor and obvious repairs were done and I can not say I really understood what I was doing at that point. Then I entered work properly working for Pye Telecoms in Ardwick and apart from swapping valves in Cambridge and Vanguard radios never really had to learn about them it was not cost effective to spend time on sets. Gear quickly moved to semiconductors and I have probably looked at valve tech for 40 years until recently.

Anyway I waffle.

I have downloaded Les' circuits from a post and will keep for reference, I think the 'life' test is carp! (very fishy), Gas test worth including.
So have probably got my answers for this thread, and I have another 70 valves (old ones) arriving soon to play with.

Adrian

[David Simpson]

Adrian, apart from the Gas test which you seem to have got your head around, the other preliminary tests are for continuity & insulation. Continuity = very low resistance between the Heater pins, and Insulation = maximum resistance between all valve electrodes = tens of Mohms up to Infinity(shown on many valve tester's meter scale as a figure of eight on its side). The likes of AVO testers can also test for this with the heater hot. As well as a multi-way switch to connect the test voltages to the correct valve holder configuration, you could, say, fit 9 Bannana plug sockets & use your bench multimeter & megger.
A decent 2nd hand heavy-duty DC LV bench PSU could supply Heater Voltages, and a much more modest DC LV PSU could supply -ve Vg. A modest HT HV PSU could supply Vs up to say 300V, and you'll need a thumper of a HT HV PSU to supply up to say 400V & draw a max of 100mA. All these types of PSU's I've seen for sale, over the years, in VMARS Sales Catalogues, and in the BVWS Bulletin Magazine, & I believe bargains can be had at BVWS Swapmeets, etc. I've also seen some decent stuff occasionally at ARS Junk Sales. You could, of course, build all your own power supplies. But, believe me, I've built several, its no easy task sourcing all the components, & the chassis & cabinet work can actually take longer than the circuitry work. Circuitry work - Jesus, that can be a sod to work out, looming & lacing, tag strips & tag boards, colour coding or ident sleeving all the different wires, and so on. If all that is something you haven't done much of, then I'd advise you to seek hands-on mentoring from someone down your way. Forum thread posts are ideal for discussing concepts & getting bits of advice here & there, but an online tutorial site it is not, I reckon.

Regards, David

[AdrianH]

Hi David, there is a post of mine I have just updated on the valve tester I am building over on the Homemade section. I do use a 2 amp Thurlby Bench PSU for heaters and another small bench PSU to power a buck converter for HT and bias. I have used the banana socket for testing with a DVM with heaters on and I was thinking at first I had a cathode to grid short until I reversed the DVM probes.

Adrian

[David Simpson]

Adrian, how does the small HT PSU provide both a selection of HT's for Va & Vs, & -ve Vg's via the one Buck Converter ? If the HT PSU is "small", can it give say a 100mA steady max Ia for couple of minutes or so, whilst studying the valve's performance ? Also, from medium sized Pentodes up to big thumping O/P valves, the Screen does actually draw a few mA of Is in the top half of the Ia/Vg curve.
I would recommend separate HT power supplies for Va & Vs, and certainly for -ve Vg. From what little I know of buck converters, they exhibit a "Sawtooth" O/P. One doesn't want that on a "pure" DC Valve Tester. Have you "scoped" your DC o/p's ?
I think I've already mentioned it, but my DC Tester's separate Va & Vs are supplied via a 2 pole Relay, controlled by a spring-loaded on/off toggle switch. This avoids you forgetfully leaving the valve running at full blast, and a quick "release" is a good safety precaution. Also, as well as the tester's Ia & Is being metered internally, I've put a removable link in the Anode wiring which facilitates external monitoring of Ia with a decent AVO8 MM(with its great magic quick acting "Cut-Out"). All "Belt & Braces" I know, but it safely avoids phooking a rare or expensive valve, particularly if its not yours.

Regards, David

[AdrianH]

Mods please close this thread.

Adrian