diyAudio (
-   Tubes / Valves (
-   -   What is the definition of a 'good tube'? (

andyjevans 6th April 2009 09:55 PM

What is the definition of a 'good tube'?
I always thought 'good' in tube testing terms was 60% and over.

On my AVO panel meter, Good (green range) starts at 5.5mA out of 10mA

On a Funke W19 tester Good range starts at 4mA out of 10mA

These are wholly different definitions of what constitutes a 'good' tube.

Is there any accepted definition?

I would be really grateful for all opinions, feelings, facts and definitions!!! In particular, what should a tube seller sell as a 'good' tube - or does it depend on his tester?

andy evans

Poindexter 6th April 2009 10:12 PM

Andy, the definition is your definition.  When I buy tubes, I buy a lot of a type (sometimes hundreds), and I have a breadboard tester that I can quickly configure for different types.

I can test for any and every param, and I match 'em up for what I want.  If you think about it, you can use a tube that's 'weak' in emission or something if that's application optimum.  I intentionally use tubes for the 'top' (active load) of a SRPP that have higher gm than the 'bottom' (voltage amp) tube, because it gives a better balanced circuit.  For a diff-amp, I want very similar native operating point (so the current balances), and gain into load.

Don't let webgeeks like me tell you what a good tube is; it's your call.



Hearinspace 7th April 2009 12:04 AM

I like Poinz's view. More relaxed and adaptive is good I think.
When a tube enters end of emissive life the current at a given Vp-Vg will start to go down and then that degrades further rapidly. (ie. there is a knee in the lifetime emission curve) The trouble is that some tubes at the beginning of life can be low emitters for a while and then start to improve in performance. Then there are really just sub standard tubes that don't do what they're supposed to.
I've had many 'bad' tubes that I figured out how to get a reasonable use from, but others that were nasty enough that they couldn't be trusted not to take a valued piece of equipment with them if they went, so I just pitched them.
While many NOS tubes will test better than book spec for new, I don't freak out if they're a bit low.
6N1Ps seem to have a relatively short sweet sounding life - in my amps they ran for something around a year before they started to call attention to themselves.
About your 4mA out of 10 - I bought some supposedly NOS Valves that had come out of the British military from WWII. With respect to book specifications they tested miserably but the boxes had stamped on them "Serviceable" - meaning I believe that they were basically crappy but if the gerries were raining artillery down on you and you needed a valve in your wireless to call for help then well, they should do well enough . . . . . .

andyjevans 7th April 2009 12:16 AM

I'm interested in general views about this.

My point of departure was being sold some tubes tested on a Funke W19 where the GOOD scale starts at 4mA out of 10mA.

This seems pretty low to me, but it plainly says that on his tester.

I've asked colleagues what they think and the results are:

'My personal view is that gm is the value that should be used in assessing whether or not a valve is good. I have as some of you may know, measured many valves (when I had my AVO 163). Even new valves varied widely in terms of mA passed under standard test conditions. gm is arguably a better measure of "Q" or "goodness." Of course, some of these ancient types have very low gm even in perfect condition though a good valve tester should be able to provide suitable ranges for these "antecedents!"

I used to define a "good" (i.e. usable) valve as one which delivered 70% of gm - and incidentally - of emission! I know that Rod Burman and John Giacomelli operated on a similar basis. (These are two big English tube dealers)

And also:

"Normally minimum good is about 1/2 to 2/3 of nominal Gm value and I guess current."

And another view:

"I have a Weston 981 III (transconductance) that provides a "reject point" for a particular tube. By looking at published tubes specs and doing a little math, the "reject point" on this tester is 65% of the published Gm spec for a particular tube."

So we have
1) 70%
2) 50-60%
3) 65%

These are approximately the definitions I am used to



Hearinspace 7th April 2009 02:31 AM


My point of departure was being sold some tubes tested on a Funke W19 where the GOOD scale starts at 4mA out of 10mA.

This seems pretty low to me, but it plainly says that on his tester.

Andy , I agree with you that it's low,
If you take a look at Philips and Telefunken data sheets where end of life specs are given it seems clear (to me anyway) that 4mA out of a possible 10mA would be considered bad. For example, for the E180CC, Philips gives a current of 6.3mA -10mA with end of life pronounced at 5mA with S for good at 5.3 - 8.1 and end of life at 4mA/V. The Telefunken sheet for the ECC803S gives good at Ia =1.25mA and S = 1.6mA/V and dead at Ia= 0.8mA with S=1.05mA/V.

On the other hand, while I wouldn't be happy if they ended up in my lap either , it's hard to find fault with a seller who genuinely believes the meter when it says good and sells them as such.

And just an opinion,I don't think a typical tube tester is worth anything anyway. (or at least not much) The only way to know if a tube is going to work the way you want is to test it where you plan on running it. In my experience so far, building a jig or adjustable circuit (like Poindexter) or using one of the fancier completely adjustable curve tracer sort of testers is the only way to go outside of just building and measuring. For single ended triodes a pot adjustable current source on the plate and a rheostat for Rk to fish around for the sweet spot is just dandy!

SpreadSpectrum 7th April 2009 05:43 AM

If I'm not mistaken, as a tube ages, the rp goes up but mu is not affected much. I guess one's threshold for 'badness' could depend on whether a low rp is essential to the correct operation of the circuit.

coresta 7th April 2009 10:51 AM

Hi toobzlovers! Many other parameters must be seen in order to evaluate the "real" state of a tube : I use a Metrix U61b that is the Rolls Royce , or the Bentley of tubetesters :) This one show the grid current , and vacuum , especially useful for low signal tubes, emission ability : For example , i always test E188CC in a short "saturation" burst so that i can see if is able to drive up to 40mA , and not only the 5mA ... You can see the real max Pa acceptance of any toob to get the red on plate , as George likes to do ;) and so on .... :) Cheers Pierre

andyjevans 7th April 2009 12:02 PM

Here's another view:

"For me, to test good means that it is still meets the original spec. If
it's doesn't then it is, for all practical purposes, a different valve."

And another:

"Generally I think that any valve that passes 50% of its rated current at the standard test point is considered as "working" My statement about 70% - especially of gm - is really a reference to what I would prefer to use in an amplifier I have built. After all when one calculates resistor values one does expect certain voltages at given points!"

And another:

"With DHT on some testers you need to take the filament voltage into account as one side is connected to 0V internally , for medium or high mu types this will seriously affect the target
result , ditto for filamentary rectifers such as 5U4 or 5R4 . Any tester that measures in mA and nothing else is junk . A target test value should at least be in mA/V or mhos (transconductance) and also , preferably with power valves anode current . The transconductance is the best way of determining how good the valve is . For me good starts at 66% for both anode current and transconductance , I've sold upwards of 20,000 valves and have very rarely been pulled up over test results ."

Conrad Hoffman 7th April 2009 12:40 PM

FWIW I still use my old Tek tube scopes. The manuals were very specific. A good tube is one that works in the circuit. The results of tube testers were not to be taken as indicators of a good tube. They recommended substitution as the primary troubleshooting technique. I can confirm this for some of the plug-ins, where selection of matched tubes is particularly fussy.

pointy 7th April 2009 01:55 PM

a bad tube is the one in a batch which has a different label and you can't gm marry it with any other tube from the batch.
(for stereo you know)

the tester i had was a avo mk3
a big gray thing the one with the roll knobs

All times are GMT. The time now is 06:33 AM.

Search Engine Optimisation provided by DragonByte SEO (Pro) - vBulletin Mods & Addons Copyright © 2017 DragonByte Technologies Ltd.
Resources saved on this page: MySQL 18.75%
vBulletin Optimisation provided by vB Optimise (Pro) - vBulletin Mods & Addons Copyright © 2017 DragonByte Technologies Ltd.
Copyright ©1999-2017 diyAudio