I have searched for answers to this question, but to no avail.
Now that I have a working Aikido using 6sn7's, I have been on the prowl for tubes and have come across all kinds of test results.
What do these results actually mean? Usually the sellers say that a good new tube would be say 2600 measured on a Hickock with other measures being in the middle 100's.
What are these measures? If 2600 is for a new good tube, then what would a tube of, say 2050 mean in terms of sonic quality / life-span, etc.
Thanks,
Charlie
Now that I have a working Aikido using 6sn7's, I have been on the prowl for tubes and have come across all kinds of test results.
What do these results actually mean? Usually the sellers say that a good new tube would be say 2600 measured on a Hickock with other measures being in the middle 100's.
What are these measures? If 2600 is for a new good tube, then what would a tube of, say 2050 mean in terms of sonic quality / life-span, etc.
Thanks,
Charlie
Hi Charlie,
Readings like your 2,600 figure are in umhos, a figure of transconductance. That you will find on a data sheet. There is a "bogie" or target number. Normal manufacturing variances will cause this number to vary. It drops as a tube is in use.
The mid-hundreds numbers are off a percent scale. Those meters normally only measure emission. Those are not terribly useful tests of a tube (unless it's a rectifier).
-Chris
Readings like your 2,600 figure are in umhos, a figure of transconductance. That you will find on a data sheet. There is a "bogie" or target number. Normal manufacturing variances will cause this number to vary. It drops as a tube is in use.
The mid-hundreds numbers are off a percent scale. Those meters normally only measure emission. Those are not terribly useful tests of a tube (unless it's a rectifier).
-Chris
I'm afraid such numbers are essentially meaningless.
Attempting to describe the health of a complex component with one number is doomed to failure. Valve tolerances are such that one new valve might easily have 50% more transconductance than another new valve. As anatech says, transconductance falls over the life of a valve. How will you distinguish between a new valve with lowish transconductance (but plenty of life in it) and a very tired but initially high transconductance valve?
Secondly, any measurement is only of value if you know the circumstances under which the measurement was made. I can obtain wildly different values of transconductance for a single valve simply by changing the point at which I make the measurement. A more helpful measurement might give the conditions: Va = 250V, Vgk = -4V, then the results might be Ia = 8mA, gm = 2.5mA/V. Note that the units of measurement have been specified. If you had a series of measurements like this, you might be able to decide which valves were the out and out clunkers. But it's still not ideal because...
There are various ways in which valve testers work, and some make some pretty amazing assumptions/approximations. Thus, the same valve could easily test completely differently on different testers. A better measurement might say: Va = 250V, Vgk = -4V, Ia = 8mA, gm = 2.5mA/V, measured on an AVO VCM163.
But even that's not ideal. How do you know that the geriatric test equipment is reading correctly? Remember that any valve tester is likely to be at least forty years old, so there's plenty of scope for it to be at the very least out of calibration, and possibly faulty. (My AVO had four independent faults when I first acquired it.) What you want to see is: Va = 250V, Vgk = -4V, Ia = 8mA, gm = 2.5mA/V, measured on a recently calibrated AVO VCM163.
Are we out of the woods yet? Not quite. Did the person making the measurement know how to use the tester correctly? You'll have to make your own judgement about that by looking at any supporting evidence.
Now you see why I say that such numbers are essentially meaningless. But you'll be able to recognise numbers that might be helpful...
Attempting to describe the health of a complex component with one number is doomed to failure. Valve tolerances are such that one new valve might easily have 50% more transconductance than another new valve. As anatech says, transconductance falls over the life of a valve. How will you distinguish between a new valve with lowish transconductance (but plenty of life in it) and a very tired but initially high transconductance valve?
Secondly, any measurement is only of value if you know the circumstances under which the measurement was made. I can obtain wildly different values of transconductance for a single valve simply by changing the point at which I make the measurement. A more helpful measurement might give the conditions: Va = 250V, Vgk = -4V, then the results might be Ia = 8mA, gm = 2.5mA/V. Note that the units of measurement have been specified. If you had a series of measurements like this, you might be able to decide which valves were the out and out clunkers. But it's still not ideal because...
There are various ways in which valve testers work, and some make some pretty amazing assumptions/approximations. Thus, the same valve could easily test completely differently on different testers. A better measurement might say: Va = 250V, Vgk = -4V, Ia = 8mA, gm = 2.5mA/V, measured on an AVO VCM163.
But even that's not ideal. How do you know that the geriatric test equipment is reading correctly? Remember that any valve tester is likely to be at least forty years old, so there's plenty of scope for it to be at the very least out of calibration, and possibly faulty. (My AVO had four independent faults when I first acquired it.) What you want to see is: Va = 250V, Vgk = -4V, Ia = 8mA, gm = 2.5mA/V, measured on a recently calibrated AVO VCM163.
Are we out of the woods yet? Not quite. Did the person making the measurement know how to use the tester correctly? You'll have to make your own judgement about that by looking at any supporting evidence.
Now you see why I say that such numbers are essentially meaningless. But you'll be able to recognise numbers that might be helpful...
Hi EC8010,
You can ony look at the appearance of the tube to get an idea of it's run hours. Experienced eye required.
I will say that Dynamic Transconductance is the best quicky tool a technician has. Beyond that you need to perform lab tests. Be aware that a tube in storage will measure low and increase a bit, then begin a long, slow decline. A new intitial reading doesn't say anything about life (reduced heater current "life test" may give more clues). A tube that measures too high is just as defective as one that measures too low.
I use a Stark 9-66, the same as a Hickok. They are calibrated in umho's, but these conditions are not the same as the manufacturing plant testing conditions or the operating conditions in circuit.
-Chris
Fully agree.
Absolutely!
You can ony look at the appearance of the tube to get an idea of it's run hours. Experienced eye required.
I will say that Dynamic Transconductance is the best quicky tool a technician has. Beyond that you need to perform lab tests. Be aware that a tube in storage will measure low and increase a bit, then begin a long, slow decline. A new intitial reading doesn't say anything about life (reduced heater current "life test" may give more clues). A tube that measures too high is just as defective as one that measures too low.
I use a Stark 9-66, the same as a Hickok. They are calibrated in umho's, but these conditions are not the same as the manufacturing plant testing conditions or the operating conditions in circuit.
-Chris
I found this site a while back, very good info on tube testers and their limitations. B&K's are good too, but only certain model numbers can perform transconductance tests.
http://tone-lizard.com/Tube_Testers.html#advertising
Even so, try the tubes in another amp before you chuck them, if the filaments are good and they pass the shorts & leakage tests.
http://tone-lizard.com/Tube_Testers.html#advertising
Even so, try the tubes in another amp before you chuck them, if the filaments are good and they pass the shorts & leakage tests.
Hi cbutterworth ,
Hi all ,
If you want to know and to understand almost everything
about Tube Testers , I strongly recommend the book :
“ TUBE TESTERS and CLASSIC ELECTRONIC TEST GEAR
by Alan Douglas “ , a comprehensive history and technical data ,
the limitations , drawbacks , advantages and calibration data ,
for the most commons Tube Testers ever made ( in U.S.A. ) .
Edited by Sonoran Publishing , Alan Douglas is a recipient of the
Antique Wireless Association’s Houck Award for Documentation
of radio history .
The book has a second part containing informations about VTVM,
signal generators , oscilloscopes , bridges , decade boxes , etc.
It’s not an advertising , it’s only a big and good “tip” , about a good and well ilustrated book .
Regards ,
Carlos
Hi all ,
If you want to know and to understand almost everything
about Tube Testers , I strongly recommend the book :
“ TUBE TESTERS and CLASSIC ELECTRONIC TEST GEAR
by Alan Douglas “ , a comprehensive history and technical data ,
the limitations , drawbacks , advantages and calibration data ,
for the most commons Tube Testers ever made ( in U.S.A. ) .
Edited by Sonoran Publishing , Alan Douglas is a recipient of the
Antique Wireless Association’s Houck Award for Documentation
of radio history .
The book has a second part containing informations about VTVM,
signal generators , oscilloscopes , bridges , decade boxes , etc.
It’s not an advertising , it’s only a big and good “tip” , about a good and well ilustrated book .
Regards ,
Carlos
- Status
- This old topic is closed. If you want to reopen this topic, contact a moderator using the "Report Post" button.