Intrigue & Controversy - Mutual Conductance vs Electrona

[cwmoser]

I read some comments on websites about the Hickok vs Precision - i.e. Mutual Conductance vs Electronamics test methodology - as to which is better. I was not around in the era when this was debated and wondered if folks could post more about the historical setting and what it was like back in the hey days when Tube Tester manufacturers competed against each other.

To my semi-trained brain, it seems both have merits but I know folks today seem to favor the Mutual Conductance tube testers.

So, what is all this about? Is it "much ado about nothing"? What were the perceptions back in those days?

Thanks

CW

[Alan Douglas]

Precision (and Eico and Jackson) used a "dynamic emission" circuit which was one step above a simple emission tester. Instead of connecting all grids and plates together and operating the tube as a diode (simple emission test) they applied different AC voltages to each element, and put the milliammeter in the plate circuit. It's not Gm by any stretch (unless you're an advertising copywriter). "Electronamic" is a meaningless name. But the circuit does work, and will find problems that the simple emission testers won't.

[Chris108]

I can tell you in one word what we thought of mutual conductance and dynamic tube testers in the TV repair shop where I moonlighted 30 years ago while in college: Bunk!

When we encountered a circuit that was malfunctioning, the first step was to rule out any known or visible component failings, then swap the tube with a known good spare. If the performance improved, the set needed a new tube. If not, you went looking for problems elsewhere. Tube testing was only done if there was a suspected bad tube and no spare was on hand, which did not happen very often.

In the tube manuals published by RCA and others, there is some important information which is implied--they don't come right out and say it. The specifications for mutual conductance (for triodes) or transconductance (for other kinds of tubes) are given at or near the maximum operating levels for the tube type. In other words, emission and every other measurable aspect of a tube must be nearly perfect, otherwise you won't get the specified voltage or power gain!

There were tube analyzers built which allowed this kind of "extreme" tube testing, but they were so huge and costly that nobody could afford them outside of tube factories and labs. Later on, tube curve tracers were developed to do the same thing graphically, but most were so expensive that no ordinary repair shop ever considered buying them. (Used lab grade tube curve tracers are still highly sought after to this day, BTW).

Mutual conductance tube testers are portable, scaled-down versions of lab tube analyzers. The biggest difference is that they do not give you much choice of operating voltages, so the transconductance numbers are "in their own world." They do not usually represent what will happen in an actual circuit, nor are they the same as the tube manufacturers' numbers.

Dynamic testers use a different circuit to accomplish essentially the same results. All of the elements in a tube have to "work" in order to get a good reading. Since there is no gain measurement that can easily be converted to transconductance, you get a "bad-good" scale, and maybe an arbitrary numerical scale that relates to average performance.

Most of us who worked in the field found dynamic testers perfectly adequate for general radio and TV repair. Where mutual conductance testers came in handy was in situations where the actual gain of a tube had to be within close limits in order for a circuit to work correctly--something that was almost never the case with consumer electronics. Such circuits are found in some kinds of test equipment, analog computers, servo systems, and the like. Much has been made of matching tubes for push-pull audio stages, but IMHO this has been greatly overblown. Yes, the tubes have to match (within reason), but then the rest of the circuit has to match too!

The downfall of all conventional tube testers, regardless of type, is that they do not test most kinds of tubes anywhere near their full ratings. Consequently, you have only an educated guess that a tube might actually deliver its full measure of voltage or power gain when it is called on to do so. This is why we did not bother testing tubes unless we had no good spares available to try first.

[k6sti]

[Chris108]

Usually we'd take a NIB tube off the shelf and try it in the set. If it fixed the problem, the customer bought it (assuming they had the set fixed). If it did not help, the original tube went back in the set and the new one went back in the box until needed again. Sometimes we could also temporarily borrow a tube from a TV that had been junked for a different problem, etc., etc.

[k6sti]

Sounds reasonable. But did you ever encounter a new tube that wasn't quite up to snuff? And how would you test for that?

Makes me wonder how many new old-stock tubes are still good today. It's the possibility of air molecules sneaking past the metal/glass seals over the decades that worries me. Kind of like my electret headphones, which eliminate the HV bias supply by using a material with trapped charge. But trapped for how long?

Brian

[Lou deGonzague]

Most NOS tubes are still good unless they have been stored in very bad conditions. I have 4 WD-11 tubes that still work good and they are about as old as they get. At the Lafayette store I worked at they had a high end tester for communications and high end audio gear. I agree with Chris on his post(very good info), way more important to have a known good tube for testing a stage then a tubetester. Sometimes after restoring a set I might run a few of the used tubes in my stash and see how they work, then label them "set tested".

[Bruce Hagen]

Tube failures of the shelf were extremely rare but if we suspected one we simply tried another. Even though we had twice daily delivery (talk about being spoiled!) we typically had 10 each of the popular types at hand. There were a few tube numbers such as the RCA 6FQ7 that were outside of this comment.
_________________
Bruce

[gadget73]

I use my tester as a go/no go test to see if the tube is doing something or nothing. Very seldom have I seen it show anything but 100% or 0% emission. Its a really basic emission tester though, so I don't expect miracles. It is useful for weeding out open heaters or showing shorts though. More often I swap tubes around to see if anything changes. I've put in tubes that my tester said were 100% and they were flat horrible in the circuit. HV rectifiers for TV use are a particularly notable example. I went through all of my 6 1B3 tubes before I found one that would give me a normal picture without blooming out with the brightness control anything over 0%. Every one of those tubes tested fine on the checker though.
_________________
Saving vintage electronics, one dumpster at a time.

[Chris108]

It absolutely did happen that we'd take a brand-new tube off the shelf, put it in a set, and find it was bad. Once in a blue moon, you'd get one where the getter turned white, or one that would make the symptoms worse than they were before.

More commonly though, you'd change a horizontal output tube or a damper, and the new one would spark internally due to debris or loose elements. You had to get it out of there without fail because it could decide to spark later on, pop the set's B+ fuse or circuit breaker, and cost you a customer.

The way we used to "test" old output tubes and dampers, BTW, was to rap them gently with the handle of a screw driver. If the tube flashed internally, it usually meant that things were getting loose inside, and replacement would be indicated to avoid further problems. This was really the very best kind of 'tube tester' to use when a TV came in with the circuit breaker or fuse blown for no apparent reason, and it came right on and started playing after a reset. To this day, I am still in the habit of giving rectifiers and power tubes a tap or two while they are in the equipment or on a tube tester to see if anything "flashes" or behaves erratically.

[Alan Douglas]

[Dave Doughty]

Here are links to Precision's published propaganda sheet on their electronamic method of tube testing claiming it is "more than" just mutual conductance testing.

http://i30.photobucket.com/alb ... ision1.jpg
http://i30.photobucket.com/alb ... ision2.jpg

Dave

[dsbk]

The gain of 12AX7 and triode to triode balance of 12AT7 is important in what I do with guitar amps.

Finding defects in all tubes that I use for anything is also important.

While my tester is very good at detecting shorts and grid-emissions, I have found little correlation between Gm readouts and the actual voltage gain of a tube in circuit. RCA does tell us that Gm is the most important paramater for small triode testing.

Maybe a low-level preamp tube is an extreme case, but testing at various service tester parameters, on whatever part of the curve a particular tester uses has proven to be unreliable for me.

I am currently testing these two types with an 1Kc audio signal at 400 volts B+, in a circuit that RCA, Mullard and Amperex have documented the gains for.

My Gm tester is up and down all over the place with Gm. In the real-world 400 v. tester, the spread on gain readings is much more narrow. Also the section that showed as the weaker for Gm is often the stronger in real world tester. I have modified my B&K 707 to have more appropriate parameters for the 12AX7, but still its Gm readings don't tell the whole story or the whole truth.

I have got to conclude that Gm is only one parameter, and that it becomes misleading to trust such a test on the wrong part of the curve.

The service-tester tube-matchers have got to be buying into an illusion, IMHO.

The good news is that my tester does do exactly what it was built to do, and does it very well. I trust it completely as a service-tester, not a tube matcher.

I buy tubes from folks with many other kinds of testers. Same story when I test what I bought.

Alan research, as shown in his book, seems to indicate that the electro dynamic method is not so hot either, and can sometimes beat-up a tube. I think I did once ruin a tube in my emission section.

The Cardmatic is interesting, but still not very high plate volts.

I wonder what volts are available in a tester like the AVO. Anybody using an AVO?

ds
_________________
"Be bold and mighty forces will come to your aid". Basil King
"Boldness has genius, power and magic in it." Goethe

[BPlus]

I still like the "dynamic" testers and particularly the Jackson 648, even if there's no evidence that they work better. I think this discussion goes back to the days when Chris H was still with us.

I'm firmly convinced that my Sencore MU140 is the best thing to test small-signal audio tubes since it uses a 5 khz signal and automatic bias. But I have no actual results to back that up, I just like believing it.

I think that the Jackson 658 must be good for audio output tubes since I vaguely remember that Erik K's writeup mentions higher voltages. Someday I'll have to actually test it. Or not.

I'm sure there's a TV-7 around here somewhere. I don't know why I'd choose it over any other tester.

Most of the time I use a little B&K 667 or a Mighty Mite.

A real good test for 12AX7s is to plug them into my Fender Champ. I know how it should sound. Sounds good, no smoke, good tube.

You can go nuts thinking about tube testers, but they really don't make that much difference in getting your radios or audio equipment working. Any well-known brand with the right sockets and a complete tube chart will do. Find the dead and shorted ones and it's done as much as you can expect.

-Steve W.

[dsbk]

[gmcjetpilot]

Here is a tube tester folks

http://www.amplitrex.com/index.html
http://www.amplitrex.com/at1000.html

It is only $2550 BRAND NEW!

It plots as well. I want one and I don't know why?

Don't laugh, you can buy a used Hickok and get it all calibrated, tricked out with digital meters and it will cost the same money, but it will still be an antique. This at1000 does way more.


I agree with all the comments above but I still think as flawed as Gm measurements can be, it is DATA and good info. I think a radioman should have both an emission and mutual conductance tube tester. I am pretty happy with my Mercury 1000. It comes in handy when buying tubes that are suppose to be strong or NOS. My tester gives Gm spec numbers when the tube is NOS or new. It gives something less than spec when they are used. I bought 4 tubes and one of them was weak weak. So I got a partial refund. It does indicate something more than OK. What does it indicate? Well that is the debate. You have to use it with understanding. I have some low reading tubes that work fine in the radio.

[renaissance.man]

If you want to learn all about tube testers and testing and plotting tubes, get a copy of Alan Douglas's book, Tube Testers and Classic Electronic Test Gear. There is info about building "the ultimate tube tester". Also how to convert SS curve tracers for vacuum tube use. Trust me, this is some great stuff.
_________________
Ed

[SamD]

[neutronic]