What's Your Approach for Testing Tubes?

[ayan]

In Scott's thread about the JJs, I talked about how I tested 12AX7s: amp clean, or when in OD channel pretty clean as well, no effects, loud volume. I strum or "claw" on various chords, as consistently as I can, up and down the neck listening for harmonics. I try to avoid brushing the strings or using a pick with any nuance whatsoever, as this may create additional harmonics. I listen for even order harmonics or lack thereof, and for odd order ones as well (when you get amusical feedback, not related in pitch to the notes being played), as well as for microphonics and other noise. Based on what I hear I basically bin the tubes in 4 buckets: (1) Tone dead or hopelessly microphonic... garbage bound; (2) Inert -- maybe good for an FX loop or PI application (3) Good; (4) Gem. I always stand in the same place when doing this and can only take 30 minutes at a time or so before my hearing begins to suffer and everything starts sounding the same.

Would you share with me how you check your tubes? I just put in an order for a dozen 12AX7Bs to see if I can find me some good tubes, as my supply of good sounding ones is dwindling.

Thanks,

Gil

[nickt]

Hi Gil - no help from me

but on another note...

I found a DIY tubetester at http://members.aol.com/sbench101/#TubeTest and have been thinking of putting it together. Seems the real ones are getting pretty pricy and rare.

Since we're basically interested in a narrow range of toobs a "lite" DIY version could be useful in slapping down objective measurements to share. Waddaya think?

cheers
Nick

[Bob-I]

I must say my method is no where as detailed as alyn's. I put the tube in, let it run for awhile and test it at a variety of settings, loud, soft, overdriven, clean and decide if I like it over a few days. I usually play my Red PRS 24 fret Standard on the bridge PU, or my Epi Sheriton with Jason Loller PU's. I often don't use a pick so that I'm not hearing only the pick attack.

Now you've got me thinking I need the bins and labels. I do that with sax reeds, write on the A B C F and really only use the A's.

[heisthl]

Fantasy tube tester - get an old Duncan convertible (the ones with the little boards with tube sockets that were plugable for different styles) and make a bunch of plugins to test the 12AX7's at all different voltages and bias points in the same amp. You could probably nail down the good ones in less than 5 minutes each.

[nickt]

Fantasy tube tester - get an old Duncan convertible (the ones with the little boards with tube sockets that were plugable for different styles) and make a bunch of plugins to test the 12AX7's at all different voltages and bias points in the same amp. You could probably nail down the good ones in less than 5 minutes each.

Hmmm... can't find that via google. There is a Seymour Duncan Convertible Amp...

Back on topic - obviously ears are the final test but never been able to afford to buy tubes in big lots to pick over. Then again things are looking up just now cause the AUD is approaching 1:1 USD - maybe next order shouldl be bigger than my usual ones 'n twos.

[Normster]

I should pay a lot more attention to tube grading than I do. First test for me is plate voltage, especially on preamp tubes. If they're too far out of whack, I scrap them. Second is much like Bob's approach. I put them in the amp and play for a few days. If the amp doesn't sound right, I relegate the offending tube to "loop" duty much like Gil does. Noisy tubes get marked on the box and stashed. Not sure why, but I've had some really noisy EH's that work just fine in Fender amps.

I like your approach, Gil. Hopefully I won't develop a taste for NOS tubes.

[heisthl]

Hmmm... can't find that via google. There is a Seymour Duncan Convertible Amp...

http://www.diyguitarist.com/GuitarAmps/Convert.htm

Look at the power amp schematic - 12VAC heaters and a bias supply that can put out -110, not to mention the bizarre PI arrangement. I remember working on these years ago but I didn't remember how weird they were.

[nickt]

http://www.diyguitarist.com/GuitarAmps/Convert.htm

Look at the power amp schematic - 12VAC heaters and a bias supply that can put out -110, not to mention the bizarre PI arrangement. I remember working on these years ago but I didn't remember how weird they were.

Ahhhh... now I'm with you...

[Fischerman]

I usually already have a set of tubes in the amp so I'll let it warm up and listen to them briefly...then switch to SB and swap the one preamp tube. I usually just focus on one tone...either clean or dirty based on my intent for the amp (so for my ODS I test it dirty because that's why I built it). I use both a pick and just fingers because I only use a pick some of the time and my 'pick tone' is very different than my 'finger tone'. I listen for feedback/sustain as well as whatever properties I'm looking for in that tone.

It's really more of a 'do I like this better or no?" test. And I only test one or two tubes at a time...then I'm spent.

[ayan]

I like your approach, Gil. Hopefully I won't develop a taste for NOS tubes.

I had been -- still am -- a minimalist and wouldn't pay attention to tubes too much in the past. If a tube was microhpnic, I would replace it and that was that. The problem with my approach was that I was spoiled without knowing it, because in the late 90s I had bought a boatload of Eis (and mind you, I had bought them because they were cheap) and it just so happened those tubes really rocked.

Recently, as my supply began to dwindle I went to the next chepest thing, which were the JJs. At around the same time, I started to see that my amps didn't sound so good anymore... They were stable, but some were kind of lifeless. I remembered what plenty of people had told me long ago: Dumble amps are very sensitive to the types of tubes you put in. And that's when I started checking my supply of JJs, only to realize half of them were 100% toneless...

As I am getting ready to build another amp, I will need some more tubes and that's why I have a dozen of the 12AX7Bs coming my way. I will go through those and report back, but tech friends of mine, as well as Alan, seem to be in agreement that they are the best thing out there.

Cheers,

Gil

[stelligan]

FWIW, I have over a dozen JJ's in the house. My experience with them has been uneven. 50% of the time I find a good one. The others are as you and others have said. Reliable but lifeless. It seems I don't have microphony or failure issues, just tone disparity. The ones that I have ended up liking are ones that draw more current and bring the plate voltages down. My experience has showed them to vary widely there. A few months ago I spent 2 days going through the 4 d-style amps I own with every tube I had in the house. I am not a NOS tube guy so, most of what I had were Sovtek, Tung Sol, EH, & JJ. That event brought me to EH in V1 & V2 on most of the amps. For what I had on hand - those seemed most consistently satisfying tone wise. Some higher gain JJ EC803s ended up getting the nod in V1 in my non HRMs. I should try some Ei. Never have.

[ayan]

FWIW, I have over a dozen JJ's in the house. My experience with them has been uneven. 50% of the time I find a good one. The others are as you and others have said. Reliable but lifeless. It seems I don't have microphony or failure issues, just tone disparity. The ones that I have ended up liking are ones that draw more current and bring the plate voltages down. My experience has showed them to vary widely there. A few months ago I spent 2 days going through the 4 d-style amps I own with every tube I had in the house. I am not a NOS tube guy so, most of what I had were Sovtek, Tung Sol, EH, & JJ. That event brought me to EH in V1 & V2 on most of the amps. For what I had on hand - those seemed most consistently satisfying tone wise. Some higher gain JJ EC803s ended up getting the nod in V1 in my non HRMs. I should try some Ei. Never have.

Another thing I found about the JJs, in addition to some being tone dead -- and I guess our rate of sucess with them is pretty consistent, Stelligan -- is that they can be very mismatched from one half to the other. I tried some in the PI and I couldn't even begin to bring the plate voltages close to one another, and that's in spite of the Dumble trademark 10K trimmer. A few of those tubes would get the voltages only within about 10 VDC of one another. In my experience with these things, although the effect is subtle in the grand scheme of things, bringing the voltages to about 3 VDC of one another is where I tend to set the trimmer.

Gil

[glasman]

I spend hours on end trying tubes in amps. I just got a sampling of several new tube type from New Sensor to evalute.

My procedure...

1. Test in old Hickok 752.
2. Install in V1
3. Measure plate voltage looking for consistancy (with other reference tubes).
4. Thump on them several time looking for microphonics.
5. Let them burn in for a coulple of hours and then listen for noise. I typically toss about 25% or more for noise.
6. Play and listen to the complexity of the tone. (once again to reference tubes)

7. Replace/Repeat as ncessary.

I am planning on building a 20 tube burnin unit when time permits. Cook the tubes for 40 to 80 hours before testing. Should save me some time in the end. I still have in the back of my mind to run more exhasutive testing with respect to frequency response (with varying plate loads and grid resistors) and phase response. There must be a way to SEE what we hear .

My choices, as of today...

Selected JJ for PI
Selected EH for V1, V2 and Loop
JJ or Chinese 12AT7 for my reverb slot.

Of course this may change tomorrow.

FWIW, the MAN recommends Ruby Selected chinese tubes (C5) and SED 6L6's.


Gary

[Funkalicousgroove]

I use a VTV dual triode characterizer, I can go through about 50 tubes/hour and mark the duds, it also tells me which one are balanced, and which ones have matching triodes. it lets me listen to each triode individually.

After that I plug them into an amp and give a final listen, but the VTV tends to be pretty accurate.

[nickt]

I use a VTV dual triode characterizer, I can go through about 50 tubes/hour and mark the duds, it also tells me which one are balanced, and which ones have matching triodes. it lets me listen to each triode individually.

After that I plug them into an amp and give a final listen, but the VTV tends to be pretty accurate.

Hi Funk - this was what I was getting at when talking about some sort of tester. Clearly $1200 is not going to be spent by mere hobbyists like yours truely on a tester even if it works great. (now $200 in parts and $35,000 in DIY labour is another matter ).

Looking at the VTV website there's not much info on what the thing actually does. My guess would be it's hard wired to provide the most useful numbers/indicators for audio easily.

Since you find the triode characterizer pretty accurate (and your amps are good enough for the likes of Larry Carlton!! ) seems that a tester can replace ears at least for the sorting of "probably good" from "definitely bad".