• WARNING: Tube/Valve amplifiers use potentially LETHAL HIGH VOLTAGES.
    Building, troubleshooting and testing of these amplifiers should only be
    performed by someone who is thoroughly familiar with
    the safety precautions around high voltages.

Matching tubes, what parameter to use?

Status
This old topic is closed. If you want to reopen this topic, contact a moderator using the "Report Post" button.
I don't match anything but output tubes.

With them I measure bias voltage and gm at the Valve tester manual suggested anode current value (and at the manual suggested anode and screen voltages) I collect up into different piles those which match gm within say 5% (if I have >10 tubes I might match tighter than this)

From each of those piles I then choose pairs with the closest bias voltage. So I match for gm first and then bias voltage second. Just my method. Others may do things differently.

Oh! I occasionally test twin triodes for balanced triode sections for Phase Splitter duty, again I match gm first and bias volts second.

Cheers,
Ian
 
Last edited:
I think THE best way to match tubes is to use a tube curve tracer and match tube curves. However its obviously unrealistic to test tubes this way. The more common approach is gm matching.

That's what brought this question up for me.

I was testing my tubes with a uTracer.

I have a pair of two year old Electro-Harmonix 12AU7 tubes that are so close that the plots from both sections as well as both tubes all overlay each other.

I have a second set of 12AU7 thats are labeled with only 'China" that are very nearly as good. That was a surprise since they came with a kit and I had just put them in my tube box to use a backup tubes.

Then I have a 'New" pair of Electro-Harmonix 12AU7 tubes that were bought as matched sections and matched against each other that are not matched either in sections or against each other.
 
Matching a dual triode tube that is in a well designed 'LTP near perfect current sink' phase splitter might not be as important as . . . Matching a pair of very high transconductance output tubes in push pull output stage that uses a common cathode self bias resistor. Yes?

The application is often what determines the importance of a match.

For a pair of output tubes on SE mono-blocks that only use one output tube per amp, you might not need a close match as long as you have a good gain-balance control on the preamp.
 
Last edited:
Matching Phase-Inverter halves can be critical in Non-Feedback amps such as AC30 circuit.. However it is not as critical in feedback amps like Marshall, since the tail of the phase inverter is receiving error correction feedback that will push the signal into balance...Provided that things are not grossly out of wack in the first place....
 
TonyTechson,

I use Eurotubes.com for my tubes. They take JJ tubes, and retest them, match them, etc. They do very extensive testing of several factors. I know the test equipment they use, and also the designers of some of those measurement devices (all specifically made to test tubes).

Eurotubes works very closely with JJ, and a couple of them there have been at JJ in Slovakia several times.

I call Eurotubes, tell them what I want, and then because I am so lucky (blessed), I just drive over and pick them up.

I have had excellent performance results and reliability of those tubes.

Given all those facts . . . With very few exceptions, I will not design a tube amp that does not use any thing but JJ tubes from Eurotubes. (a few tube types I use are not made by JJ).

When I design a circuit that needs a phase invertor, I always use a pair of triodes, tie the cathodes together, use an extremely high impedance solid state current source, match the plate load resistors, and match the grid resistors of the next stage. That intrinsically matches the gain of the two phases.

If there is any need for further matching, it is the output stages that are not as intrinsically matched. I use well matched output tubes, but if they are not, I can slightly un-balance the phase invertor by putting an additional resistor in the plate load of only One side of the invertor. Find out which phase amplitude needs adjustment to balance the sum total of the invertor and outputs. A low Ohms rheostat in series with the plate load, or a high Ohms rheostat across the plate load. When I do that, I measure the fundamental, 2nd harmonic, and 3rd harmonic in real time on an FFT. Tweek, look, find the null, done! And done rhymes with fun. Time to listen.

Oh, if you are going to use negative feedback, be sure to do the null test without negative feedback. Make the amp linear before applying negative feedback. Negative feedback hides a multitude of bad characteristics, but do not make negative feedback have to work harder than necessary.
 
Status
This old topic is closed. If you want to reopen this topic, contact a moderator using the "Report Post" button.