Do you need matched output tubes? Is it worth the extra bucks? Are unmatched tubes just as good? There are at least three major things to match in output
Do you need matched output tubes? Is it worth the extra bucks? Are unmatched tubes just as good? There are at least three major things to match in output tubes that affect the sound of a pair of tubes in guitar amps. “Matched” may mean a number of things. And you may or may not want perfectly matched tubes.
Today’s tubes vary a lot. The vagaries of mechanical tolerances, materials chemistry and perfection of vacuum translate to variations in the DC bias point of the tubes, the AC signal gain of the tubes, and the distortion characteristics of the tubes.
Most of us are familiar with biasing to eliminate crossover distortion. We adjust the grid bias voltage so a little DC current flows with no signal. That current lets the two tubes hand off signal cleanly as the signal changes from positive to negative, and it is what you set when you DC-bias your amp. The tubes not only need to have a little current flowing with zero signal, but that current also needs to be the same for both tubes. That’s DC matching – selecting two tubes which happen to have the same current for the same bias voltage.
Some amplifiers have a bias adjuster for each tube that can effectively DC-match un-selected tubes. This is important enough that I put per-tube bias controls into the Workhorse amplifiers with LED indicators so you can DC-match your own tubes in the amp. If your amp doesn’t have separate per-tube biasing, you may need to pay the extra money for matched power tubes.
Each tube is also responsible for amplifying its own half of the signal smoothly. The AC gain (or “transconductance”) of the tube does this. If the AC gain of one tube doesn’t match the gain of the other tube, the result is distortion as well. Unfortunately, most tube testers can’t measure gain at full power, so gain matching on a simple tube tester may not reflect what you get in your amp. On the bright side, this is asymmetrical or evenorder distortion, and human ears perceive a little bit of this as a “sweetening,” not as distortion at all.
A high AC-gain tube on one side of a pair will hit overload more quickly than a lower gain partner will. A lower-emission tube may simply run out of electrons to pass, and may also overload earlier. Both of these may cause a lower clean power output from the amp.
Little quirks of the elements inside a tube can cause another form of distortion – little bends and twists of the signal within what is otherwise a linear part of its operation. This is the kind of difference that helps make an EL34 sound different from a 6L6. A lesser degree of this difference exists between any two tubes of the same type.
Distortion matching can be done, but this is a very muddy area. I’ve talked to some tube suppliers who claim to do this, and they have all been reluctant to discuss this area of matching. I’m guessing that most of the “distortion matching” is really a combination of DC and AC gain matching.
Some suppliers provide tubes matched by a grouping scheme, the idea being that if the tubes in your amp are from the “#5 group,” you can replace them with new pair from the same group without re-biasing. This is O.K.-ish, but since the tubes in a group like this have a range of values in them, don’t count on perfect matching from set to set. And bias is not directly tied to distortion, so I can’t see how the grouping system can do anything about overload or distortion.
Automated tube curve tracers exist that can run the tubes at full power, and record what the tube does at every point over its entire range. This kind of testing is great for hi-fi tubes, and is likely to be the future of tube matching and testing. Unfortunately, these are quite rare today.
Tubes also drift with heat and age. Tubes that were matched when you bought them can drift apart. You can see this graphically in the Workhorse amps by looking at the LED indicators after you have adjusted the bias current. The tubes will drift in and out of perfect bias slightly – which is actually O.K.
After all that, the answer to whether you need to buy matched tubes is a resounding “maybe…” You need equal DC bias currents per tube. If your amp is not one that lets each tube be adjusted independently, you need to buy matched tubes. Beyond that, your ears may or may not like the slightly sweetened sound of imperfectly AC gain matched tubes better.
As a musician, you need to check it with your best test equipment – your ears.
R.G. Keen
Cheif Engineer
Visual Sound
www.visualsound.net
