Interviews

Tu-be or not Tu-be
Questions and Answers on New Old Stock Tubes

Part 4 - Tips for Buying N.O.S. Tubes

If you are looking to buy a new old stock tube from a N.O.S. tube dealer, what types of questions should you be asking about a particular tube before you decide to buy it?

First, you want to make sure the tube is genuine. Unfortunately, I don't know how to tell anybody how to find out when a tube is genuine from a dealer. If you don't have a lot of background in tubes, you're basically at that person's mercy. My best advice is to ask around to make sure the dealer is somebody you can trust and who knows what he's talking about. I specifically recall a guy advertising Amperex JAN 7308, ten for $90, in Audiomart. I found out that they were really re-labeled Russian 6DJ8’s. I called to tell him, but he didn’t seem to care. He got lots of calls.

In terms of test results, if somebody professes to test a tube for low noise, I would ask specific questions regarding the methods used. If they say "I plug the tube into my preamp and listen to it", well, there is something to be said about that as far as revealing really bad tubes. However, I don't know about you, but it would be hard for me to quantify the degree of tube noise coming out of a speaker simply by listing to one tube after another, allowing them to settle in and heat up before doing the critical listening. I just don't think my memory is that good. I would ask the seller of the "low noise" tube if he has a way of applying some type of a numerical value to the degree of tube noise.

What types of tests are typically done on a new old stock tube to determine whether it's a good tube or a bad tube?

The most important factors for the tube consumer to consider in making a N.O.S. tube purchase should be (1) to make sure the tube is new; (2) to make sure that it has been tested for good drive capability; and (3) if it’s a dual triode like a 6922 or 12AX7, to make sure that it has good section-to-section uniformity. Beyond that, two other things to be concerned about are the inherent noise being generated by the tube and the tube's sensitivity to microphonics. All tubes are microphonic to a degree, but what really counts is whether the tube is microphonic to the point that it won't effectively perform in a circuit. Finally, if it's a brand new tube, you want to make sure it's not gassy. You have to keep in mind that gassiness in power tubes can often be burned off, so this is not always a huge problem.

What types of tests do you perform to determine those characteristics?

In testing tubes for tranconductance and for section-to-section uniformity in twin triodes, I use Hickok 539C, the super rare 580 and 580A, and a Triplet 3444A tube testers. I use a Tektronix 570 curve tracer when I match power tubes.

To test the tubes for noise and microphonics, I have my own set up. I use a Kaye Labs small signal tube tester as a platform to plug in both the tube to be tested and my headphones. The Kay Labs tester is then hooked up to a Spectral Dynamics FFT Analyzer, which provides me with numerical readouts for many of the tests I perform.

To check for microphonics, I developed a rather unique series of tests. I typically start by placing a speaker running a signal amplified by the tube being tested near the tube itself. This tells me whether the tube is prone to feedback problems. Next, I'll flip a variety of switches near the tube and take peak readings through the Spectral Dynamics analyzer to determine vibration sensitivity. I can perform an assortment of other tests on the tube as well, depending on the customer requests.

As far as noise level goes, I check that using the Spectral Dynamics unit. The beauty of this setup is that it allows me to assign an objective numerical value to the tube's performance at any particular frequency.

While these tests do provide a certain degree of useful information, I think it's important not to get too tied up with the microphonic and noise performance of N.O.S. tubes. The ultimate proof is in the listening. I've heard many vintage tubes that are by no means the last word in either of these two categories, yet sound more musical when used in actual audio equipment than a number of modern test-bench champs.

Unless you're an expert, is there any way to tell a new old stock tube from a used tube?

[Kevin picks up a box of used tubes and we go through them]

Yes. If you look very carefully, you'll notice shadowing in the glass between the bottom of the tube and the bottom spacer. A lot of tubes will get burn marks in this area if they've been in use for even a short period of time. You can also look for shadowing between the upper plate and the "getter" (the silvery coating of material on the glass inside the tube). With power tubes, the getter will go from being silver to turning kind of a milky gray with use. The first thing that happens with use is a rainbow-like band of coloration will appear around the edge of the getter. Over prolonged use, the rainbow coloration will continue to grow while the getter just slowly fades away.

What are matched sets of tubes and why is tube matching important?

Tube matching typically involves matching the bias voltages of power tubes or the section-to-section uniformity in small signal tubes.

Imagine that you have a four-engine race boat. For ideal performance, you would want to make sure that all of those engines were capable of the same top speed when under load. If the engines are not matched, the overall performance of the boat could potentially suffer. This doesn't mean that the boat won't go fast, in fact, small differences in performance capabilities would likely go unnoticed, but it's sure nice to have uniformity. Tubes perform in much the same way.

Tube matching is much more critical in power tubes than in small signal preamp tubes, especially in power amps having only one bias adjustment for a pair/quad of tubes or, as in certain older, lower-priced amps, a fixed bias.

In simple terms, bias is negative voltage applied to the tube's grid to keep it under control and prevent it from destroying itself. The application of negative voltage, lets call it idle current, allows the user to adjust the operation parameters of the particular tube. On most amps, when you turn the bias screw/adjustment, you are reducing the negative voltage and allowing the tube to run hotter and harder at idle. That's why if you use a meter at the tube bias points, you see the readings in milivolts go up as the bias that keeps the tube under control is reduced.

Some tubes, due to manufacturing variations, will be "current hogs". These tubes will run hotter than the rest of the tubes in amps having only one bias adjustment since they need a different amount of bias to produce the same current. That's why tube matching is much more important in amps with a single or no bias adjustments than in amps where the bias of each tube can be individually adjusted.

In order to match power tubes, with all other operating parameters remaining constant, you need to bring the tubes to the desired idle current and measure the bias voltage. Matches between tubes are then made by comparing their respective bias voltages. Unfortunately, these tests cannot be performed with most tube testers. Very few testers will get anywhere near the output voltages of a power amplifier and allow you to take the bias voltage measure and plate current while the tube is under load. For these tests, I use the Tektronix 570 curve tracer.

How close should the matches typically be?

For transconductance within a dual triode, no more that 20% within the tube. Between pairs, it doesn't hurt to get them within 5% to 10%. Bias points on power tubes should be within 5%.

What are the consequences, if any, of using unmatched tubes in a particular component?

You may never notice whether or not tube pairs are matched. I think it certainly doesn't hurt to have them matched, but I don't know that it is as big a deal as everybody makes it out to be

I have heard a lot about tube washing and re-labeling in terms of turning a modern tube into faux N.O.S. by removing the markings and replacing them with N.O.S. logos. Are there any kind of telltale signs for the layman to look out for in terms of that particular practice?

Re-labeling is something that has gone on since the early '60's. The original Marantz power amplifiers were shipped with Siemans EL34 tubes labeled "Made in Germany". Well, Siemens wasn't making EL34's in Germany in the early '60's, the tubes were actually relabeled Mullard tubes.

Tube washing and relabeling originally occured when a company didn't want to disclose where their tubes were actually made. Nowadays, tube washing is also practiced by people buying common tubes, washing off their labels, and re-branding them with exotic names in an effort to increase their value. If you find someone selling large quantities of expensive tubes like Amprexes or Mullards at bargain prices, I'd be highly suspicious. Beyond that, the only way you can really identify the true manufacturer of any tube is by looking at the internal structure and the batch codes on the tube itself. Fortunately, the manufacturers of the most sought-after tubes like Telefunkens, Amperexes, and Mullards placed indelible marks on their products. Telefunken molded an outline of their famous diamond logo into the glass at the bottom of their tubes. True Amprex and Mullard tubes have batch codes etched into the glass.

Interestingly enough, the Amperex Bugle Boy brand is going to be marketed again by Richardson. I think tube consumers inclined to purchase "new" Bugle Boys should be aware of what they're buying and should probably confirm the identity of the actual manufacturer of these tubes with Richardson as a factor in their tube-buying decision. When people mention Bugle Boy tubes, they are typically referring to those tubes manufactured in the late 50’s/early 60’s in the Philips Holland or Mullard U.K. plants.

Are there certain tubes that tend to be faked more than others?

Telefunkens aren't faked much because of the diamond. Amprex and Mullard seem to be the favorite names because, after all, who would know how to identify a real Mullard or Amperex? I have written an article for Stereophile that will be coming out soon that is going to spill the beans on tube ID, meaning what the batch codes are and what they mean and who was really doing what.

Is there anything like a Bluebook or other kind of resource out there for valuing NOS tubes?

Not really. Charlie Kittleson put out the Vintage Hi-Fi Spotters Guide and Price Guide, and it does have a little blurb in there about tubes, but, I don't know if that's helpful.

So the N.O.S. tube market is a more of supply and demand-type situation as opposed to a fixed values…

Yes. You can walk into a TV shop tomorrow and buy a bunch of stuff for $0.50 each, and the tubes could be worth $50.00 each.

I sense that we are at the end of this deal- two years from now the supply of NOS tubes is going to be dust. It is like a meal that was just eaten by a snake, as it passes through the snake, the big lump finally just goes away. We are at the end of the lump right now. Tubes that I took for granted a year ago, I kick myself for not buying today. Like I said, I don't buy a lot of stuff in the US anymore. To get the tubes my customers demand, I am going to real small, out of the way countries- it takes a lot of time and a lot of poking, but that is the only way that you can get the tubes lately.

...continued in Part 5 - Care and Maintenance of N.O.S. Tubes