Anyone Here Ever Purchased A Tube Tester?

I found a Healthkit TT-1A tester a few years ago that had been recently restored by a shop that understood this device. I don't use it often enough but it's great and (relatively) easy to use. It will test *almost* any tube made through 1978, including DHT and Nuvistors. It's a bit of a cumbersome beast. The paper rolls are in good shape but I find it easier to lead digital files on my laptop or access from a tablet.

Yesterday I tested all the 19 tubes in my preamp with the Beck RM1 tester. I found three 6922 tubes that measured very weak, and replaced them with spares that tested ok. I was rewarded with better sound. What is interesting is that I had not really noticed that the sound had become a poorer, more muffled, closed-in. Until I replaced the weak tubes. Then, the benefit was obvious. So I told myself - good idea to do this more often!

I know that the RM1 is limited, but in this case it worked well. The meter readings corresponded to the sound. So, I will have to buy three 6922s, not a full set of 18 6922s plus 1 12au7 (expensive for NOS). RM1 also catches bad tubes, including 6as7g, before I put them into my amplifier. I avoid tubes creating errors in the amp or preamp. RM1 doesn’t test transconductance or noise, but has actually saved me money over some years of use.

Whatever the tube tester, you may be able to develop your own system for testing the things it doesn’t test. Like the background tube noise level. I have a strict procedure, measuring the noise level of the tubes in my system with a db meter, compared to a standard reference (with volume loud, not playing music, with the tubes in my very noise-sensitive phono preamp). In case of dispute I can send the measures to the seller, to get my money back. A year ago I got a quad Telefunken that sounded good but had too much noise. I had to return them. I was sad and the seller was irritated ("but they test strong!").

My noise level test is done in a very demanding context. The phono preamp has three tube gain stages, and the first is stretched to the utmost - amplifying the weak signal from the cartridge. Each stage has four 12ax7 tubes, two per channel. This is a very puristic solution and when all is right it sounds wonderful.

I know that I can use "mild offenders" in the phono noise level test elsewhere in my system, without obvious defects. If a 12ax7 sounds good in my phono it will sound good in my headphone amp too, and there, I don't hear the noise problem (or not much).

I have a quad of NOS Philips Herleen 12ax7 that measure very strong and have good sound. At first, the noise level was quite acceptable, but over 1-2 years it became higher, and I had to take them out of the phono preamp. I am not sure what to do with them.

The background tube noise, in my system, is always there, if I turn the volume towards max, with no music playing. Almost all the noise comes from my phono preamp, even if the preamp and the amps - both tubed - contribute too. In practice the digital input sounds silent while the analog input has noise, more or less. To hear this noise, unless the phono tubes are noisy, I have to turn the volume up. It is not so easy to hear when playing music, although it appears in quiet passages. It is my strong conviction that this background noise is NOT what I want, even when it is not so easy to hear. It intrudes on all the rest. The music becomes less stable, less at rest, less undisturbed. Often called "black background". Not sure why, since I dont hear it as "black". What I hear is less noise.

Athough the diagnosis may be complex - what noise, exactly, and why - it is easy to hear, with the volume up loud. It may start with a slight sense of 'wind' in the sound, developing into a 'storm' in the noisy cases. The tube noise can also resemble cooking. When a tube has developed cooking sounds it is usually on the way out of the kitchen...in my system.

What I don't know, is whether it is good practice to use moderately noisy but good sounding tubes in less critical positions in the system. Or is it best avoided.