Which current production 12AX7 have short plates and spiral filaments?

[Malcolm Irving]

I’ve just been comparing various 12AX7 / ECC83 in V1 position in a JTM45 type amp for buzz and micro-phonics. (All the samples were short-plate type.) No guitar plugged in, vol. at max., and a cheap SPL meter up against the speaker cloth. I tried about 6 tubes of various types (including some ‘top of the range’ new production). The clear winner was a relatively inexpensive Electro-Harmonix 12AX7EH. With this one in place, no buzz or hum at all, just the hiss you have to expect with the standard Marshall/Fender grid stoppers (two 68k in parallel for the high gain input) and minimal micro-phonics.

Looking into the characteristics of this particular tube, I noticed it is a short-plate type with spiral filament. I know that short-plate tends to reduce micro-phonics and spiral filament reduces hum/buzz. (I’m also aware that a lower gain sample would tend to appear quieter in this test.)

So, my question (for those of you who have stayed with me thus far) is can anyone tell me which of the current production 12AX7 / ECC83 (in addition to the EH) have short plates and spiral filaments?

[Jammin'John]

Spiral filament tubes include the 12AX7LPS, 12AX7EH, Tung-Sol 12AX7, Mullard 12AX7 reissue, and JJ ECC83s/12AX7.

JJ

[soundmasterg]

Spiral filament tubes include the 12AX7LPS, 12AX7EH, Tung-Sol 12AX7, Mullard 12AX7 reissue, and JJ ECC83s/12AX7.

JJ

Of course some of those you mentioned are not short plate tubes though. The LPS and Tung-Sol come to mind.....

Greg

[Jammin'John]

I only worry about having a short plate if there are microphonics.
Yes,some of those are long.

JJ

[Malcolm Irving]

Thanks Jamin’John and soundmasterg for the information. The modern production datasheets don’t tell you much.

[Jammin'John]

No they don't.

JJ

[km6xz]

Without a reference in voltage gain, your test reveals nothing about the inherent noise contribution made by the tube. If one tube is a lot higher mu than another you would expect a lot more noise despite normalized stage gain measurement of noise might be lower.
A better test would be for equal gain as a whole unit. Use a signal generator for steady tone and measure the power output (SPL) with your sound meter in a fixed position(don't touch it after taking the first reading), then short the input and noise the noise without adjusting any control. Try the next tube, with same input signal, and adjust the gain to equal the SPL of the first test, then short the input and read the no signal and subtract the no-signal reading from the with-signal reading for signal to noise ratio. Repeat for all the tubes.
But none of this is needed. Just play and don't obsess about it, you might like one of them for a variety of subjective reasons and use it, instead of a bunch of tests. There are too many variables to distill the selection down to the amplifier configuration that you just like. What sounds best is dependent more on your mood than any technical trait of a tube.

[Malcolm Irving]

Without a reference in voltage gain, your test reveals nothing about the inherent noise contribution made by the tube. If one tube is a lot higher mu than another you would expect a lot more noise despite normalized stage gain measurement of noise might be lower....

Well, yes, I did realise that (I mentioned that a lower gain sample would appear quieter in this 'test' in post 1).

However, the noise I was trying to eliminate was buzz. White noise was at the same level for all the samples (subjectively to my ear anyway). The 12AX7EH was the winner because it had the same level of white noise, but no buzz.
The amp was an attempt to create a close replica of an early JTM45. It therefore has 68k grid stoppers (fairly remote from the tube pins), two of which combine in parallel when you use the high gain input jack. The white noise coming from a 34k grid stopper should be higher than that generated in the tube, and if it produces the same level at the speaker this suggests that the gain of all the tubes tested was similar. (OK, based on a subjective assessment - a measurement of gain would have been better.)
The wires coming from the input jacks via the grid stoppers to the grid pins are not screened (except by staying close to the chassis where possible).
On completing the build I was a bit disappointed by the level of buzz (OK for a gig, but a little bit intrusive at home) and decided that the lack of screening on the V1 grid wires was probably the problem. I was all set to replace them with screened cable, but thought well I'll just try a few different tubes in V1 just to see if it makes any difference and was pleasantly surprised that the EH made no buzz at all.
The heater supply has a PT centre tap which is grounded, but not elevated. I can only conclude that the buzz on the other tubes was heater to cathode leakage inside the tubes (either resistive or capacitive).

[nickb]

Something else to consider is that the variation in hum between samples of the same type is large. I recently tested about 100 of a batch of the same type and there was about 20dB range in hum. This would exceed the variations due to construction by some margin, I would think.

[Malcolm Irving]

Something else to consider is that the variation in hum between samples of the same type is large. I recently tested about 100 of a batch of the same type and there was about 20dB range in hum. This would exceed the variations due to construction by some margin, I would think.

Yes, I was wondering about that. Maybe just a lucky tube. I've bought a couple more of the EH tubes to see if the lack of buzz is consistent. Haven't tried them out yet, but will report back if I find anything interesting.