Where can I get RAM super low noise tubes?

Roger, thanks for dropping in. It is a pleasure having someone of your caliber joining the conversation. I hope you don't mind me asking you a few questions and throwing a few comments / observations your way while we have your attention.

Even with checking tubes for noise levels, gain matching, etc... how do you account for variables in sonics from tube to tube and brand to brand ? Obviously, not every tube, even though they might have come out of the same batch and production run, will sound or measure the same. That is why i stated that tubes "almost offer TOO many variables", even amongst the same brand.

I know that people that work on / design gear for a living tend to look at things like this as kind of a science. I also know that the many years of experience have taught you a LOT of the "in's and out's" of what to look for. None the less, i still think that it is almost impossible to know exactly what tube will work best in a component in a specific system without actually trying them out. Obviously, the work that you do somewhat "pre-qualifies" the tubes in that they are not "junk" or "noisy", but that does not specifically tell anyone what sonics they might possess in a specific piece of gear.

As you know, changing just a few of the operating points within the curve can DRASTICALLY alter both the readings and sonics that you get out of a device. This is true of either tube or solid state devices. As such, how do you select the specific settings that you choose to use when testing tubes ? How applicable are those results on a somewhat "universal" basis ? Sean
>

http://www.electron-valve.com/reading/amplifier/tubechange.html

Don't know if you've seen this article by Roger, or if it answers your question, but it is very interesting...

Does not look to me, that the above mentioned article answers Sean's very well founded question. I sincerely hope RM will come back to this thread. He, by the way, was the designer of a wonderful preamp marketed by Beveridge, which a hatchet job review of HP in favour of the SP6 unjustifyably drove off the market and Beveridge into ruin. Cannot remember when that was, must have been in the seventies, It was then, I think, that RM started RAM tubes and his RAM Labs enterprise.

Roger has been responsible for a LOT of very good designs and products. Anyone that has read IAR from the beginning knows his name and would recognize the influence that he's had on the industry as a whole. Let's hope he pops back in and shares some of his knowledge and experience. Sean
>

Sean,

Thanks for the question and thanks to Bullwnkl999 for pointing you to my article "Why and Amplifier's sound changes when you change the tubes". I started designing circuits by ear (and my oscilloscope) at a very early age, before I had studied electronics formally. I felt then and still feel now that circuits affect sound more than parts. Yet parts interact with circuits, and that's the rub. As a mature designer, I design so that part variations have the least effect on my circuits. Therefore, you may hear the least variation when you swap tubes in Music Reference products.

In the examples given in the article mentioned above, the changes were both audible and measurable and due to circuits which, in my estimation, are too sensitive to tube variations.

Now, does a certain tube, say the smooth plate Telefunken always sound the same in every circuit? I don't find it so. Does a tube have frequency response properties of its own? Definitely not. Can a tube change the frequency response of a circuit? Yes it can, but it shouldn't, not in a good circuit.

Although we can't fix overly-sensitive circuits, we can screen the tubes for operating parameters such as gain (mu) and plate voltage for a standard bias value. This is better than just a transconductance test in a Hickok as it measures where the tube will "bias up" in a typical circuit. I find that the self-bias point is the most important factor in sound as it affects distortion directly. We have rejected many of the recent 12AX7 variants as they do not "bias up" anywhere near the accepted (RCA, Sylvania, GE, Telefunken) values. As those companies used the same published curves (have a look they are all the same), their tubes biased up to the same point (except for tube to tube variations). So I agree with your last statement wholeheartedly. We keep a tight watch on operating points, and even have customers savvy enough to request tubes from the upper end, lower end or the middle of the range, as they know what they like. Since all the data is on every tube, we can easily pull what they want from stock.

With all this information on every tube, why buy random tubes where you don’t know anything about the noise, microphonics, gain or operating point.

I'll share a little story on noise. When I was visiting the EI factory in Yugoslavia I mentioned to the engineer that the folks at Sylvania (in Altoona) didn't know how to make a low noise 6DJ8, 12AX7 or any preamp triode. It's mostly a function of cathode coating. When I saw how crude the cathode coating technique was, I wondered how a low noise tube gets made anywhere. They just mix this powder up in a bucket with water and spray it on the nickel cathode sleeves. So I asked the engineer at EI how they got such low noise cathodes (theirs are the best I've seen). He said, "Well we don’t know for sure, its not a repeatable thing, we have learned to mix the coating a few days before we use it, it may be the water, it may be that the janitor takes a piss in the bucket at night". Well, they say it’s the water that makes New York bagels the best. Go figure.