[Roger A. Modjeski]

My apologies for being away from the circle for some time. Expansion at Ram Labs has taken my attention but we are about done for awhile. As I often do, I am writing on a topic that has come to my attention from several sources in the past few weeks. Must be in the air....

MATCHED!! Yea sure. What vendor is going to say they aren't? Problem is they match triodes for Gm (cause that's what the better testers tells them). What we want to know is Mu (voltage gain). The channel balance of your preamp is dependent on Mu and matched Gm does not insure matched Mu. Recently a friendly competitor (I do test for some who care) sent me tubes where the sections read 20% unmatched for Gm. However, their Mu was very close. Those guys who make dual triodes (12AX7, 6DJ8 etc) appear to watch the Mu more than the Gm because they know Mu is more important. I know that too and all RAM triodes are measured and matched for gain and it's right on the label.

Please keep in mind that even the venerable Hickok Gm testers were made for TV/radio repair shops. They are not laboratory instruments. Would a laboratory instrument indicate a Gm the was almost twice the typical value? That's what I see all the time on eBay. That's 100% error... it that precision...?

[Roger A. Modjeski]

I'm getting the feeling no one is interested in this topic yet it is a very important one.

[JoshK]

It is important and interesting.  I guess the question I will ask is if you match for mu and not gm, then won't the Ibias be very different across the two stages?

[Roger A. Modjeski]

Josh, I'm not sure I understand "across the two stages"? Do you mean between the two sections? 

In any event, bias is not strictly determinded by gm or mu. That's why we measure bias separately and give the actual plate voltages that the tube biases up to in a typical circuit. We do want that to be reasonably close.

Until recently, when BAT changed their phono circuit from manual balance to automatic balance (of plate voltage) we provided them tubes with no more than 3 volts difference between the plates of section 1 and 2. We still provide these to customers and they are advantageous even in the self adjusting circuit.

[dangerbird]

---good read--food for thought.

[Roger A. Modjeski]

---good read--food for thought.

thanks for bumping the topic.

[neobop]

I find this topic particularly relevant. The guy who designed my amp said that almost no one knew what they were doing when it came to tubes and biasing. I bought around 10 Siemens EL34 tubes and he sorted them much differently than anyone else. I didn't understand plate voltages and stuff like that, I still don't. But I know one thing, after spending a few hundred on NOS tubes, next time it will be strictly RAM. Maybe sooner than later.
neo

[rbwalt]

good choice on getting your next tubes from ram labs/ music reference! i do not think there is anyone else out there that tests and grades tubes like roger does. i will only buy tubes for my music reference amp from roger.

rob.