|
Audio Asylum Thread Printer Get a view of an entire thread on one page |
For Sale Ads |
150.148.218.189
In Reply to: Re: Question for Ralph on MP1 posted by Legendre on August 14, 2005 at 16:32:37:
The problem is current; all high Gm tubes need a lot of current. (Want high Gm, because in a cascode, gain = Gm*RL, where Gm is the transconductance of the bottom tube section, as you know but maybe others don't.) I came to the conclusion that the Atma-sphere solution in the context of the MP1, to use a pair of 12AT7s in parallel to form the "bottom" part of the dual-differential cascode is a pretty good one, because 12AT7s will give pretty good Gm without needing such large amounts of current. To do better than that, one would probably need a separate supply. (You tell me; I doubt that the tranny in the MP1 can produce the current needed at 250 to 300V, but maybe I am wrong.) So, the tubes I had in mind for the bottom of the dual-diff amp might include the 5842 or the ECC99 or 5687, if there were more current available. There are other more esoteric ones too.
Follow Ups:
I think the problem you run into is whatever amount of current that is possible through the bottom tubes has to go through the top as well. At this point the only way to get the current is to reduce the plate resistance, which will increase bandwidth, but at a loss of gain.So here is where you need to boost the B+ (and maybe B- voltages), partially to get the current but also to manifest the voltage swing with the same or larger plate resistance at the increased current that these other tubes will allow. Of course, the 12AT7s will allow for that too. In several different ways you run into practical considerations that result in limited improvements. If you had some planar triode tubes (from the last years of tube development), they had some interesting qualities- gain of 200 with high current (@ 10mA, that sort of thing) but they are impossible to obtain. They would be interesting as so much gain is available from a single tube section.
More recently, cold cathode carbon nanotubes have shown more promise, although they have seemed quietly on the back burner of research for the last 20 years.
You wrote, "I think the problem you run into is whatever amount of current that is possible through the bottom tubes has to go through the top as well."I thought about that problem, too. Actually, there is a way to shunt some of the current that passes through the bottom stage so that it does not all have to pass through the cascode tube section, using resistors and a bypass cap. It's well described in Morgan Jones' book; I don't remember the details, and I have no idea whether it would per se affect sound quality.
There was a time after I first installed the Koetsu when I was thinking night and day about this problem, in my spare moments. Then I gradually came to find, as the cartridge broke in, that the gain was acceptable as is. Only during the last month of this miserable summer have I had problems with gain (as I said, probably due to the effects of humidity on the mylar in my M1 spkrs), which has started me thinking again about the input stage of the phono section. Possibly, things will get better again in the fall. Possibly also, as you imply, one could squeeze a little more gain out of the stock 12AT7-based stage, just by increasing the voltage and current another notch or two. Maybe also my 12AT7s are aging; they have seen quite a bit of use but under the very low stress conditions of the MP1 circuit.
Hi -Don't know how much gain you're currently getting from the 12at7 part of the cascode stage, but wonder if maybe a 12av7 would buy you anything. Wouldn't have to rewire the filament, at least.
I once looked into the more esoteric 12A_7 series of tubes, like the 12AV7 and the 12AY7. Can't recall whether the 12AV7 affords higher Gm and similar or lower Rp vs the 12AT7. Do you know? If and when I get serious about this, I will re-visit the 12AV7 possibility. Beauty of the ECC99 is that it also plugs right into a 12A_7 socket.
This post is made possible by the generous support of people like you and our sponsors: