Tube DIY Asylum

SE input stages and a phase splitting interstage

[ Follow Ups ] [ Post Followup ] Thread: [ Display  All  Email ] [ Tube DIY Asylum ]

This Post Has Been Edited by the Author

I have a PP 6C4C amp (like 2a3) and a 300b SET. Input sections for both are 4P1L driver stage. Input stage of the PP amp is 26, and SE amp is another 4P1L stage. For a while the PP amp - which is balanced through all 3 stages - sounded better than the SET. I thought I was done with SE amps at that point - the sound was excellent. Then gradually I worked on the 300b SET and now I wouldn't go back to PP.

What I learned in this process was:
1) Only use DHTs, no directly heated tubes anywhere.
2) Only use filament bias for the input stages. That meant choosing between 26 and 4P1L. The 4P1L is slightly better.
3) Use the best interstage transformers you can buy.
4) SE input stages sound better than PP stages if you use filament bias.

So I ended up with the schematic shown. Out of all the interstages I tried I liked the LL1660/18mA after the input and the Hammond 126C after the driver. The Lundahl is clearer and has good treble detail, and the Hammond is bifilar and adds a certain warmth and tone. It's better than two LL1660/18mA that way - don't ask me why but it just works.

I haven't built a parallel SET yet - I'm building one as we speak with multiple 4P1Ls in filament bias in the output and O-netics OPTs!!

So at this point I'd say build a parallel PP output stage but use a phase splitting interstage after the driver. Use two stages of 4P1L in filament bias and then a Tango NC16. I haven't used this myself but for the power I think this would be simpler. Stick with the 300b outputs.

I believe the input stages are the heart of the amp and that's where I've put all my development experiments. I avoid larger value cathode resistors and their bypasses like the plague, and I've never found a fixed bias that sounds as good as filament bias, not even battery grid bias. So that's where I am at this point. I absolutely love my amp - voices are just so real they "talk" to you and I listen mainly to vocal music as a result. I never got this with the balanced PP amp - it has all the tone and the detail and is a fraction cleaner on orchestral climaxes. But when I listen to voices it just annoys and irritates me now because that special extra dimension isn't there. Just gotta have my SET fix!! It's really like a drug.

The schematic is substantially correct, though I've played about with some of the values, and also I use glow tubes for voltage regulation of the input and driver stage. I also use choke input DC supplies to the input and driver section filaments, and carefully select the wirewound cathode resistors. I put a lot of work into optimising the first two stages. It's worth it. You may be surprised to see a cheap Russian tube here, but it's kicked out 10Y, 46, 26 and all the other DHTs I've thrown at it! The 4P1L, to give it proper credit, is actually a German Wermacht tube that the Russians copied! Some of those European DHTs were the dog's dangly bits.


Edits: 09/09/12   09/09/12   09/09/12   09/09/12   09/09/12   09/09/12

Follow Ups: