I have a Williamson octal chassis (6sn7-6sn7-el34x2) that I would like to change to a long tail phase splitter for comparison. It appears that most Mullard circuits use ef86 at the input stage. Are there any good octal pentodes for that purpose (don't want to mess up the original chassis sockets)? Any one tried 6sj7? TIA.
I think the EF6 (precursor to the EF86) was available as an octal, though I normally see it using the 'P-Type' socket. The metal EF12 is another version which uses a European 8-pin socket which looks nothing like an octal - I have one of these in my stash and have never seen a socket for sale.
You can also try the 6J7/EF37/6C6/310
or as you said, the 6SJ7 - it was a popular american tube for plenty of amplifiers back in the 40s.
Many years ago I built a SE amp using a 6C6 (grid cap) pentode driving a RCA 2a3 - very nice little amplifier for the money.
You can also try the 6J7/EF37/6C6/310
or as you said, the 6SJ7 - it was a popular american tube for plenty of amplifiers back in the 40s.
Many years ago I built a SE amp using a 6C6 (grid cap) pentode driving a RCA 2a3 - very nice little amplifier for the money.
The 6SJ7/5693 is a nice tube, but its gm is lower than that of the EF86/6267. I recall seeing EF86 into 12AX7 LTP in a Mullard circuit. The closest you'll get to the 'X7 in Octal is the 6SL7. A 6SJ7 into a 6SL7 might come up short on gain.
In any event, high gm types make BETTER phase splitters (both LTP and "concertina"), as vulnerability to slew limiting is reduced. Also, high gm types have low Rp, which gives them better drive capability. AFAIK, you will not do better than the 6SN7 in the gm dept., when using Octals. If a 6SN7 is used as the LTP, the issue of insufficient gain is exacerbated. You can resolve the gain issue by using the high gm 6AC7 instead of the 6SJ7. If you look at the 6AC7 data sheet, you'll notice a remark about DC heater power when amplifying low level audio signals. In the power amp case, I/P is at line level. Biasing the heater off B+ should be adequate to control hum.
In any event, high gm types make BETTER phase splitters (both LTP and "concertina"), as vulnerability to slew limiting is reduced. Also, high gm types have low Rp, which gives them better drive capability. AFAIK, you will not do better than the 6SN7 in the gm dept., when using Octals. If a 6SN7 is used as the LTP, the issue of insufficient gain is exacerbated. You can resolve the gain issue by using the high gm 6AC7 instead of the 6SJ7. If you look at the 6AC7 data sheet, you'll notice a remark about DC heater power when amplifying low level audio signals. In the power amp case, I/P is at line level. Biasing the heater off B+ should be adequate to control hum.
Eli Duttman said:
Thank you for pointing out the potential gain issues. You are right that the Mullard 5/20 uses EF86 into ECC83. I was trying to follow the EICO HF-22/35 examples that have the EF86 into 6SN7. Since I do have a couple of 6SL7s at hand, probably will try those first before looking for 6AC7s.
I use EF37A (a low noise version of EF36) as an octal eqiv to EF86. Very nice sounding and much quieter than the 6J7 or 6SJ7.
They are also bright red, and were designed for early British computers. In fact, the first ever electronic computer used EF37A to break German codes in WW2...
http://en.wikipedia.org/wiki/Colossus_computer
They are also bright red, and were designed for early British computers. In fact, the first ever electronic computer used EF37A to break German codes in WW2...
http://en.wikipedia.org/wiki/Colossus_computer
- Status
- This old topic is closed. If you want to reopen this topic, contact a moderator using the "Report Post" button.