Hi,
It seems I've been spending too much time on Ebay again, this time resulting in a quad of NOS 6S4S (soviet 6B4G equivalents) being on their way from Hungary.
As I have a slight case of PTSD after chasing hum in a 6B4G SET that I built twenty years I also ordered a pair of Coleman filament regulators.
My speakers have a near zero tolerance for hum...:
At first I reckoned my James JS-6112 HS OPTs (2,5k/3,5k:4/8R) would be a perfect match, then I got some indications that people seem to prefer a less steep loadline for the 6S4S.
That made me think of my old, unused LL1688s that I bought fifteen years ago for a transmitter tube amp project that never happened. According to the datasheet they can be connected as 5K:8R. Absurdly oversized, but wouldn't it be fun...?:
It also seems that 6S4S can appreciate a bit of grid flogging, a.k.a. class A2 operation. This rules out the basic "input triode RC-coupled to output triode"-topology, leaving us with either some form of direct coupled follower or interstage transformer coupling as the only realistic options. I succesfully used high gm cathode followers to push some triode wired 807s into deep class A2 in a recent project, so it would be interesting to try interstage transformers this time. Some digging in my storage room rewarded me with this:
The LL1660 seems to have a bit of a shady reputation as an interstage transformer (but supposedly excellent as line output wired as 4,5:1) and the DCR is a bit on the high side for pushing any real grid current. Those LL1671s seem like a better choice here, although the 50mA airgaps calls for some rather meaty driver tubes.
Speaking of driver tubes, here's a short list of some candidates from the shelf behind me:
As for input tubes, I have a small batch of 2C22 / 7193s that would probably do just fine here, being ~equivalent to 1/2 6SN7 with dual top caps.
My mind gravitates towards something like 2C22 - LL1660 - 6AV5 - LL1671 - 6S4S - LL1688.
Surely an ambitious idea and I would very much like to hear any opinions you might have have about it.
Best regards
Daniel
It seems I've been spending too much time on Ebay again, this time resulting in a quad of NOS 6S4S (soviet 6B4G equivalents) being on their way from Hungary.
As I have a slight case of PTSD after chasing hum in a 6B4G SET that I built twenty years I also ordered a pair of Coleman filament regulators.
My speakers have a near zero tolerance for hum...:
At first I reckoned my James JS-6112 HS OPTs (2,5k/3,5k:4/8R) would be a perfect match, then I got some indications that people seem to prefer a less steep loadline for the 6S4S.
That made me think of my old, unused LL1688s that I bought fifteen years ago for a transmitter tube amp project that never happened. According to the datasheet they can be connected as 5K:8R. Absurdly oversized, but wouldn't it be fun...?:
It also seems that 6S4S can appreciate a bit of grid flogging, a.k.a. class A2 operation. This rules out the basic "input triode RC-coupled to output triode"-topology, leaving us with either some form of direct coupled follower or interstage transformer coupling as the only realistic options. I succesfully used high gm cathode followers to push some triode wired 807s into deep class A2 in a recent project, so it would be interesting to try interstage transformers this time. Some digging in my storage room rewarded me with this:
The LL1660 seems to have a bit of a shady reputation as an interstage transformer (but supposedly excellent as line output wired as 4,5:1) and the DCR is a bit on the high side for pushing any real grid current. Those LL1671s seem like a better choice here, although the 50mA airgaps calls for some rather meaty driver tubes.
Speaking of driver tubes, here's a short list of some candidates from the shelf behind me:
- Sylvania 6AV5GA, the (in)famous 6B4G replacement that has been discussed thoroughly on this forum.
- 4P1L. Adding another DH tube would complicate things quite a bit, and my gut feeling tells me they would prefer ITs with a bit more primary inductance (21H estimated)
- 12B4A. Also a bit of a stretch considering the 50mA airgaps
- 6EW7. Dual triode with one low mu/low Rp section, though I only have 3 and they are rare in Europe.
- 6S19P. Got plenty, not sure how good they are as drivers.
- 807. Not the most linear as triodes, but at least they look quite cool.
- PL36 (25E5). Nice triode curves, slightly inconvenient 25V heaters.
As for input tubes, I have a small batch of 2C22 / 7193s that would probably do just fine here, being ~equivalent to 1/2 6SN7 with dual top caps.
My mind gravitates towards something like 2C22 - LL1660 - 6AV5 - LL1671 - 6S4S - LL1688.
Surely an ambitious idea and I would very much like to hear any opinions you might have have about it.
Best regards
Daniel