Using alternative tubes with Reduction + Integration

Pflugshaupt · 4164

0 Members and 1 Guest are viewing this topic.

Offline Pflugshaupt

  • Jr. Member
  • **
    • Posts: 26
First of all, please excuse my ignorance, I just started to get into tubes about a year ago, built two bottlehead kits (crack w/speedball and reduction w/integration) and now I'm fooling around with tubes, eager to learn more about how these things work. I have a question about using different tubes with reduction + integration using octal adapters of various kinds.

I'm just trying some things coming from modding my crack to use a 6SN7 tube. So just for fun I put two 6SN7 into the reduction + integration and was frankly amazed it worked. However of course gain is quite low as 6SN7 have a mu factor of 20 while the 6992 in the reduction have 33. So with my noob approach I started looking for some other tubes and I really like octal sockets for some reason. That's how I found out there are 6C8G tubes somewhat similar to 6SN7 that have a mu factor of about 36 and use the same 300 ma/6.3v heating current the 6992's use in the reduction. I couldn't resist trying some of these in the reduction... but as you probably have guessed by now - it doesn't work - I get heavy distortion and low output. Now my big question is what is the main reason for this and would it theoretically be possible to adjust the reduction + integration circuit to make them work?

From reading the sheets I think the big difference between the two types is the plate resistance/current. So secretly I'm hoping I could somehow lower the current somehow and that would magically make things work. But it's of course quite likely I'm looking at this the wrong way so feel free to smash my hopes :)

For reference these are the numbers I looked up (maybe others are more relevant?):

6992 <-> 6C8G(VT-163)
heater voltage: 6.3 V <-> 6.3 V
heater current: 0.36 A <-> 0.3 A
plate vmax: 130 V <-> 250 V
plate current: 25 mA <-> 3.2 mA
plate dissapation: 1.8 W <-> 1 W
plate resistance: 3 kOhm <-> 22.5 kOhm
mu factor: 33 - 36





Offline Paul Birkeland

  • Global Moderator
  • Hero Member
  • *****
    • Posts: 19316
Reply #1 on: April 30, 2018, 04:22:25 AM
For the 6SN7 in the Crack, you can go all the way back to the #27, #37, #56, and #76 that are ancient indirectly heated triodes (the 27 was one of the first indirectly heated triodes in fact).  These morphed into the 6J5 which was essentially the same tube in an octal base, and then the 6SN7 which was two of the same tube in an octal base.  The 12AU7 and 6CG7 came out a bit later as tubes shrank and 9 pin bases became the norm.

I go through all of this because there is no octal tube before the 6922 that is its predecessor. 

The 6C8G has about 15% of the transconductance that the 6922 has.  This will make it a lot noisier, cause the plate voltage to be way too high at the 6922 operating points, and its higher internal resistance will be woefully unable to drive interconnects and the RIAA EQ properly.

The biggest issue you'll have with substitutes is the RIAA EQ, which will be intolerant of any substitutes with higher plate impedance.   Even the 6N1P, which is very close to a 6922, isn't quite close enough to run without some light mods.

Paul "PB" Birkeland

Bottlehead Grunt & The Repro Man


Offline Pflugshaupt

  • Jr. Member
  • **
    • Posts: 26
Reply #2 on: April 30, 2018, 05:26:53 AM
Thanks heaps for this information! Basically it's as I feared. I indeed was trying to find a predecessor of the 6922 with an octal base, but I couldn't find anything closer than these 6C8G tubes.