Quick question - I just ordered the bits required to make a Gary Pimm battery biased CCS, and I was wondering if it's possible to simply use a CCS loaded 6SN7 as a single stage preamp - each half as a channel. Would this arrangement work and be able to drive cables / input stages?
It depends on the cables and the load. Assuming there's no cathode degeneration in the 6SN7 circuit, the source impedance will equal plate resistance. For usual operating conditions, that will be between 7-10k. That's pretty high. If your power amp has a high input impedance (more than 100k) and the cables are short, you can probably get away with it.
If there is cathode degeneration, that makes things worse; the effective plate resistance will be increased by a factor of 21 times the unbypassed cathode resistance.
If there is cathode degeneration, that makes things worse; the effective plate resistance will be increased by a factor of 21 times the unbypassed cathode resistance.
I see. I sort of got it into my head it would be able to swing roughly the bias current into the load if needed - if only things were that simple. I guess I'll be using a cathode follower!
Or you could just choose a more appropriate tube for the job. Same topology, but with a 5687, will give you roughly the same gain and a Z out of about 2 to 2K5. A 12B4 will give you less gain (which might be a good thing) and an even lower Z out.
Maybe a mu-follower would be better for your needs? It would use both halves of the 6SN7 (or other double triode), would give practically the same linear performace and gain as a CCS load and would have the low OP impedance of a cathode follower.
The biggest advantage, though, is that you wouldn't need to use the Gary Pimm CCS, so you could save it for another project. 😀
The biggest advantage, though, is that you wouldn't need to use the Gary Pimm CCS, so you could save it for another project. 😀
Hold on - If I use the mu-output of Gary Pimm's battery CCS won't that basically achieve the same thing?
Will,
If you use Gary Pimm CCSes to load 'SN7 sections and take advantage of the mu follower O/Ps, you will end up with too much gain and a "hair trigger" volume control. Expect a voltage gain of 20X, which is the full mu of the type.
You can convert that excess gain from a liability into an asset. Use Sowter model 8940 3:1 step down trafos at the O/Ps. 😉 Talk about being able to drive a hellish load. 😀
If you use Gary Pimm CCSes to load 'SN7 sections and take advantage of the mu follower O/Ps, you will end up with too much gain and a "hair trigger" volume control. Expect a voltage gain of 20X, which is the full mu of the type.
You can convert that excess gain from a liability into an asset. Use Sowter model 8940 3:1 step down trafos at the O/Ps. 😉 Talk about being able to drive a hellish load. 😀
Thanks for the responses everyone 🙂 Another quick question - Will running L & R through each half of 6SN7 result in unacceptable crosstalk?
In theory, it could do so. However, there are several cases of people doing this with no audible cross-talk. The Leak Stereo 20 used a shared L-R 12AX7 as its input stage.
Aha, excellent to hear 🙂 Those separation curves look good enough SY, and as for the leak stereo 20 I can attest to its good channel separation because I use one! 😎
While I'm here - Is it possible to decrease the gain by putting some sort of feedback between the anode and grid?
<snip>
Forgot that the load is a current source, cathode degeneration actually has little effect on the gain in this scenario. Choose a tube with lower gain. (Maybe a 56, 76 or its octal equivalent the 6P5 - all singles.) Feedback like you propose is possible, but the additional loading imposed defeats the purpose of the ccs in the first place, and it creates issues with the choice of volume pot as the input impedance will now be rather low if you want good ac performance.
Sitting out in the open like that it is likely that it is picking up stray fields, and it could also be oscillating if there is no grid stopper resistor close to the socket. Most of my bouts of oscillation occurred in bread boards that had a lot in common with yours.. Long leads randomly arranged, etc. 😀
Forgot that the load is a current source, cathode degeneration actually has little effect on the gain in this scenario. Choose a tube with lower gain. (Maybe a 56, 76 or its octal equivalent the 6P5 - all singles.) Feedback like you propose is possible, but the additional loading imposed defeats the purpose of the ccs in the first place, and it creates issues with the choice of volume pot as the input impedance will now be rather low if you want good ac performance.
Sitting out in the open like that it is likely that it is picking up stray fields, and it could also be oscillating if there is no grid stopper resistor close to the socket. Most of my bouts of oscillation occurred in bread boards that had a lot in common with yours.. Long leads randomly arranged, etc. 😀
Kevin, I haven't done this specifically with the Pimm ccs, but with other types, I found that cathode degeneration greatly increased distortion.
Another odd thing I noticed
If I held the disconnected end of an rca cable connected to an amp and just held it near the tube and the leads carrying the output signal, I could hear the music and it seemed like all frequencies were present from what I could tell. Is this oscillation?
If I held the disconnected end of an rca cable connected to an amp and just held it near the tube and the leads carrying the output signal, I could hear the music and it seemed like all frequencies were present from what I could tell. Is this oscillation?
- Status
- Not open for further replies.