Playing with the Songwriter Preamp

[bcmatt]

I had a cheap amp with 2 12ax7s and 2 6V6s that I tried out the Blumlein Garter circuit for cathode biasing.
https://ampgarage.com/forum/viewtopic.php?f=6&t=37548
I ended up switching to much of the Songwriter Preamp (from what I think I could find out about it so far). I wouldn't mind some opinions on a couple things from those that had some experience with Songwriter type builds.

Garnet 45B Oct 25 Mismatched Voltages.jpg

I'm loving the simple and useful controls so far, but a couple things I may be doing wrong:
Is the bass pot not supposed to grounded?
Does anyone have an opinion on NFB? I haven't implemented it yet at this point, but I'm thinking of trying a couple values on a switch. Any ideas what to try for this circuit?
The other thought I have is whether there is maybe an alternate V1 cathode situation that I could put on a switch for an option to give the gain a boost. Any thoughts on what to try? I imagine that v1 is probably pretty centre biased already, so perhaps a smaller cap value... That I would imagine would change the voice a fair amount as well. Thoughts?

[bcmatt]

It's a fun combo setup to play, but it's the biggest pain in the butt to get the chassis out; have to remove the speaker baffle to get the chassis out to work on. That's why I'm going for a bit of inspiration before opening it again.

20241026_091604.jpg

20241026_115625.jpg

20241026_093246.jpg

[Colossal]

The bass pot is NOT grounded. It is just providing variable series resistance with the 100n cap.

For NFB, the original has 220k/4k7 on the end of the tail of the PI (4k7 in series with 39k tail, not 47k). You could certainly replace the 220k with 250kB with a fixed amount of resistance at one end.

For V1, since it is run as a parallel triode with 220k load and 1k5/1uF cathodes, if it were a single triode, this would be 440k load resistor and 3k3/500n cathode; quite a flat load line. The original amp's V1ab B+ is around 293VDC. Gain is about 37dB, pretty high.

[bcmatt]

The bass pot is NOT grounded. It is just providing variable series resistance with the 100n cap.

For NFB, the original has 220k/4k7 on the end of the tail of the PI (4k7 in series with 39k tail, not 47k). You could certainly replace the 220k with 250kB with a fixed amount of resistance at one end.

For V1, since it is run as a parallel triode with 220k load and 1k5/1uF cathodes, if it were a single triode, this would be 440k load resistor and 3k3/500n cathode; quite a flat load line. The original amp's V1ab B+ is around 293VDC. Gain is about 37dB, pretty high.

Thanks Colossal!
I did not know about expected voltages for V1. I was playing with my dropping resister values and where they feed from, but I think I need to put the 1M resister to ground at the end of the dropping string... maybe that will balance things out a bit more since my v1 plates are at 140V. Do you know what the PI plates should be expecting in the original design? I don't have a choke, but I will maybe make my dropping string line up a bit more with the songwriter values.

I'll experiment with those NFB values... although I would need to adjust for only having an 8 ohm tap. I'll see if I can find a 4k7 for the end of the tail (39K tail) and then I'll experiment with a 100K resister in series with a 100Kb pot and see what it does.

[Colossal]

The math says that for V1ab with a B+ of 293VDC, the plate voltage (across 220k for paralleled triodes ) would be around 119V and the cathode around 1.3VDC.

The PI supply voltage for the SW30 is 282VDC and the plates should be around 200-210V or so. 47n output coupling caps, bias is 1k2 and the tail at 39k/4k7.