martin manning wrote: ↑Tue Apr 22, 2025 11:10 am
Matt, what is the garter circuit in the EL34 amp discussed in this thread?
That is it, Martin. There are also a couple 220uF Cathode Bypass caps from Cathode to ground as well (only one per cathode).
The grid leak (I think that's what they are called) values are 220K instead of 300K. I've settled on 360 Ohm Cathode Bias Resistors... So I have ordered 4 of them.
I will update my schematic to represent my actual amp. I'm a little behind on that aspect.
I think the discussion around parallels to various undergarments and especially the illustration in the TCJ post https://www.tubecad.com/2009/04/blog0163.htm came up short. The essential feature of this circuit is that it ties the output tubes together, more like this:
You do not have the required permissions to view the files attached to this post.
martin manning wrote: ↑Tue Apr 22, 2025 2:40 pm
I think the discussion around parallels to various undergarments and especially the illustration in the TCJ post https://www.tubecad.com/2009/04/blog0163.htm came up short. The essential feature of this circuit is that it ties the output tubes together, more like this:
Haha! Nice picture. Those really expensive tubes make it look especially classy!
Yes, that is also the article that maxkracht directed me to when he inspired me to try it on my 6V6 Songwriter (adapted from another old Canadian Amp): https://ampgarage.com/forum/viewtopic.p ... er#p470529
martin manning wrote: ↑Tue Apr 22, 2025 11:10 am
This garter circuit is interesting.
Dave, what do you mean by wiring the MV as a variable resistor? Why not use the standard PPIMV?
I'm assuming the circuit below (from Broske TCJ blog post) is what you are talking about.
Matt, what is the garter circuit in the EL34 amp discussed in this thread?
Hi Martin,
For the garter bias circuit, the 220k/220k grid leaks, which would be the PPIMV, are not ground referenced but elevated by the second bias resistor in the garter circuit. So this would mean just terminals 1 and 2 are connected (or 2 and 3 are connected with the output at 2). This is what I am thinking:
[edit: drawing removed to avoid confusion for future readers]
bcmatt wrote: ↑Tue Apr 22, 2025 1:37 pm
Once I got the proper value of 1n5 in that spot, I've found I've been pretty happy with it in my amp. I had been thinking the same thing about adding some sort of control (I also like the Songwriter's Bass and cut combination), but I found that with both cabs I've used it with, I found the bass well balanced now. I've liked this amp with m-65 Creambacks, and have used it with both an open 2x12 and a Mesa Roadster 4x12 (which combines a closed back and open-backed half). Of course, your miles may vary, but you may find you will either are just happy with 1n5, or you could tweak it slightly to your taste and not need to have control. I like that this amp on the clean channel is capable of slightly more gain than the songwriter and I wonder if it's that lack of bass control contributing to a slightly more lively signal. This one having no NFB would suggest that's maybe the reason, but the songwriter's NFB is so small that it's maybe indistinguishable. Of course, I'm comparing my more bastardized versions of these two amps, so ... grain of salt and all that.
That said, I do think the Bass/Cut combo of the songwriter is a beautiful idea. I converted my 5e3 into a 6G3 bright channel with no trem (but keeping the cathode bias). I found it to be such a nice upgrade. I am planning to try the Songwriter Bass/Cut combo instead of the tone control to hopefully give it a bit more control without losing the potential gain on tap. I need to make it a new faceplate anyways.
Hi Matt,
Ok, thanks, that is helpful! If I use the Songwriter bass circuit, I would probably choose 1n for the base value and 47n in series with a 500kA pot. I found that with a 3M-RA pot, most of the useful travel and bass was from 12pm to 3pm on the pot anyway. Yes, that bass control and the 330k/220p treble peaker that follows in the Songrwriter circuit does soak some gain. A 6-position rotary switch might be a good choice for bass as well in this amp.
I also added a similar bass control to a single channel 5E3 once and it was useful. I used a 100k pot with 47k in parallel between the second and third lugs for a custom taper. It worked well and you could hear the sweep in the bass. I used a 22n cap in that amp and it was full but not wooly or bloated sounding.
OK, now I am confused... and it's because of that Blumlein picture I posted. I don't know what that other side of the grid part of the diagram represents where there are caps. I have the suppressor grid (pin 1) tied to the Cathode on my EL34s. My Screens still have just their 1K power resistors. It's the Control grid that I have going to the opposite cathodes resistors through the 220k grid leaks. Does the diagram just represent the control grid on both sides of the diagram?
bcmatt wrote: ↑Tue Apr 22, 2025 6:56 pm
OK, now I am confused... and it's because of that Blumlein picture I posted. I don't know what that other side of the grid part of the diagram represents where there are caps. I have the suppressor grid (pin 1) tied to the Cathode on my EL34s. My Screens still have just their 1K power resistors. It's the Control grid that I have going to the opposite cathodes resistors through the 220k grid leaks. Does the diagram just represent the control grid on both sides of the diagram?
The TCJ illustration shows triodes, just control grids are cross-connected, and it shows two connection to the same grid element. Dave wants to do a PPIMV, where the pot element replaces the grid leak resistor connected to the bias voltage.
So with the schematic I marked up above the PPIMV won't shut the volume all the way off. You still have the lower Rk forming a divider with the pot element, putting the pot wiper a few hundred ohms from ground. It won't be loud, but to cure that you'd need to put caps on both sides of the MV pots.
