A New 12XXX for Ted B.

[Ted B]

Yes, I agree in that no bright cap seems to be a bit too far in the other direction. With the 100pF however, I didn't notice any perceived loss in gain. I'll wait a few days and repeat the testing.

I noted earlier that I (very carefully) clipped in an additional 16uF filter cap at the PI while playing. I didn't notice any difference. I'll look at a larger cap and try the screens ... maybe.

I plugged in a pair (V1 and V3) of the current production Russian 'Mullards'. I find they sound rather good.

[vanhalen5150]

The answer to question #1 is, your amp would be noticeably stiffer with less squish.

Fair enough. I'll play with it a bit more and try to decide where my preference lies. It still has a touch of an 'organic' feel at this point, which is just enough to let you know it's an old circuit. I'm thinking the stiffer feel may be a little bit too much 'polyester', but I'll ponder it.

I'm no expert with this stuff, but upping the bias tap or PI filter cap might be an easier place to start instead of the screens if you want a little more head room. I guess it depends on what you are going for. If you have a single 32up for the PI, A/B a dual 32/32 and see what that does. Interesting about the bright cap, with it, my amp seems to jump in at around 3 on the volume. Still, with it in, when you get to 9 or 10, the magic presents itself in spades, more so than when the bright cap is not there.

I often wonder about that bright cap. I think its something that translated over to stock none MV amps from the MV amps like the 2203/2204. If your running all tubes opened up, I don't see the benefit of having that there, I'm not familiar with the MV amps but is that a correct assumption? Did that end up being put into the stock values based on MV amps and running amps through attenuators because they do suck the high end out of your tone.
Am I making sense?

[neikeel]

Nice amp, I like the black chassis and board. The external bias looks useful too.

I think I can chip in re: bright caps. I have 5,000pF RS ceramic on my '68 and 5,000pF RS tubular ceramic on my '67 SL (ie both stock). On both these amps the main effect is as a boost of all frequencies at lower volume settings such that on the '68 the amp is almost full tilt at 2.5 on the bright volume (original CTS 1meg pot). The 67 is shared cathode and with humbuckers only really breaks up around 6 on the dial (original RS 1meg) and has a nice gentle taper so nice peudo clean sounds between 4 and 6 on the dial.
The difference appears to be on my clones and more modern metal panels. They are much brighter, possibly louder with a bit more distortion. I use Metro or Marstran trannies on the clones and Drakes/Dagnalls as per year/model on the metal panels. I found 5,000pF on a clone of my SL and one of my 50watters unbearable, like you. Fine if you use a 30' long curly cable I guess and/or need to cut through the fug of loud stage volumes in a metal band . I agree that 100pF gives the best 'sparkle' on these amps (I use 100pF on my JTM45 clone one of my 50watters and 500pF on my 45/100 clone all found by trial and and error). Similarly my ' original 68 50watter also runs a 5000pF Lemco ceramic disc and it is not harsh or overly bright at all (I usually go straight in to Hi 1 without jumpering). To show there is an exception to this I do have a 1987/2204 switchable with Marstran iron that has 5,000pF ceramic disc (cheap no name) in 1987 mode and 1,000pF on the 2204 and it is as sweet as a nut .

Regarding the 'boost cap' on V2a cathode I generally prefer a 0.68uF with 47k/8ohm NFB and PI output couplers no bigger than 0.047uF as I think that although it does give a wide frequency range boost I lose some definition and clarity if I try to clean up with the guitar, made worse by big PI output couplers (0.1uF). Obviously if you are running 100k/4ohm NFB then the response will be different and if the PI caps are 0.022uF too.

I like 48uF in the PI of the 12 series amps with 16uF on the screens. I found 32 or 64 made little difference but 100uF definitely stiffer. But like low screens filters I think it is the feel and response of the amp/guitar under your fingers that is different that affects the way you (by that I mean I play ) not sure it would sound different in clips etc.

Sorry to ramble, just sharing my experimentation results

[Ted B]

Your thoughts are much appreciated Neil. It definitely gives me different things to ponder.

The 5000pf Ch1 bright cap is a subject within itself. I can see that an earlier circuit with shared cathode, larger coupling caps, greater NFB, etc., it would be more forgiving, but with my amp, the on/off switch effect and added brightness is just too much to be useful. In an outdoor setting, where treble is the first thing to be lost, I can better see the benefit. In a 20'x20' room (where my amps reside presently), and without any peripherals that potentially suck a bit of treble (effects, attenuators, etc.), it's not working. My hearing still works to beyond 18khz, so that may be a factor as compared to some other users. Another thing to consider is when used with the '003' Pulsonic cones used by Celestion at the time the this value bright cap may not sound quite as bright as the '1777' Muellers, and reissue G12M and G12H as well. If one is using bass speakers (55hz resonance), the obtrusiveness is further mellowed. In my situation, the older 100pf value seems to constitute a good compromise, and the amp's volume control has a greater range of sound.

About the V2 cathode 330uF fat cap, I've decided I like it because I like the added balls it gives at moderate volumes. As the volume is increased however, the lower frequencies it passes tend to become too distorted. But since Marshall's tone controls are subtractive, I've found that I can effectively dial it out. I've A/B'd the amp, leaving only the 0.68uF and clipping the fat cap in and out. I prefer the fat cap, but I'd feel differently if I couldn't dial it out.

I observed no real audible difference between 32uF and 48uF on the PI, so I left it as-is (32uF). Presently, I'm using 16uF (32x32 in series) at the screens, and I like the sound as well. It doesn't seem noticeably flabby when pushed, so I'm not inclined to increase the filtering there.

In another recent discussion [http://forum.metroamp.com/viewtopic.php?f=13&t=36686], I noted the bright screens on the EH 6CA7s while playing. I know this is 'normal', but it's a bit more noticeable than I recall. This has me wondering about the 4-ohm, 100k NFB setup I'm using, as this could be an artifact of low nfb. My old amps from this era used 8-ohm, 47k, but went this route to check out Friedman's observations. Given the sheer amount of gain this amp has on tap, I'm wondering if the 8-ohm, 47k configuration might take some of the fizziness out of the highest volume settings. I'd like to hear from anyone who's compared the two before I break out the soldering iron.

So there, a few ramblings of my own!

[Ted B]

Well, it seems I found the reason why my amp seemed noisier than expected. It's this particular (100k-4ohm) NFB configuration:

http://forum.metroamp.com/viewtopic.php?f=4&t=36695

[StuntDouble]

I'm wondering how much 100pf caps paralleled across the p.i. plate resistors would help with that? I've heard that it knocks off some of the harsh high-end and helps clean up the noise and fizz.

[Ted B]

Good question. I just tested it while playing.**

It made a subtle change, and did dumb down some of the higher frequencies slightly, but not in a way that I liked.

Every way I've sliced it, I keep going back to the lower noise, less fizzy 47k - 8ohm NFB configuration. I think it's going to stay that way.


** Randy Kilowatt Says: Don't try this at home, kids!

[StuntDouble]

Interesting, what affect did it have on the sound when you tried it? Have you tried a 100pf p.i. cap as well? I'm thinking that with a 1.75k OT, 100k on the 4 ohm tap may just not work.

[Ted B]

The test I described in my post just above was a 100pF PI cap vs. 50pF.

[StuntDouble]

The test I described in my post just above was a 100pF PI cap vs. 50pF.

Ok, gotcha; I don't think bumping the p.i. cap up from 47pf to 100pf does much of anything; I was curious if someone else was hearing the same thing I did when I tried it. How did paralelling the 100pf caps across the 82k and 100k plate resistors, affect the sound? What did you notice?

[Ted B]

As I indicated, clipping an additional 50pF in parallel made little difference, and yielded nothing useful. If one is looking to tweak the character of the amp, I didn't find this to be worthwhile.

[StuntDouble]

Yeah, bumping that p.i. cap up to 100pf doesn't do jack...at least to my ears. I was referring to paralelling a 100pf cap across the 82k and the 100k, so that each of the 2x 100pf caps are in parallel with each of the resistors, but that's ok...nevermind. The build looks very cool.

[Ted B]

Yeah, bumping that p.i. cap up to 100pf doesn't do jack...at least to my ears. I was referring to paralelling a 100pf cap across the 82k and the 100k, so that each of the 2x 100pf caps are in parallel with each of the resistors ...

No, that just makes a low resistance shunt around the plate resistors that introduces signal into the power supply. I don't think you'd want to do that.

[StuntDouble]

I'm not convinced it would work either.

http://forum.metroamp.com/viewtopic.php ... ss#p348315" onclick="window.open(this.href);return false;

I was reading this and found it interesting; I was just wondering if it did help stabilize the NFB loop and suppress noise in the high end.

[Ted B]

That seems unconventional at a glance. Basically, this shunts high frequencies from opposing sides of the PI into the B+ feed. Apparently however, those frequencies get neutralized in filter cap hell, where they are never seen or heard from again.

After playing my '#1' (shown below) as if I were in a stadium (no attenuation) through the NFB settings where I've put them (47k - 8ohm), I'm liking everything right where it is. Even with Tom Holmes H453 PAFs (8.2k - A4), the amp is positively ripping your head off at 3/4 volume with very hard driving, but controlled distortion. The key to finding the sweet spot there seems to largely depend upon the Bass knob setting. Roll the guitar's volume back, and it cleans up nicely, and not half bad sounding either. Classic rock textures can be found at various points in between, thanks to the flexibility of the 100pF Ch 1 bright cap, which IMO, is far more useful than that 'on/off' switch 0.005uF. I do like the extra gut punch the V2 cathode fat cap gives. It audibly rattles my 4X12 1960TV cab. What's nice is it doesn't have the faintest hint of the plastic, processed sound typical of modern cookie cutter hardware. It still sounds (and smells) like a vintage plexi, but this one gets really, really angry.