Has anybody ever try the diode trick?

[Dandrix]

Tone trick,

Insert a diode after resistor #2 (ab763 fender supply cap section). IN4005 with the stripe towards C supply cap. The diode is between the end of the 1k resistor and C. This "rectifies" the preamp section of the amp, keeping the power supply from "pulling down" the preamp when it makes excess power demands. It would make the amp more distinct and quicker to repond. reduces sag???

so, i tried this mod with my meter hooked up to C supply and then D supply.
played an E run texas blues shuffle thing, and the voltage would still dropped. between 5v and 10v depending how hard i would get into it. i found the tone tick (sag reducer) not doing what it was going to do.

In theory it was to add at C supply more capacitance, would increase voltage and reserve for the drivers (preamp section?) better power, less cross-over disortion and unbalance. It improves "transient response".

any thoughts???

[daz]

any thoughts???

Yes....thats exactly why i stopped using a seperate preamp and power amp rig for about a year in the 90's. When the pre and power amp have seperate power supplies it kills one of the most important (THE most for me) aspects of tone IMO....dynamics. When you play soft, then hard to vary tone and OD amount, the draw on the PSU is what gets you those dynamics. When you make the preamp stiff like that you kill any dynamics generated there, and thats even worse with a cascaded pre. Give it a try if you wish, but to my way of thinking that would be like purposely making your amp worse. You'd be sorta turning it into a HI-FI. I mean, theoretically i know nothing about that trick as far as whether it'll work as you say, be safe, etc etc. I'm not a tech, just a player who likes to mess with and build amps with minimal knowledge. But i do know about tone an dynamics is a huge part of what a guitar amp is about. It in fact separates the great amps from the dime a dozen.

[Chuck H]

There IS a new tone emerging in guitar amps. That of the multiple cascade preamp with a fairly clean power amp. With an EQ placed between it affords the uber gain thing with huge frequency response options. I'm with you in that dynamics are more fun with the power amp and preamp both clipping and getting their relationship on. The amp is really another instrument at that point. But the modern sound is really about harmonics and compression. Modern uber gainers are more about signal processing. Some of the metal tones used are actually sort of, um, blatty? Just flat and bland. Sure the pinch harmonics jump out as loud as the fundemental, but at what cost!?! This goes along with what your saying about a stiff preamp. Because in such amps the power amp is NEVER clipping. So there is not much power supply sag. If it did the huge bass associated with bedroom monkey tone wouldn't be possible. But I do like SOME of what's being done by contemporary players with this uber gain sound. There is no absolute wrong in tone. The limitation is vision. The best players can hear something musical about anything they plug into and use it. I aspire to be more like that. And I have a bunch of ideas for uber gain amps too. Maybe I'll even get around to building one some day. Until then, I'll stick with the tre cool power tube clipping, high gain beast that I have

[daz]

Well, yeah, i wasn't considering heavy metal/thrash etc. The stuff where your guitar volume never goes below 10 and the gain on the amp usually doesn't either. I always just see things from my perspective which is guitar as versatile instrument meant to do many things. And I think most players fall into that category to some extent at least. But those who play the stuff i mentioned, yes, you're right. Dynamics mean nothing when high gain and chugging the low E is all important nearly 100% of the time. But man, without the PA and preamp interacting as they fight over available current, theres just no reason to play electric for me. In fact Chuck, thats why i built my own amp to begin with....to get the kind of dynamics and feel i love more than i was with my marshalls. If i were stuck with that stiff tone i'd sell everything except my acoustic and be done. In any case, i'd never heard of this, but i can see why it would be getting popular.

[Chuck H]

Sure it's getting popular. Well... Getting is dated. It's BEEN popular longer than I've been out of touch!?! But players have experimented with a fully saturated sound since the early 60's. It's better now than it was then I guess. And it's trendy. I suppose we all just got tired of the same old guitar sounds. Kids these days with all their crazy rock music (Ha ha! ).

As far as getting that uber gain sound and still having some dynamics I think the amp design world peaked with the SLO 100 and Dual Rectifier. Two amps that are VERY good at achieving all the harmonic mayhem one could want while still offering some thunder and lightning. Everything since (and trying to over do those designs) is just a signal processing mess with a flat square wave. No balls or character. Personally, I prefere a tad less gain. Just enough less that I can still tell what type of guitar I'm playing through . Most modern uber gain designs are just freakin' blenders set to frappe! My personal amp will (almost) do modern metal tones. But I get bored in about five minutes and dial the gain down and the volume up to satisfy my personal cravings to hear some wood.

[daz]

Sure it's getting popular. Well... Getting is dated. It's BEEN popular longer than I've been out of touch!?! But players have experimented with a fully saturated sound since the early 60's. It's better now than it was then I guess. And it's trendy. I suppose we all just got tired of the same old guitar sounds. Kids these days with all their crazy rock music (Ha ha! ).

As far as getting that uber gain sound and still having some dynamics I think the amp design world peaked with the SLO 100 and Dual Rectifier. Two amps that are VERY good at achieving all the harmonic mayhem one could want while still offering some thunder and lightning. Everything since (and trying to over do those designs) is just a signal processing mess with a flat square wave. No balls or character. Personally, I prefere a tad less gain. Just enough less that I can still tell what type of guitar I'm playing through . Most modern uber gain designs are just freakin' blenders set to frappe! My personal amp will (almost) do modern metal tones. But I get bored in about five minutes and dial the gain down and the volume up to satisfy my personal cravings to hear some wood.

I like hi gain tones. Mine will do them with a clean boost and sounds fabulous. Harmonic complexity off the hook and just sweet as hell. And i DO like that as well as SRV Lenny tones and everything in between. But the high gain stuff it'll do while sweet and lush isn't the same. That stiff chugging high gain stuff is particularly made for todays 7 strings. I think thats the main reason they are doing these kinds of designs.

I'm the same as you....i like just enough gain to hear the wood still and every guitar is so individual that it's nearly like comparing a trumpet to a tuba. Then you roll down the volume pot and cleans start coming and it's dynamic and you can get dirty and clean back and fourth and drastic tonal changes all with your playing style. I honestly believe that the main reason there are guys play amps and music style like we're discussing is because they've never experienced what it's like to control a great amp's tone and dynamics with your hands. Nothing is more satisfying, versitile, and complex. The other day Guthrie govan discussed that exact thing and tried to make the point as to how incredible it is to have it all at your fingertips. He was talking to those who haven't gotten that yet.

[R.G.]

It is nice to see the guitarist community at large waking up to some items that techies know semi-instinctively: if there's different tones to be had by clipping/compressing/combobulating here, there or everywhere, do all of them you like.

That's what switches are for.

[exclamationmark]

I'm gonna go against the grain here I would assume since the advent of the solid state rectifier and adequately sized filter caps, that sag affecting the pre-amp section doesn't really happen to a noticeable extent. There are generally quite large dropping resistors in the power supply, and the minimum load the tubes themselves have on the filter caps could just be taken as the plate resistor itself (which I were I think the OP's sag may be coming from - since the diode should only allow current to flow into the filter cap). All in all, there are huge time constants sitting in the power supply (especially in the pre-amp supply) of a lot of today's amps, which means unless something catastrophic happens, the note attack (which probably lasts <100 milliseconds) is unlikely to affect the pre-amp tubes.

I think it's much more likely that power supply cap directly after the rectifiers is getting hammered by the power tubes, and the PT simply cannot keep up. The main filter cap then has 2 choices of where to source it's replenishment current - from the filter caps downstream (the pre-amp filters), or from the overloaded transformer. The problem is, is that there is probably 10-50k of resistance between successive filter caps, which means the time constant is quite large - it cannot instantaneously sap current from these filters, hence the pre-amp tubes should be safe, unless the 'problem' persists for several seconds. The poor power tubes on the other hand have to essentially starve until the transformer recovers. That's my theory, anyway.

I would really like to test this out in practice, but I haven't got the time to do so. I figure all you'd have to do is scope one of the filter caps, stick a big resistive load on your amp, turn it to full, and play (or use your computer) through it.

[Steve Conner]

I've done that. The voltage on the downstream filter caps slowly sags over a period of several seconds, as you say.

[R.G.]

It gets complicated.

Even with solid state rectifiers, the first filter cap's average voltage sags. This is because although the rectifiers do pull it back up to nearly the peak voltage of the incoming AC at each half-cycle of AC power, once the diode turns off the cap discharges at a rate determined by the load. So between half-cycles, the ripple gets bigger by its lowest voltage sagging down, and the average of the DC voltage, which is what gets to the preamp circuits, also sags down.

This is less than happens with a big resistance in a tube rectifier; a tube rectifier's internal drops mean the voltage peaks when charging on each half cycle is lower. But it still happens, and the amount of sag on ripple is the same for the same first filter cap's value and loading. So first filter cap sag on loads is still there, just smaller with solid state rectification.

[deci belle]

That diode mod can stop high frequency oscillation in some instances (mine) when I put a 5f6a with an extra gain stage in a BF Showman. Without the diode the HF squealing was just in my hearing range, but I didn't hear it with the diode between the power and driver filter caps. Maybe if I had bigger main filter caps (mine are 80uf), the HF noise wouldn't be an issue.

Anyway, that's something I noticed right away that the diode mod is good for.

I couldn't tell too much difference in sag (in that amp) with the diode in or out (it has both tube and SS rectifiers), but it sounds more alive (a little looser) without the diode— so I pulled it after all. I just can't turn the gain pot up too high.

Maybe a tweed amp with less filtration would be more affected sag-wise by that hi-fi trick?

[Gingertube]

decibelle,
When you add an extra gain stage you invariably need an extra B+ decoupling stage for it. You can run 2 stages off one B+ node as long as they are opposite phase signals (which they will be for normal cascaded stages) but you must not try to run 2 stages with the same signal phase off a single node.
The next thing is that the PI has both phases and so it needs a B+ node all of its own.
The MOST common mistake in adding an extra preamp stage is running it from the PI B+ node and the amp invariable does what you describe, burst into oscillation above a certain gain or volume setting. How do I know this?, well I've made this screw up myself. I converted a Fenderish pre to a Trainwreck Express pre and just wired the extra stage to the PI B+ node, sounded great, right up till about 12 o'clock on the volume knob where it turned to custard. One more R and one more C to decouple the B+ for that new stage fixed the problem.

All,
That diode trick is I think now dated - it was perhaps appropriate for old low capacitance supplies with lots of power amp compression/sag to hold up the preamp supply. That would tend to preserve the attack in the preamp and focus the compression and sag into the power amp., I'm not a good enough player to be sure, but I think Chuck and Daz above are giving valuable advise in saying that is normally not what you want (thats what I like about this forum).
Getting just enough capacitance on the main power supply filter is quite tricky. These days I mainly do lower power amps with 6V6 pairs and 6V6 quads using mostly vintage 6V6G (ST shape) and usually 6SL7 octals for the preamps and PI. I buy a lot of 22uF/450V Panasonic ED and just add additional of these in parallel until it sounds about right, usually 2 off for 44uF on a pair of 6V6 and 3 off for 66uF (or occasionally 4 off for 88uF) for a quad, then more of these 22uF, one per B+ node down the power supply chain. I have also done this "pseudo-scientifically", run the power amp into a dummy load at just below clipping, look at the B+ with an oscilloscope (X100 probes and AC Input) and add capacitors to get get the 100Hz (120Hz for US) ripple peak to peak down to 5% off B+ value. I say "pseudo" because I'm not sure how that relates to just doing it by ear but I suspect that this "method" errs on the "TIGHT supply" side.

Keep in mind I'm not building uber gain shreaders but rather rock and blues musical instruments for grey haired (if any) old farts like myself.
It is hobby work for me and all "word of mouth" as in "can you build me an amp like that one you did for so and so".
I am also finding that as I learn more the builds tend to get simpler. The quad 6V6, with full Power Scaling, adjustable SAG, Fixed / Cathode bias switched, pentode/ triode mode switched power amp has given way nearly always fixed bias 6V6 in pentode mode (for 15 Watts) or cathode biased EL34 in pentode mode (for 32 Watts) if they want a bit more power.

Cheers,
Ian

[deci belle]

hi Gingertube~
Thank you for that note and also for describing your routine for getting the filter caps dialed in the power section too!!

I have the normal ab763 pwr node topology until after the normal channel's 22uf cap (D). That node feeds an 8uf cap (E) for the CF triodes and its normal gain triode then another 8uf cap (F) feeds the extra gain stage triode (four total triodes added).

Chuck intuited that maybe some sort of rectification was being accomplished by that diode. I'm glad I tried it cuz I know it isn't that great for my amp's tone— even if it arrested oscillation generated by the extra gain stage. As you have pointed out~ who'd ever need it on a Showman!

The good news is that after putting a 4.7pf 1kv ceramic cap between the plate and grid pins of the first (extra) gain stage triode that gets the guitar's signal from the jack, (last pwr node), the parasitic oscillation was eliminated. Yay!!

When I was first learning about adding stages on this amp (only two years ago) I'd initially fed the 5f6 preamp tubes from the screen's filter cap (through that convenient little hole that used to go to the Vibrato tube) for some hefty motorboatin' haha!! I've learned a lot from this old Showman so far.

[soundguruman]

I have stacked as many as 42 diodes in series to drop the voltage. Pretty cool.
They used to do that in old radio transmitters.

[J M Fahey]

I have stacked as many as 42 diodes in series to drop the voltage. Pretty cool.
They used to do that in old radio transmitters.

Mmmmmhhhh, interesting.
So the "too high" original +B (say, 2000 V for an 813) is now 42x0.7V=30V less.
Or taking a smaller tube, an 807 800V DC supply now also is 30V lower.
Sure that 4% voltage variation will drastically change the operating point or anything else.

I bet the phase of the Moon has more influence than that.

Just putting things in perspective