over the top gain, help me tame the monster!

[centervolume]

I have completed a home brew build and the gain is completely off the charts. The design is a single end output basically imported from gibsonette GA-8 (but with 5881s and toggle for parallel single end operation).

The front end is a fender 5F8-A but with a cathode follower driving the tone controls. The preamp tubes are very low voltage (12AU7 and 12AT7 under 50 volts on the plates each). The 2nd triode of V1 is direct coupled to triode 1 of V2 (the 2nd triode of V2 is a post tone stack recovery stage prior to outputs).

I've been dialing in the tonestack section and I'm happy with the EQ balance and controls, but turning the volume past 2 is to summons the gods of hot lava, just scorching for the wattage, guitar feedback starting up, the whole 9 yards. If I could get what happens between 0 and 2 on the volume control to take place instead between 0 and 6 or 7 I'd be there.

The volume pot is 1MEG, should I drop to 500K? What about putting resistor/capacitor as coupling rather than the direct coupling setup as per the cathode follower?

Any help appreciated!

[92Volts]

Is the volume pot an audio/logarithmic taper pot? These give a slower increase for the first 1/2 of the turn (and less volume by the halfway mark) compared to linear taper pots. Ears perceive volume relatively... if 1 to 2 is a doubling of volume, that's the same perceived change as the doubling from 5 to 10. Audio taper pots account for this.

[shooter]

under 50 volts on the plates each

a working schematic would be great.
What are the 5881's biased at

[kagliostro]

A split load resistor on V2 recovery stage or a voltage divider or a voltage divider before that stage ....

BUT

a working schematic would be great

is not easy to imagine the whole circuit details

Franco

[centervolume]

This is a recent schematic, a couple updates since then but they are mostly in the tone control

[shooter]

What voltage you have at the cathode of the 5881's?
try sticking an AU in V1's slot for testing

EDIT: also try gatorcliping a 47ish uF cap across the cathode R for the 5881's

[centervolume]

ok - I put in an audio taper 1 Meg volume control and I increased the resistor off the vol pot wiper to 330K. I also put in a 5Y3 rectifier (I think I'm within tolerance on the 5Y3 - voltages below). This seems to be much better. I also put in a 12AU7 in V1 and a 12AT7 in V2 (due to sound and perceived greater robustness regarding the direct coupling, I'm probably off on this since the DC voltage to grid is at issue not overall plate voltage handling).

Here are the voltages
B+
1: 396
2: 387
3: 155
4: 117

V1
p1: 29
g1: 0
k1: .71

p2: 29
g2: 0
k2: .76

V2
p1: 54
g1: 29
k1: 30

p2: 55
g2: 0
k2: .8

V3: 36ma
p1: 390
g1: 387
g2: 0
k1: 33

V4: 35ma
p1: 390
g1: 387
g1: 0
k1: 33

[shooter]

V1
p1: 29
g1: 0
k1: .71

You are probably starving the tubes some, I would take the 75K dropping R to 22K

I think your problem is more about distortion than gain, with the low volts on the pre tubes it's probably distorting long before you get to overdriven

[VMS]

You can reduce some gain by removing V2b cathode cap C7

[centervolume]

I think your problem is more about distortion than gain, with the low volts on the pre tubes it's probably distorting long before you get to overdriven

I can try that, although it should be noted that it is at least both because with a thinline hollowbody es335 type guitar like I have been using to test the amp, I was getting feedback from the guitar itself at very low volume levels, and the amp was surprisingly loud. Now I have that under a little bit more control where I can hear the cold voltage distortion you are referring to. That was initially part of the design (ironically) as an anti-feedback approach given that the amp is designed for use with harmonica (hand held mic) and low voltage preamps have shown a good approach. Passing that type of low voltage distortion through the gain stages was as per design. Now I am a little confused and feel I have stepped into an alternate dimension to have screaming gain for guitar with low volts and 12au7 etc.

[centervolume]

You can reduce some gain by removing V2b cathode cap C7

Thanks VMS - will put that on my list to try out as well

[92Volts]

You do have a lot of gain after your volume control... the cathode follower driving the tonestack limits the amount of signal lost there. You have yet another gain stage driving the power tubes where a phase inverter would be, but that stage has more gain than a PI would (with the same tube, at least). If you're looking for preamp distortion before (or without) power tube distortion, why not add a master volume?

[shooter]

the amp was surprisingly loud

good loud?
I have a very similar amp, PSE 6V6, "close" to your pre-amp, (I have 1 more gain stage)
V1 has 100VDC and 80VDC
my PS taps are 340, 300, 280, 270
those might be helpful as "bench-marks"

Guessing you don't have a scope?, you might be able to jack in music, set your vol/gain knob to 7ish, adjust your "music in" volume til it just starts distorting.  How "loud" did it get?, did it sound pretty good?

[centervolume]

Shooter,
I do have a scope. That's a good idea to put it on a scope and see where things are getting too crazy. It sounds really good between 1-3 or so on the volume. Above 4 or so and it's just too much (I dont need 6 notches of out of control distortion and power on the volume control, I want 1-6 on the clean side and then 7-10 as overdrive).

I messed around a little this afternoon with the dropping resistors as was mentioned above, and generally getting V1 (which is coupled to 1/st half of V2) at more common plate voltages but it just sounded bad. I'll see if I can find my dB meter to see how much volume its really putting out. I may try a 500 or 250K Ohm pot for volume instead of 1Meg

[PRR]

> This is a recent schematic

OK.

For reference, a classic Fender will make full output with 20mV on the input. Taking a rough-number "20V" needed to slam a power tube, we need gain of 1,000 from input to power tube.

Rough-counting on thumbs, I see gain of about 10,000 in your plan. WAY too much.

It seems to be inspired by 5F6a. But the 5F6a has a mixer (2:1 loss) after the volume control. 5F6a does not cathode-bypass the stage before the tonestack driver. 5F6a uses a longtail final driver which is half the gain of a single-tube driver. 5F6a takes just-significant NFB around the power stage. And 5F6a works the final in fixed AB mode which needs more slam than SE class A.

So your plan has easily 10X more gain than 5F6a. Yes, super high gain is a thing. For solid-body head-banger players in large rooms. Hollow 335 at home, it could be dangerous above "2". As you report.

Also the 5F6a does NOT have a 75K B+ dropper. This is something to experiment with; but let's get a "good stable" amp first, then experiment with laming various stages for more "color".

[centervolume]

Hi PRR
This is consistent with what I am experiencing. I suppose 5F6a is the basic reference for the front end even though I was looking at the 5F8a twin, it's pretty likely they are nearly identical until it comes to the power/output section. For this design I wanted to experiment with parallel single ended possibilities, using a toggle switch etc. Thanks to your message I can see the 5 points where gain is going over the top (no mixer, cathode bypass before tonestack driver, longtail PI driver, no NFB, AB mode). Since I am running class A operation, 2 of the 5 items are not areas where I can make adjustments.

I'm really happy with everything but the gain issue. Meaning that the tone and texture I am getting at vol settings of about 1-3 are really nice and they disappear when I take steps to get the V1, 2 plates in "normal" operating range (i.e. above 100VDC). I'd be thrilled to find a way for the dB range I am currently getting from 1-4 on the volume pot to basically become the full range available on that pot, so I may experiment with 250K as a pot value rather than the current 1M. I guess I'm looking to reduce overall gain by 75% (or preserving 1/4 of what I have).

I did reduce value of the first dropping resistor in power supply from 76KOhms to 47K, and my plates are now around 50VDC. I like the idea of running those cool and I am interested in finding ways to attenuate.

I'll apply your suggestions from the mark up schematic and report back. Thanks again!

[shooter]

I guess I'm looking to reduce overall gain by 75%

since you have a scope tweaking is easier;
Your "bias" is 30ishVDC
Monitor with scope the Plate of V2B, make your adjustments in the pre for a "target" of about 40ish Vac-rms clean-ish, reasonable looking sine wave, with your gain pot at 10.  That should give you good clean tone to around 7ish, then let you spank the 5881's some.  Once you push SE without NFB hard she gets ugly, but when you find that sweet spot, you'll smile

[centervolume]

I'm interested in nfb in terms of where I would make the connections.

That said I have followed the recommendation from PRR (took the first dropping resistor down to 12K, then eliminated the 2 cathode bypass caps and inserted the extra resistors from those grids to ground - what term would be used for those added resistors?) and the results are really spot on. Fantastic! breakup starts about 7 as I had hoped.

My tone controls no longer seem effective but perhaps I disrupted a connection or do I need new values?

Another question, if I wanted to restore one of the cathode bypass caps for extra gain, I could remove the new resistor and put the cap back and reclaim some of the gain?

[shooter]

followed the recommendation from PRR

He's helped me tune-up many a build.

fix the tone knob, play long enough to know which way you wanna go, then tweak, till your next build