Fender Pre and Bassman Volume questions

[Firestorm]

Just re-read your post: where is the 12AY7? You don't have it on two stages do you?

[M Fowler]

Are you sure a mixer or recovery stage is not need at this junction prior to the PI?

[Subjecttochange]

So you're saying that not using a bypass cap and raising the cathode resistor won't decrease the gain? Or did I misunderstand what you mean.

The 12AY7 is V3a&b. I can make either A or B a 12AX7 though if need be since I will have an extra triode (not including the mixing stage).

That's why I put "mixer" in quotes- because instead I'd be using the old mixing stage as an additional gain stage to the fender channel- and not part of any mixing.

I'd like to keep the NFB if I can.

I may try something as high as 470k. I am colorblind- so I have to check all components with a DVM before I put them in- but at least I don't get my 10's place mixed up! ...but they're definitely 220ks.


And no, M Fowler- I'm not sure that a mixer is not needed. I'm pretty convinced that it is needed- I just cant get it clean yet. It's adding unwanted clipping in the PI. I'll post what I have as of right now.

[Subjecttochange]

This is the current state- minus the fender input being grounded.

Also- found a decent schem of a recovery stage. Does this look like a good one?

http://www.prowessamplifiers.com/schema ... matic.html

[Firestorm]

Are you sure a mixer or recovery stage is not need at this junction prior to the PI?

Fender never used one in any non-reverb amp. There's nothing to mix, nor anything to recover. You could certainly add a stage here if the objective was obnoxious gain (like in a TW Express), but you'd have to set the gain to near zero to keep it tame.

[Firestorm]

This is the current state- minus the fender input being grounded.

Also- found a decent schem of a recovery stage. Does this look like a good one?

http://www.prowessamplifiers.com/schema ... matic.html

I don't think I'd copy anything from a '70s CBS amp. That stage is mostly there for the Bass Channel (compare the AA165 circuit where the guitar channel doesn't go through it). It's got the NFB loop connected right to the PI input with no load resistor, so there's pretty strong attenuation there. And there's local feedback around that third stage (the 220K from plate to grid).

In you latest schematic, you've got 2 stages of 12AY7, with the second one unbypassed. That will be wimpy as hell. I'd make it a 12AX7 and bypass the cathode resistor on V3B. That is your tone recovery stage (and you need it because the tonestack will drop 15-25dB depending on frequency). If you want a third stage for either channel (or both), that's where I'd put a 12AY7 or a 12AU7, except I suspect you have only one triode free. You could make the cathode follower stage in the Rocket a 12AY7 since there's no gain there anyway and use the remaining triode as the final preamp gain stage. Keep it unbypassed, with a fairly big cathode resistor (at least 10K) to keep the gain under control. Since that will reduce the voltage drop across the plate resistor, it increases the risk that the stage will clip with a big signal. You could take the signal off the cathode (another cathode follower), but that would have no gain. And with only one triode to work with, you'll still be mixing the two channels through a pair of 220K resistors, so what's the point, really? I would start out simple with just the 220K (or 470K) iso resistors, 12AX7 on the Fender channel, and see where that gets you.

[Firestorm]

Oh, I see you've drawn your inputs as shorting jacks. With nothing plugged in, those inputs are always grounded all by themselves.

[Subjecttochange]

Well I tried it anyway between my last post and yours. It actually works quite well on the rocket channel (which is now being mixed into the "recovery"... whatever). Since I have the fender input grounded- I am going to make it a live jack- so I can switch between them to test both.

By the way I love the "obnoxious gain" comment about the express- I still love that design... but so true.

I think I may switch out the 12AY7 for an 12AX7- don't really care about the tube types... just want it to sound good.

I'll let you know how this test goes. The soldering iron is warm again.

[Structo]

Well this is now really getting over my head.

Something that might help the Fender side would a local feedback loop as done on the Dumble on the second gain stage.
That can smooth things and take away some of the rawness.
In fact, I put a switch on mine so I can use it or not.
I find without it, the tone is a bit more raw.
But again, the ODS circuit is a cascaded preamp so it may not apply to this.

I find the interconnecting resistors and such between stages kind of mysterious. I guess I don't understand a lot about them although I know you can tame some high end by using a larger resistor.

Is that mainly used for impedance matching between stages?

I'm glad that a few other people have joined in the discussion because this is quickly turning into a learning thread for me.

[Subjecttochange]

Ok so...

Sounds good right now as in the schematic shown below. I did also replace V3 with a 12AX7. It's much louder due to the gain- but breaks up at about noon. Is there any way to fight against that?

No problems with any sorts of weird distortion coming from the PI, PPIMV or anywhere else. Which is cool.

I was just running through an A/B/Y switch switching between inputs and comparing the sounds. The channels actually sound quite similar- except the "Rocket" channel starts to break up sooner. Tonally though- they're close. Any suggestions to make them a little more distinct from each other?

The amp is much nearer to a place where I want it though- and close to a time to think about relays again.


*Structo* you're talking about the 2x22M and 0.05uF that sit in between the grid and the plate of the second triode? I dont have any resistors remotely that high... maybe a 5.1M or two left... I am curious to try it though.

[Firestorm]

To delay the breakup, try making V2 a 12AY7 or a 12AU7. If more reduction is needed, increase the size of the cathode resistor on V2B.

To change the sound of the channels, you can fool with the bypass caps (make one really small, for example) or modify the tone stack: change the slope resistor, increase the size of the mid resistor/pot, change to a Baxandall/James tone stack (or a Skyliner), etc.

[dave g]

The 500pf coupler into the PI sure seems small. As in it would block most all bass frequencies.

Not quite. The input impedance of a LTP PI is larger than that of a common cathode amplifier. Without any global negative feedback whatsoever, you're looking at an input impedance of at least 2 meg. So, with a 500pF input coupling cap, 1/(2*pi*2E6*500E-12) = 159.15 Hz. Add a hefty amount of NFB like the AA864 has, and the -3dB frequency drops even lower.

Contrast that to the cutoff of the first coupling network in the rocket with a 500pF coupler and a 500k volume pot: 1/(2*pi*500E3*500E-12) = 636.62 Hz.

Intuition is fine, as long as you verify it

FWIW, I like a PI input cap of .002 and power tube couplers of .02. They are as large as you would ever need them to be in a guitar amp, and they won't hold on to a lot of charge when the output tubes are pushed into grid conduction. Or, just use .02s everywhere and make life easy

[Subjecttochange]

Thanks dave- that threw me for a loop for a while. But I think I've gotten it all ironed out in my head and have a .002uF in there now.

I should have been more clear on what breakup I want to reduce... I'd like to keep the breakup on the rocket side of things- and keep the fender side as clean as possible.

I think I am going to try the Baxandall tone stack in the "fender" clean channel tomorrow. Though I guess it'd be losing it's fender DNA pretty fast. I also may reduce the first cathode bypass cap on the rocket channel to make it sound a little more middy/in your face.

[dave g]

Having two relatively clean preamp sections feels a little redundant to me. Why not lose the mixing stage and use that extra triode to rewire the Fender preamp as an Express preamp?

[Firestorm]

You could raise the plate resistor on V1B to 150K or even 220K for more breakup on that side. You should probably make the cathode resistor 2K with a 150K plate and 2K7 with the 220K plate to keep the bias in the right place. If you want to keep the Fender channel Fendery, put a Baxandall-like thing in the Rocket channel. You can set it for more mids way better than the FVM stack. It doesn't have to have pots in it. You could design it with just caps and resistors, sort of the way the fixed value Rocket stack is now.

To keep the Fender Channel clean, you could stick a voltage divider after V3. In fact, you could put a pot there and dial in the amount of signal you want.