Yet another Bassman pedal

[KarenColumbo]

Dear all, I recently jammed with a few friends, and the guitar player plugged his strat into an old Fender Bassman that was growing mold in a corner ( a real bottle tank),  and it sounded not bad at all. So I thought I'd build me a pedal that sounds as close to this as I could fake it.

This is what I came up with on breadboard:


As you can see it's a straight adaption from the "original" 5F schematic as I found in Aron's gallery and other places in the web. I stole the LED/diode trick at the forst two transistor bases from ROG's "Azabache" and the BAT85 thing from somewhere else, I can't remember, I think it was Dead Astronaut's take on the Bassman.

I like it, it reacts very lively to different Strat pickup combis and the volume pot. With my Seymour Duncan hotrail humbucker it gets a bit into Zen Drive territory. All in all I'm quite satisfied.

There's some issues, though (ain't there always?):
1. There's a lot of dirt coming through - I know, it's pretty high-gain, but there's a constant high note coming through (I guess it's the wallwart), quite a bit of single coil hum and some hiss. I can get rid of most of it with the "FB" (feedback) pot but then I kill the treble content completely. Do you see some insertion point for additional "debris filtering" in the schematic?
2. The tonestack isn't very effective. I took the values from the "Dr. Boogie" pedal which are roughly a tenth of what is in the original because the high values killed the life out of everything. Maybe it's a thing of impedance mismatch/missing decoupling? Can you point me in the right direction? Foremost I'd like a bit more bass content.

[blackieNYC]

I would suspect an oscillation and not the wall wart. Try another wart.  You may need to drop and few picofarads here and there. 

[Kipper4]

No input or output caps?

[Kipper4]

Sometimes a 1nf or less from D to +V helps with high freq noise.

[anotherjim]

Can get away without an input cap, but it should have an output cap. Before the output volume pot is a good place.

That NFB looks like trouble. Series control never goes to zero feedback. The signal path inside the loop contains several RC phase shifts so at several spot frequencies it's way off - possible cause of oscillation. Note the RC network present in real amp NFB returns to correct relative phase enough to keep stability.

Try...
Disconnect feedback pot. Does it sound better?

With the feedback connected and disconnected...
Disconnect C2. Does it clean up better than introducing feedback did?
Treble emphasis from C4 & C5 -  are either or both really necessary?

[KarenColumbo]

Thanks, gents, for the hints! I will try them in a few hours' time! What I already tried is removing the feedback pot - without it noise is at it's max. There's only a small range inside of which it works fine - haven't got to measure this - will probably put a 10k pot there with some series resitance to preserve this exact range. FB pot is a real asset to dial in the correct measure of clipping, too Would hate to see it go. Will try all the other things next:

- Input & Output caps
- Checking out C5, 7 and 8

Ah, and I put some 20 pF from R14/C9 junction to ground - bit of hiss and hum gone, oscillation quieter, but still there

[KarenColumbo]

Hm, just pulled up a schematic from the gallery: On this one the FB pot is connected to Q3 Gate & Drain

[KarenColumbo]

Okay - Strange things happen.

- C4 left out, FB pot in
When I omit C4 all treble content is gone (also the oscillation). I can compensate this with the feedback pot to dial in more treble (oscillation included).

- C4 left out, FB pot left out
I ripped out the fb pot and abandoned it - same result but with no chance to dial in this treble content. This thing need C4, it seems.

- C4 back in, no FB pot
I tried down to 330p with C4 - this leaves treble there but damps the oscillation a bit. At 220p the oscillation is almost gone but I notice dullness creeping back in.
Just for the heck of it I put in 15 nF - full noise AND pretty much 10 times the gain/distortion. As if I turned the gain pot at least 20 degrees or so. Strange.

- C5 ripped out -> no noticeable difference. Abandoned.

- Bypassed drain resistors on all 201s with 1nF (drain to V+) - no noticeable difference

Obviously this thing needs feedback to cancel out the noise it picks up on the way. But the point where I take off the signal simply isn't there, is it? Junction Q2/Q3 did not work. Tried Q3's source - it reversed the gain pot action And no clipping at all. Q4 gate and source just mute the signal.

By the way - the "Bass Boy" schematic I posted before has Feedback taken from the junction of Q2/Q3 to Q1's drain AND Q3's drain to right after the feedback pot. Tried that -> muted signal.

Edit: Ah - as a search in the forum brought up: If I put a simple opamp buffer pedal between this one and the guitar, the oscillation almost vanishes. Hm - should I just buffer the input? Maybe a simple BJT thing?

Edit 2: Just audioprobed it. Oscillation begins right at Q1's output. Even without feedback cap @ source. Hm.

I'll leave this on the board 'til Monday (it's Saturday 8.30 PM here) and hope for a solution. If all fails I will reboard it on another breadboard and switch out all of the components - maybe I got some faulty part on it?

[anotherjim]

I'm going crosseyed, I meant C8 not C2 in my post above. I don't think D3 will work as intended with a bypass cap. I think the idea is that rising signal will lead to rising source and D3 don't conduct unless the FET saturates and signal continues to rise and then D3 will limit the overdrive on the FET gate. With the source bypass C2 present and signal being AC, then D3 will conduct everytime the input goes positive enough and clip things prematurely. Could even be causing a motorboat noise action.
Of course, I meant seeing if removing C2 would show that It might be better to reduce gain in the first stage rather than adding NFB. If it did, the a control could be added to vary the effect of C2 and thus the gain.

[KarenColumbo]

Ah! Now I see how that's working! I will try this asap, thanks!

[KarenColumbo]

Still on it. Tried a few other schematics with the general "Tweed"-gist and siphoning tricks.

Very interesting discovery: Some (quite prominent) builder inserted the guitar signal both at gain stage 2 AND stage 1 to have feedback. Strange, but effective. Sucks up lots of gain, but I wanted this crunchy a max, not full-on distortion. Problem: sounds nothing like a '59 Bassman at all.

Other variations included Mu-Amp-like stages - but their gain is far off my map. Nice heavy, but transparent overdrive, though.

Another take on this included two long-tailed pairs in the curcuit to induce feedback (I think). But this would be very, very far off the original circuit, so I just gazed at it with suspicion

I even tried to keep everything as clean as possible (with 9V no easy task) and introduce clipping by feedback diodes in an op-amp. But that's cheating.

[ericlaw02]

Did you end up finishing the design on this circuit? Been looking for Bassman preamp pedals to build and stumbled on this post.

[KarenColumbo]

Actually yes, but I completely changed the method. Mu Amp was the ticket ... and it doesn't sound too much like the amp, but it works fine. I'm away from my archive, but I can post the schematic of the working build tomorrow.

[roseblood11]

Catalinbread used mu amps in the 5f6, too.
Pedalpcb.com sells a pcb ("Tweed Man"), and here's a veroboard layout:
http://guitar-fx-layouts.42897.x6.nabble.com/Catalinbread-Formula-5F6-td45068.html

[KarenColumbo]

Those guys really know how to lay out a PCB. Mine is almost twice the size and still cramped here and there.

[tubegeek]

Work in progress:

https://www.diystompboxes.com/smfforum/index.php?topic=106605.0

[Mark Hammer]

The HAO Sole Pressure pedal is reputedly a Bassman-in-a-box pedal.  I made one.  It's okay.  My own 59 Bassman is currently on the fritz, or else I'd do a comparison and tell you how close it comes.

[printer2]

One thing to note, the tones stack does barely anything until you turn the mid pot down to 5k and less, more like 2k. With the 10k you have a range of maybe 25% of the bottom travel. A log pot helps if 10k is desired to get the least attenuation.

[iainpunk]

maybe this gainstage might work well, i have liked it in every build i did with it, from soft overdrive to searing fuzz