79 UL Twin Cascade/modernizing

[GlideOn-Designs]

So my buddy and I are working on overhauling (already been re-capped, and retubed) his beat-to-hell 79' Fender UL Twin Reverb tomorrow and I'm gathered all the necessary parts and am proposing the following 'modernizing' changes:



- Normal Channel tonestack changes (Marshall JTM/Plexi spec)

- 'Brilliance' control on now-defunct ground switch: NFB adjust 420 ohm to 10k and High Cut up
to 22nf on Concentric 100k pot.

- Celestion G12H75 speakers swap.

- Optional: PPIVMV 250k LarMar instead of current Pre-phase to help make better use of high gain at lower practice volumes.



I'm fairly confident in my ability of the above mods having built a Champ
Style amp, retrofitted an old Gibson GA amp head into a Bassman style amp and turned my 1968 Traynor into a Marshall JMP/JCM800 amp.

I've done Cascade with the Traynor amp by putting the V1 Pin7 triode onto a DPDT switch which series to a 470k attenuator resistor and I have effectively switchable Cascase/Modern amp distortion tones.

I wanted to do the same "switchable cascade' for the AB763/Twin circuit, but wasn't completely sure how as the V1 and V2 is very different to the 5F6A circuit I'm familiar with. I understand that there isn't as much inherent gain on the preamp channels as there is in more typical Marshall style circuits, hence the desire to Cascade/Series.

But based on a few findings and my own experience, I came up with this diagram to have switchable Cascade (and necessary attenuation) via a push/pull Mid pot. Nothing too fancy or extensive and easily reversible I hope.



This should work, right?

 

[GlideOn-Designs]

Well I followed an Uncle Doug YouTube page that gave crystal clear instructions on how to remove the pre-pi master volume and subsequent trashy reverb driver boost with it.

However, the inverse slanted faceplate makes it impossible to fit a dual-pot LarMar style Post-PI Master, so it'll have to be a crossline MV maybe with 820ohm resistors in series to remove the common 'scratchy' artifacts down low and common with this MV type.

 

[Wally]

When I revoice that Normal channel, I separate the pin 8 cathodes on V1&V2 so that gain can be manipulated. I maintain the 820 ohm cathode resistor in the revoiced channel and install A 1.5kohm resistor in the other channel to maintain the classic AB763 type of channel. I like to apply reverb to both channels as well. This will add gain to that modified Normal channel, too.

 

[GlideOn-Designs]

When I revoice that Normal channel, I separate the pin 8 cathodes on V1&V2 so that gain can be manipulated. I maintain the 820 ohm cathode resistor in the revoiced channel and install A 1.5kohm resistor in the other channel to maintain the classic AB763 type of channel. I like to apply reverb to both channels as well. This will add gain to that modified Normal channel, too.

There is no shared cathode in the Silverface circuit. It's never been connected to my knowledge.

There is in the Bassman style though, which I've done before to give two distinct channels/volumes. 2.7/0.68uf on one and 1.5k/25uf on the other.

I've already give the voicing changes, it's surprisingly gainy an Marshall-y.

I can't believe this isn't more popular or as documented to do. Haven't even messed with the phase invertor or mixing resistors

 

[Jowes_84]

Wally has it and tons of experience. So listen to him.
Just to add to be complete on „cascading“ those MV Push Pull boosts:
You should already have a boost circuit on your Twin and a nice SPDT 1Meg push pull pot - the master volume.
Most people don’t like the boost in stock form, me neither.
You could potentially use that switch.
But..
It is possible to adjust that stock MV boost circuit slightly and mimic what Rivera and Jahns did on the smaller 80‘s amps (Super Champ, Princeton Reverb - see schematics online)
Some people also do not like that sort of lead sound, but it should be possible to tweak to taste easily. You will need to add a 10-39k plate resistor to the reverb driver and adjust from there..
Just an idea, you would use what you already have and not change the amp by much.
The wires leading to the MV pot switch should be kept away from reverb and trem, maybe shielded. Most of these amps suffer Tremolo ticking so that might benefit your amp anyways.
I’ve done it for my 74 100W Vibrosonic Reverb (non UL) - sounds are usable - but to be honest - I do not use it often.
Disclaimer: I did not review your schematic/layout carefully enough.
On PPIMV for UL, a Feedback adjust and a presence control-not sure, you now have three controls each of them messing with your amps feedback on an amp designed for ultra clean, I would maybe only pick one of these mods, listen, and then go from there.
(I have done too many mods on amps in the past and do have regrets )

 

[GlideOn-Designs]

Appreciate it, I just didn't see what he was talking about at first but yes the cathode is technically shared after the tube from V1 to V2 pins 8 instead of across the socket like most amps.

I had to look at it for awhile to see what was going on, I apologize. My experience thus far is primarily 5F6A circuits. I have almost religiously studied the Rob Robinette pages and apply my experience with this currently.



Correct me if im wrong, but I think this is what I'm understanding:



- V1 and V2 pins 8 (cathode) seperated.

- V2 (vibrato channel) gets 1.5k to reset center bias for that stage.

- V1 gets new 820ohm...to ground? Bypassed with a capacitor (like the V2 .68 cathode boost on early JMP amps?)


Cascade is actually on a replaced Normal Volume 1Ma pot with push/pull, kicks in a 1M series resistor, a bit hot but works well.

Any tips on where to mount the 820ohm cathode? There's a nice spot almost directly across the pin to the Bright Switch ground...

Oh and the old push/pull MV is out of the amp now, so is the 12k/.1uf/1M resistor network halfway through the stack. It sounded like garbage and the pot has a horrible taper, wobbly too.

I already installed a Crossline MV with 1ks jumpered in series, sounds much better and is far more versatile overall. But there's no room on the front slant panel for a PEC dual pot for LarMar, the downwards angle prevents that...may end up going on the rear where the Ground Switch is, swap for a presence control or variable NFB up front.

Meanwhile I'd prefer to boost like with pedals or some cathode fat boost if possible.

 

[andrewRneumann]

Cascade is actually on a replaced Normal Volume 1Ma pot with push/pull, kicks in a 1M series resistor, a bit hot but works well.

Did you use a SPST or a DPDT? Your original layout showed a DPDT with a jumper and 1MΩ resistor. It didn't look to me like the switch would do anything--always sending signals down both channels. You got it to work?

 

[GlideOn-Designs]

Maybe my diagram isn't correct (I was a graphics major, not an enginnering major) but yes, it works and is quite funny much of a Jekyll and Hyde effect it is. Technically I used a Single Throw switch, a typical DPDT was just a few hairs too big to fit. I got the idea from the stock Master Volume which is also a Single and barely fits.

However even with 1M series' worth of signal attenuated it is still very hot and treble-y.

The tone controls also heavily influence the sound so you have to crank the mids on both sides which honestly sounds best anyhow on normal settings.

I'll likely experiemt with the tonestack a bit more - maybe lower to 4.7uf - 2uf for the vibrato to nip away blocking distortion at higher settings, snubber on V1 plate to cathode, maybe bump normal channel mixer to 470k/500pf. PI coupling cap could be lower than the 0.02 as well. It's starting to become something other than a Twin but that's okay, there was room for improvement anyhow to compliment the Celestion G12H75 speakers. Still massive transformers, high headroom and high horsepower is always universally appreciated even if it is a UL.

Cascade is very fun and useful option to have and make it stealth with no downsides or parasidic losses present when done correctly.

 

[andrewRneumann]

Maybe my diagram isn't correct (I was a graphics major, not an enginnering major) but yes, it works and is quite funny much of a Jekyll and Hyde effect it is. Technically I used a Single Throw switch, a typical DPDT was just a few hairs too big to fit. I got the idea from the stock Master Volume which is also a Single and barely fits.

However even with 1M series' worth of signal attenuated it is still very hot and treble-y.

The tone controls also heavily influence the sound so you have to crank the mids on both sides which honestly sounds best anyhow on normal settings.

I'll likely experiemt with the tonestack a bit more - maybe lower to 4.7uf - 2uf for the vibrato to nip away blocking distortion at higher settings, snubber on V1 plate to cathode, maybe bump normal channel mixer to 470k/500pf. PI coupling cap could be lower than the 0.02 as well. It's starting to become something other than a Twin but that's okay, there was room for improvement anyhow to compliment the Celestion G12H75 speakers. Still massive transformers, high headroom and high horsepower is always universally appreciated even if it is a UL.

Cascade is very fun and useful option to have and make it stealth with no downsides or parasidic losses present when done correctly.

I'm being dense I guess. I'm trying to figure out how you accomplished this with only a SPST. So when the switch is "open" the signal just goes to V6 pin 2 like the original does. When you close it, it sends the signal through a 1MΩ and to V2 pin 2. When it's closed, it is sending signal to both places--both operating in parallel? I would think you would want to cut off the signal going to V6 when you were cascading to V2.

BTW, if I am looking at the photo correctly, that 1M mounted on the back of the pot switch looks like a 100K.

 

[GlideOn-Designs]

Good eye and you are correct - the SPST when pulled bridges a series resistor when engaged, blacks any connection/normal operation when pushed in. Simple and elegant (I think).

Actually took the pic when trying a 270k (same I have in V1 plate), now it's a 1M as I had though I needed in the first place. Less k means hotter signal, but in this case the Silverface needs lots of it to be usable. The volume level is nearly dead-on same as either Normal or Vibrato channels at level.

I suspect there is some parallel signal bleed as there is still a fair amount of sound if you happen to plug into the vibrato channel, but I'm hoping it'll some be negligable or phase out more with higher k mixers and give it more midrange grunt in the process, help it stand out for solos or leads. Again, still very useable, dramatically different sound here and no problem doing thrash metal or Hair Metal with a Les Paul, the enormous NFB probably helping it be as tight as it seems to be.

Will have to experient some more in my Den later

 

[GlideOn-Designs]

So the 470k/500pf mixer for normal does indeed help with the Cascade and attenuate some bleeding of course, but I'm not sure about the tone. It really alters the character putting it over the fence into Marshall territory. Perhaps 270k or 330k is worth trying?

I also think reducing all of the V1 and V2 cathode to ground caps below 5uf will help with the blocking distortion and instability on higher settings without sacrificing those rich syrupy lows. And maybe a higher 25k mid pot for the Vibrato side, make the normal a 100k while we're at it.

But if I'm being honest, it's really the tonestack that messes with the Cascade. Any extreme adjustments really neuter or muddy the tone. The other solution is to lift it altogether, but that's extreme and invasive, I still want the Twin sound ao let's try higher Mid pots first!

I want a 'splash' of British authority but not a complete voicing change.

 

[GlideOn-Designs]

Hmm, discovering that all of the knobs are designed to meant to fit nylon shaft pots, solid shaft are no-go save for the master volume (weirdly the only solid shat pot with a screw-on knob). Tried a knurled CTS and although it clumsily fits it is still not a 1:1 like the Nylon.

That means using new knobs or using Nylon aftermarket which only have 10k, 250k, 500k and 1M. No 50k or 100k for a mid pot.

 

[GlideOn-Designs]

Update:

Fender makes 50ka knurled shaft, they fit with some plumbing tape

 

[Wally]

Update:

Fender makes 50ka knurled shaft, they fit with some plumbing tape

There are bushing shims to fit over split shafts to accommodate set screw knobs.

 

[GlideOn-Designs]

 

[GlideOn-Designs]

Before/After

Filter caps replaced with F&Ts, dropping/bleeding resistors with 2w Metal Oxide.



Tonestack Changes:


Normal Channel:

39k slope resistor

270k plate resistor (2w metal)

500pf treble peaker

.022uf for tone shaping caps

.0047uf coupling

2.7k/.68uf 1st cathode

820ohm/10uf 2nd cathode (split from V2, thanks to @Wally for suggesting)



Vibrato channel:

1uf (stock) and .022 Blackface tone shaping caps

1.5k/4.7uf 1st cathode

1.5k/8uf 2nd cathode (split)


250pf over V1 mixer resistor

PI coupling .0015uf

 

[GlideOn-Designs]

Replaced the crumbly carbon screen and grid resistors on the power tubes with 2w metal oxide and 3w ceramics. Cleaned up some oxidation around the grounds.

Freed up the ground switch too. Not a lot of room to use a control, definitely not a stacked LarMar. So maybe a combined NFB/cut control.

These are not friendly amps for stacked or sensitive controls.

I'm stuck with a Crossline in the front for now, but I would love to find some slimmer stacked 100k precision pot for a LarMar.

Edit:
Found a slimmer pot! These are slim like the Alpha pots but in 2w rating

(F #54) JH Dual 100K Audio Potentiometer 100K Ohm Taper https://a.co/d/77WOiHx