Rhodesplyr 6/13/2018 7:43 PM
Plush P1000S
I've been working on a 1971 Plush P1000S guitar head, and I thought I'd post some of what I've found. For those who don't know, it's pretty much a copy of the AB763 Twin Reverb with improved transformers.

Rumors of sketchy build quality are true. After doing the general repairs like recapping electrolytics and replacing drifted resistors, I had to go through the amp carefully to find the time bombs: joints that were never soldered properly at the factory and wires that were nicked when stripped and have weakened over time to the point where they're ready to break off.

The available schematic shows a tube cathode follower line out, but this one has a later transistor emitter follower (single BJT) line out that uses the bias supply as its power supply, so it runs at a negative voltage relative to the chassis. The downside of this approach is that because the bias supply isn't nearly as well-filtered as the B+ supply and because solid-state circuits don't have the input/output isolation of tube circuits, it feeds a lot of PS ripple back into the phase inverter, raising output hum dramatically. I'm going to recommend disabling it to the amp's owner, unless he really needs the line out, in which case I'd recommend retrofitting the tube cathode follower. It looks better on paper. At least it looks less likely to couple PS ripple back into the source. I guess the SS line out buffer saved a few bucks.

Another quirk of this amp is that it pulls the current for the preamp stages from the chassis through the input jacks grounds. I can't remember seeing that before.
 
Enzo 6/13/2018 8:24 PM
Why not just improve the bias supply filtration?
 
Rhodesplyr 6/13/2018 10:43 PM
Quote Originally Posted by Enzo View Post
Why not just improve the bias supply filtration?
They made a gesture towards this in the original build by adding extra capacitance to the bias supply, but it's simply not as effective as a choke and a row of filters. It has 100uF + 100uf separated by a resistor, and that's not enough ripple reduction for the circuit.

Also, on a SS-rectified amp, I like the bias supply to come up quickly.
 
olddawg 6/14/2018 12:03 AM
Quote Originally Posted by Rhodesplyr View Post
They made a gesture towards this in the original build by adding extra capacitance to the bias supply, but it's simply not as effective as a choke and a row of filters. It has 100uF + 100uf separated by a resistor, and that's not enough ripple reduction for the circuit.

Also, on a SS-rectified amp, I like the bias supply to come up quickly.
Can you isolate it and simply use a wall wort for a power supply if the customer needs a clean line out?
 
Enzo 6/14/2018 12:05 AM
Or run a branch from the bias over to a regulator. You will lose a couple volts but it will be clean.
 
J M Fahey 6/14/2018 4:22 AM
Of course.

For a more precise answer:

1) what is the bias source measured voltage ?
I guess some -50/60V DC.

2) What is the PNP transistor collector voltage?

3) please describe the RC string feeding it ... such as -55V raw bias - 1k - 10uF - 1k - 10uF - transistor collector voltage.

The point being that simply changing a couple R values, increasing (later) uF and adding a Zener diode will give you VERY clean DC with minimal alterations.

Oh, and please state transistor emitter resistor value.

Also, on a SS-rectified amp, I like the bias supply to come up quickly.
Oh, you wonīt touch the main bias filter cap, only downstream feeding the transistor and thatīs separated by resistors and clamped by a Zener.

EDIT: if possible , draw and post line out emitter follower circuit, which is very simple.
It can tolerate and reject quite a lot of ripple on collector IF base bias is clean, but I am suspecting a cheesy kludgy design which didnīt care about that.

Now why would I suspect that on a Plush amp?

Just making Fake Kustom amps with cloned Fender **tube** guts speaks volumes
 
Rhodesplyr 6/14/2018 7:53 AM
I greatly appreciate the offers of help.

I've already got a lot of hours in this repair making the amp reliable for use, which included tearing out a number of repairs even more poorly done than the original build, so I'm waiting for the owner to tell me whether the line out is important to him or not. Regulation had already crossed my mind, but that is more work.
 
J M Fahey 6/14/2018 8:07 AM
Oh 2 resistors, 1 cap and 1 Zener do not count as "work"

Specially compared to your idea of adding an extra *tube*

Punching a socket hole on a steel chassis, (plus 2 holes for mounting screws or rivets) , getting and wiring Filament power, , +V, grounding, a couple extra components, 400V class capacitor ... now *that* is work
 
Rhodesplyr 6/14/2018 9:40 AM
FWIW, the chassis is aluminum, and the tube socket holes are already there, empty.