More Vintage Gear: Roland AP-2 Phase Shifter

[Chuck D. Bones]

Another pedal from my brother's garage. I vaguely recall my roomie Richard owning this back in the mid 70's. I modded it for him by adding a DEPTH knob. He later sold it to me, I loaned it to my brother and forgot about it. Until a couple of weeks ago when he brought it back. Back in the day, nobody had a pedal board because there weren't that many pedals to be had. Big pedals were a thing. This baby is a boat anchor.

When I powered it up, no sound. I checked the LFO and it wasn't running either. Turns out that IC1 (the one on the right) had died. Not sure why, maybe someone hooked up the wrong voltage to it. There is no reverse voltage protection on this board. I pulled the LM2902 (same as an LM324). installed a socket and plugged in a new one. Voila! It works. As you can see from the schematic, it's a 6-stage JFET phase shifter. There's a 2-transistor buffer up front. 1/2 of IC1 and all of IC2 make up the 6 phase-shift stages. The remaining half of IC1 forms the familiar triangle wave LFO. It runs on a pair of 9V batteries. Not true bypass. The stomp switch simply disables the phase shifter by turning all of the JFETs fully on. The sketch at the top of the schematic shows how to implement more-or-less true bypass. The A25K DEPTH pot replaces R53.

 

[jubal81]

What a find.
That's a pretty good circuit. Bootstrap for input impedance. Feedback on the JFET gates to tame distortion.
It's a Rolls Royce compared to the Maestro.

 

[anevilspar7an]

This would be a pretty sweet project after updating what ICs and JFETs to use.

 

[Chuck D. Bones]

TL074s would work great. Lotta different JFETs will work. Basically any JFET that would work in a Phase 90. Could use a charge pump to make the -9V rail, or run it on just +9V and make Vref, like a Phase 90.

 

[cooder]

Classy! How do you like the sound? What would you compare it too?

 

[neiltheseal]

Any suggestions to improve the input buffer?

I’ve got o e of these and the phasing is fantastic but that buffer is dogshit. True bypass is a must but even still I’m trying to think oh how to improve it.

 

[Chuck D. Bones]

It feels like some treble is getting lost somewhere and it might be the buffer. I'm away from home for a few days, but when I get back I'll look into it.

 

[Chuck D. Bones]

Classy! How do you like the sound? What would you compare it too?

It's like one and a half phase 90s, I guess. The two additional phase shift stages definitely increases the intensity of the phasing effect. If you're used to the tone of the Phase 90 or UniVibe, then the AP-2 might seem like a bit much.

 

[neiltheseal]

It feels like some treble is getting lost somewhere and it might be the buffer. I'm away from home for a few days, but when I get back I'll look into it.

What are your thoughts on these preamp changes? I asked for tips to improve this in a facebook group.

"I'll suggest the following changes to the AP-2 preamp/mixer:
Move the top of R2 from Q1 collector to Q2 collector.
Increase value of R2 to 100k
Increase value of C3 to 100uF.
This changes the first stages to a more standard bootstrapped CE-stage without touching the phase and gain of the mixer stage."

 

[Chuck D. Bones]

One of us will need to run an LTSpice simulation and breadboard it.

 

[neiltheseal]

That's true. I will breadboard this one of these days when my brain is less full.

I converted this to be powered by 9v (with a charge pump) and made it true bypass. These are huge improvements but it needs more work.

I suppose if I do the buffer only it wouldn't be that big a job on the breadboard...