PCB build, Block Logo Phase 90. Creativity and experience requested!

[GreenTeaZ]

Ok, so I bought a vintage PCB for a block logo Phase 90, the kind with 3 dual op-amps. It's late 70s from what I know. Anyways, if you guys would be so kind, do any of you guys have any suggestions for various mods to get the most versatility packed into this little phaser? And names too, any hilarious, off the wall names are also appreciated. Thank you!

[plexi12000]

i was looking too before. cant remember what pages i read, now....but i just googled mxr phase 90 mods....or something close to that.  lots of stuff out there.

heck....prolly lots of stuff on the forum too

[Mark Hammer]

A really good phase shifter will have 5 controls: speed, depth, resonance/feedback, manual/initial/offset, and mix.

Speed, you already have.

Modulation depth is to be found in that 3M9 (or 3M3) resistor coming off the LFO.  The larger the value, the narrower the sweep.  Personally, I don't feel the need to have more than a 2 or 3 position toggle for this.  You can replace the resistor with something a bit higher than 3M9, or just stick with 3M9, and use the toggle to place other resistors in parallel to drop the effective resistance and allow the sweep to be wider.  For example, 3M9, on its own gives a modest sweep width.  Placing 22M in parallel, drops that down to 3M3, which is a wider sweep, and placing 10M in parallel drops it down to 2M8, for an ultra-wide sweep.

The single 22k or 24k resistor that folks make a fuss about, when it comes to block-vs-script, can easily be made into a continuously variable control.  Higher resonance is nice for slow and wide sweeps, which negligible resonance is better for gentle bubbly sounds.  Replacing the 22k/24k with a 15k fixed resistor in series with a 25k-50k pot will get you a nice range of intense to negligible resonance.  A word of caution about this, however.  You'll find that the unit can become noisy with high resonance.  So I'd recommend putting a 2200pf cap in parallel with the 10k feedback resistor on the last phase-shift stage.

Manual/Initial/Offset is addressed via the resistor/resistance coming off the trimpot.  The trimpot sets the bias voltage and that resistor sets the current.  Once you have the bias set, with a 1M in place, you can again use a toggle to put some other fixed resistors in parallel with the 1M to bump the range of where the notches occur up a bit.   Alternatively, you can simply replace the 1M with a 680k resistor in series with a 500k variable resistance, to move the range around from gurgly to swirly to swishy.

In this particular instance, I don't think a continuously variable Mix control makes a big heap of difference.  You can be content simply installing a toggle to lift the connection to the dry signal at the mixing point, in order to get a vibrato effect.  You will find that the vibrato effect benefits from being able to adjust the width of the modulation.

Finally - and this is a bigger step - I've had a lot of fun with adding envelope control.  I tapped the signal from the input buffer stage, and ran it to a simple single op-amp rectifier, similar to what you see in a Doctor Q.  The rectifier drives an LED, and the LED moves the value of an LDR in parallel with the Speed control, such that I can make the modulation faster if I pick harder.

There you go.  A full-featured phaser from a humble P90.

[GreenTeaZ]

A really good phase shifter will have 5 controls: speed, depth, resonance/feedback, manual/initial/offset, and mix.

Speed, you already have.

Modulation depth is to be found in that 3M9 (or 3M3) resistor coming off the LFO.  The larger the value, the narrower the sweep.  Personally, I don't feel the need to have more than a 2 or 3 position toggle for this.  You can replace the resistor with something a bit higher than 3M9, or just stick with 3M9, and use the toggle to place other resistors in parallel to drop the effective resistance and allow the sweep to be wider.  For example, 3M9, on its own gives a modest sweep width.  Placing 22M in parallel, drops that down to 3M3, which is a wider sweep, and placing 10M in parallel drops it down to 2M8, for an ultra-wide sweep.

The single 22k or 24k resistor that folks make a fuss about, when it comes to block-vs-script, can easily be made into a continuously variable control.  Higher resonance is nice for slow and wide sweeps, which negligible resonance is better for gentle bubbly sounds.  Replacing the 22k/24k with a 15k fixed resistor in series with a 25k-50k pot will get you a nice range of intense to negligible resonance.  A word of caution about this, however.  You'll find that the unit can become noisy with high resonance.  So I'd recommend putting a 2200pf cap in parallel with the 10k feedback resistor on the last phase-shift stage.

Manual/Initial/Offset is addressed via the resistor/resistance coming off the trimpot.  The trimpot sets the bias voltage and that resistor sets the current.  Once you have the bias set, with a 1M in place, you can again use a toggle to put some other fixed resistors in parallel with the 1M to bump the range of where the notches occur up a bit.   Alternatively, you can simply replace the 1M with a 680k resistor in series with a 500k variable resistance, to move the range around from gurgly to swirly to swishy.

In this particular instance, I don't think a continuously variable Mix control makes a big heap of difference.  You can be content simply installing a toggle to lift the connection to the dry signal at the mixing point, in order to get a vibrato effect.  You will find that the vibrato effect benefits from being able to adjust the width of the modulation.

Finally - and this is a bigger step - I've had a lot of fun with adding envelope control.  I tapped the signal from the input buffer stage, and ran it to a simple single op-amp rectifier, similar to what you see in a Doctor Q.  The rectifier drives an LED, and the LED moves the value of an LDR in parallel with the Speed control, such that I can make the modulation faster if I pick harder.

There you go.  A full-featured phaser from a humble P90.

Ok, so if I wanted to do all this, I'd need, in addition to the original parts,
an ON-ON-ON toggle
22M & 10M resistors
15K resistor
50k pot, what taper?
2200pf cap
10k feedback resistor
680k resistor
500k pot, what taper?
ON-OFF toggle to lift the connection
And I don't know if doing that last one is in my experience.
But WOW. I was indeed thinking of making the trimpot a normal pot and letting it fly.

EDIT: And what size enclosure would be smart so I can stuff all this in there?

[duck_arse]

the GGG circuit shows "2N4125 (2N3906)", so I'd guess any PNP general purpose. what have you got?

[Mark Hammer]

Ok, so if I wanted to do all this, I'd need, in addition to the original parts,
an ON-ON-ON toggle
22M & 10M resistors
15K resistor
50k pot, what taper?
2200pf cap
10k feedback resistor
680k resistor
500k pot, what taper?
ON-OFF toggle to lift the connection
And I don't know if doing that last one is in my experience.
But WOW. I was indeed thinking of making the trimpot a normal pot and letting it fly.

EDIT: And what size enclosure would be smart so I can stuff all this in there?

The 3-way toggle is a humble SPDT ON-OFF-ON, which is thankfully much cheaper and mroe available than the ON-ON-ON type (which are typically DPDT).  The short-handle ones that Small Bear sells are nice, and either similar or identical to the ones you'll find on ZVEX pedals.

The trimmer actually CAN be a panel-mounted 250K pot, except that only about 30% of its rotation will be useful.  The remaining 70% will move the JFETs out of their optimum bias range, such that you don't get any phasing at all.  Playing with the value of the 1M resistance is a much better solution.  A simple hi/med/lo range may work just fine for most people.

The 50k pot can be any taper but linear will probably work just fine.  The 500k pot can also be linear.

The added feedback cap IS an addition, but the 10k it goes in parallel with is already there on the board.  It is the cap that joins the input and output pins on each phase-shift stage.

I suppose your building style may be different, but I see absolutely no difficulty in fitting all of this into a normal 1590BB or one of the Small Bear Bare Box units.  You might....might be able to fit it into a 1590B if you a) omit space for a battery (sidestepping the need for a stereo input jack that takes up more space), b) use an "outie" power jack rather than an "innie" (i.e., the nut attaches from the outside with most of the jack outside the chassis), c) use a slim profile stompswitch and a strategically-located 3mm status LED, d) use the small-handle toggles that can be placed more closely together, and e) use a small pot with a slender knob.

I'm presently working on a dual P90 in a 1590BB.  I really like the idea of the MXP Phase 99, that packs a pair of P90s in one small enclosure, with the ability to run them in parallel or series, ganged to one LFO or separately swept.  I'm including that, but I also included hi/lo feedback, wide/narrow sweep, and hi/lo range, for each P90, using those sweet little PCB-mount push buttons, similar to what MXR uses for the P99.  It's remarkable how much you can pack into a box if you're willing to go with binary choices implemented using small pushbuttons.

[plexi12000]

awesome info--- thanks Mark Hammer.   I believe you nailed it. hahaha!

[GreenTeaZ]

So I have the PCB all soldered, WHEW that was a lot of work! But now, I've hit yet another roadblock. I see all the wire pads, and I know two are for power, two are for in/out, and two are for the potentiometer. I'm looking at the schematic and my head is spinning. Can someone help me figure out which wire goes to which deal?

EDIT: Got that sorted out relatively easily. Though, I'm very confused on your dry signal lift mod Mark. Where would I wire the switch to?