Phase 90 mods

[HunkFunkPedals]

What do you think the switches are? on/off/on? on/on/on? on/off?

[RickL]

The vibe mod switch will be on-off-on. It switches either the 15n (on) no caps (off) or 47n (on) caps in parallel with the existing caps for three possible combinations. Alternately it could be just a dpdt with no centre-off position, giving just two combinations, but, since the switch is the same size, there is no reason to do it that way.

The phase 45 mod switch could be either (have to see the schematic) but it is probably just a dpdt with no centre position.

A Vibrato switch usually removes the dry signal from the chain, but this switch appears to be acting as a spst, adding R18 in parallel with the existing resistor to give a smaller value. Unless there is a cut trace on the board that is not shown, in which case it will work the normal way, with the switch adding R18 in series with the existing resistor in one position, and opening the connection in the other position.

The Script mode switch does the same thing with R15, with the same possibility using a cut trace.

[HunkFunkPedals]

Does this make it clearer? The forum is in german, lol.

[RickL]

That definitely helps.

According to the schematic, S4, the Vibe mod, uses a dpdt switch to choose between 47n (stock) and 15n (mod) caps for C2 and C3. No third position. This will make the phaser sound more like a Univibe. You could actually use a centre-off switch for three variations by replacing the 47n caps on the board with 15n caps (instead of removing them, see below) and using 33n caps (to get 48n, pretty close to 47n) and some other value, maybe 15n to get 30n or keep the 47n for 62n (this is the one I would experiment with). This would give you choices of 48n (stock), 15n (univibe) and something between (30n or whatever you decide on).

S2, the phase 45 mod, bypasses two of the phasing stages to turn it into 2 stage phaser like a phase 45. It is definitely a regular dpdt switch, no third position. The two points marked "A" on the schematic are the wire that connects the two top lugs on S2.

S1, the script mod, switches R15 in and out of the circuit. R15 provides some feedback, which makes the phasing a little more intense. You could use a spdt centre-off switch for S1 and have a second choice for R15. Lower value gives more feedback, higher value gives less. Use a pot in this position, wired as a variable resistor, if you want to get fancy and make the second choice variable.

S3, the vibrato mod, does remove the dry signal, by disconnecting R18. As with S1, if you use a spdt centre-off switch you can experiment with the value of the second resister. A lower value will mix more of the dry signal in, making the phasing less intense and a higher value will mix in less, making it sound more like vibrato. Again, you can get fancy and use a pot to make this position variable.

For the script and vibrato mods to work you will have to remove R15 and R18, respectively, from the board. For the vibe mod to work you will have to remove C2 and C3 from the board (or replace them with 15n caps if you want to use a centre-off switch for three choices). For the phase 45 mod to work you will have to remove R9, R16 and C4 from the board, but note that the script mod will not have any effect when using the phase 45 mod.

[POTL]

for the univibe mod, it would be more correct to change 4 capacitor values. A 4pdt switch would be huge, but there are switches in dip packages, for example the cd4053 has 3pdt, I'm sure you can find 2 2pdt switches in dip packages or 4pdt in one package. about the 45/90 circuit,
I'm not sure that in the circuit from the German forum, phase 45 with feedback will sound normal, you can move the feedback closer to the input buffer (covering all 4 filters) as done by BOSS ph-1r and use the spdt switch to switch modes 45/90 I'm not good at hard switches, but if you have the right configuration, you can get by with an extra switch for modes 1) 45 without feedback
2) 90 without feedback
3) 90 with feedback
4) vibrato

in theory, you can use cd4066 / 4016 and a jack switch that controls it. using cmos for univibe mode can make 5 mode for this effect.
but I'd rather add a mix knob like in the Walrus modulation effects between 100% dry and 100% wet signals.
this will give the univibe mode both the chorus and vibrato positions of the original device. if you change the LFO to BOSS ph1 / 1r, you can add depth control, add center point adjustment as in flanger (manual) effects, this is in ada phaser.
Feedback adjustment with potentiometer.
replacing jfet with ldr+led

tc electronic has an interesting idea for switching the number of stages with a simple spdt on-off-on control they switch between 2-4-6 phasing stages, 45-90-135 degrees. but I still don't understand why their pedals say 4-8-13 stages.

hey is it still phase 90?

[HunkFunkPedals]

LoL....here's the kicker....I'm modifying a Joyo clone P90.....

I have the schematic for the joyo and the P90 and they are pretty much the same circuit. I was just trying to read the German about what the mods where doing to see if I was into them.

I'm new to phasers. Just trying to ruin one to see if inlike ruining them, ha!

[Mark Hammer]

Missing from the list of mods is adjustment of sweep width.  Because of the way human perception works, we prefer that when something changes very slowly, we want that change to be more obvious, and when it changes very quickly we prefer that the change be subtler.  So, three-knob phasers will have speed, variable feedback, and variable sweep width (normally or often labelled as "depth"), such that when one cranks the speed, the width of the sweep can be tamed.

The Phase 90 is a compromise design, with sweep width, sweep starting point, and amount of feedback all "preset" such that it works for any speed one selects.  Not too much sweep for fast speeds, and not too little for slow speeds.  But there can be wider sweeps and narrower ones that work well.

The resistance that dictates the width of the sweep in a Phase 90 is the 3M9 fixed resistor coming off the LFO (R27 in the posted drawing).  This limits the current feeding the gates of the four FETs.  It sums with the current coming off the trimpot.  If one reduces that resistance, more current is fed to the FETs and the sweep is wider.  Make the resistance larger, and the current is reduced, yielding a narrower sweep.  I find that a usable range goes from just under 3M to around 4M7.  A 3-position toggle provides suitable sweep adjustment.  Make R27 a 4M7 resistor, instead of 3M9, and use the side positions of the toggle to place either a 22M or a 10M resistor in parallel for effective values of just under 3M9 and just under 3M2.  If I'm not mistaken, some issues of the P90 used a 3M3 value for a wider sweep., so 3M2 will do nicely.  If you want to go nuts, use 8M2 instead of the 10M parallel resistor, for an effective resistance of 3M.

I'll note in passing that the "Color" switch on the EHX Small Stone adjusts the width and speed range of the sweep, as well as the amount of feedback, to achieve slower speeds that are wider with more feedback, and faster speeds that are narrower with less feedback.

[POTL]

LoL....here's the kicker....I'm modifying a Joyo clone P90.....

I have the schematic for the joyo and the P90 and they are pretty much the same circuit. I was just trying to read the German about what the mods where doing to see if I was into them.

I'm new to phasers. Just trying to ruin one to see if inlike ruining them, ha!

this should have started) IMHO the design of the Phase 90 is too raw and unfinished, I prefer the more thoughtful BOSS solutions to make the jfet work more efficiently. however, in your case, basic mods will do. 90/45 script/block

[Rob Strand]

Missing from the list of mods is adjustment of sweep width. 

Something else missing is just changing the caps on an MXR Phase 45/90 doesn't make it sound like a Univibe.   The LFO shape and depth (ie. the sweep) has a lot to do with it.   Changing the caps alone just weakens the phasing.

[CheapPedalCollector]

I much prefer having an active feedback control like the ROSS phaser and a mix control as well as width and manual controls. I can do without the mix control and would be fine with a 4 knob version of the Phase 90 similar to the Flanger. If going 6 knob I'd add mix and a 3 way switch for 2/4/6 stages to get all the tones from the 45, 90 and 100.

[POTL]

since we are talking about the univibe mod. I know there are easy vibe projects and a commercial product based on it by EQD Depts. A couple of years ago I was simulating replacing Phase 90 capacitors with univibe capacitors. As far as I remember, multisim did not show me similar notch values, the filtering was different. Even if we turn a blind eye to the LFO (we can try to program the microcontroller to the desired shape) and replacing the lamp with transistors / LDR. What is the probability of setting the correct filters? is it real at all? Or do I need to copy the discrete circuit?