Some ideas for the P90

[Ben Lyman]

I've been messing around with a Phase90 circuit on my bb and came up with this. I am using 2N5457's and 3x TL062 and so far everything is sounding great to me.
I wanted the bias trimmer to be a pot so I figured out a way to do that with a B50k between the extra resistors, 39k and 180k.
I also used a 15k and B100k series resistors to make a variable feedback adjustment.
I had to put a 3k on the Rate control because it will go way too fast and sound weird, then stall out. Maybe because I used a C500k pot? I still can't decide if I want to change that top speed or not.
Here's a vid: https://youtu.be/uq767wDtcaM
EDITED schematic, removed 1N914 and re-arranged bias resistors:


I am wondering about some of it though.

1) I am using my standard power filter/protection/LED plan, which is permanently installed in my bb, along with that I also threw in the 1N914 protection diode (D2) but it probably isn't necessary, right? I removed it and I didn't notice any changes, should I have noticed something?
 
2) Is it ok to use my U1 as a buffer and 1st phase stage? My U3 as last phase stage and LFO? Should I be worried about weird noises leaking through or something?

[robthequiet]

Some wise person said something like if it sounds good it is good. Sounds cool in the vid. The bias adjust is especially cool, which could be swept for an effect unto itself. Looks like you nailed the parameters in good ranges. Itsa keeper.

[deadastronaut]

sounds cool....a nice mod is to turn the lfo off, and have a manual sweep filter...

nice with OD/DIST in front...

[TejfolvonDanone]

2 things i'd like to add:
1) the 1N914 (D2) isn't necessary because the 1N5817 protects the circuit from reverse voltages and the Zener doesn't need any protection.
2) I like to put my LED after the reverse protection circuit so any reverse voltage won't damaging it.

It sounds great by the way.

[Gus]

Look at the section around R4 you have 39K in series with the Vref output and "top" of the bias

The other schematics I looked at on the web show the Vref connected to rest of the circuit without a series resistor

So you are not connecting the sources and emitter with a low resistance node
Q4 emitter current will wiggle the Vref and bias voltage

R3 the 10K sets the current in the zener shunt regulator, 9VDC - zener voltage / 10k for  3.9VDC / 10 for about .39ma

Have you tried moving the Vref to the other side of the 39K?

You might have found a different adjustment for the p90

Are you using a tant for C12?

[duck_arse]

can I arkse, what's the caper with Q1? how come it's connected to the Vref line?
sorry, half- and mis-read.

[Ben Lyman]

Thanks Gus, good catch.  My mistake in drawing the schematic... oops! it's not like that on my bb, I will fix my sloppy schematic later.
EDIT: I am using a tant for C12 15uF, is that the only kind that works right there?

Thanks everyone else too, what do you all think of the top speed? I think most people would say to let it go bonkers-fast at the top end like an alien ray gun battle. I tend to think just "musically fast" and no more than that.

Duck, I think Q1 is drawn correctly, as are all the others. It seems Vref and Vbias wiggle them all except the output mixer Q, which only gets + from Vref to e

Here's a link to the schematic I started with:
http://www.electrosmash.com/images/tech/phase-90/mxr-phase-90-script-logo-schematic-parts.png

[TejfolvonDanone]

Thanks everyone else too, what do you all think of the top speed?  I think most people would say to let it go bonkers-fast at the top end like an alien ray gun battle. I tend to think just "musically fast" and no more than that.

In my builds i tend to have a "headroom" for the controls. I mean when i build something i almost always scale the values so that i don't want to max out them (and neither go to the minimum in some cases). So i never end up with the "i just want a tiny bit more on this knob" situations.

[Ben Lyman]

At 2:10 in my vid I show top speed with the 3k stopper, I like this, it's really just slightly above any speed I would ever use.
At 4:30 I pull out the stop and show a sample of the possibility to include optional "un-musical" speeds 

[TejfolvonDanone]

At 4:30 I pull out the stop and show a sample of the possibility to include optional "un-musical" speeds 

I loved that part So i would go with a really small resistor for setting the minimum like 100R.
Being serious i would say that something between 3k and 1k would be the ideal for ME.

[induction]

2) I like to put my LED after the reverse protection circuit so any reverse voltage won't damaging it.

Reverse voltage doesn't damage leds. They're diodes. Insensitivity to reverse voltage is one of their defining characteristics.

[Gus]

LEDs have a reverse breakdown voltage so it is safer to move the led +power after the 1n5817 or you could place a Si diode across the LED to conduct if reverse voltage is applied

[TejfolvonDanone]

Reverse voltage doesn't damage leds. They're diodes. Insensitivity to reverse voltage is one of their defining characteristics.

Diodes have a reverse breakdown characteristics which essentially means that if you apply voltage big enough in the reverse direction high current can flow. Zeners usually work in this reverse breakdown area with set breakdown voltage and they are used as voltage reference and shunt regulators.
For most diodes there is a rated reverse voltage at which they are guaranteed not to break down. This value isn't always specified for LEDs and they are usually much lower than for a regular diodes. So LEDs aren't so forgiving to reverse voltage as diodes are. You may blow only the LED with a reverse voltage. Protecting the LED doesn't need a lot of effort: you just put it after the Schottky and you are done. It's better to be safe than sorry. You could put a antiparallel Si diode with the LED as Gus mentioned but why bother when you already have reverse voltage protection?

[induction]

Interesting. Learn something new every day. I've reversed dozens of leds without killing them, so I didn't realize it could be a problem, but the internet agrees with you. I stand corrected.

[Mark Hammer]

• R35 (3M9 as shown) sets the current feeding the FET gates, and ultimately the sweep width.  MXR opted for a one-size-fits-all-speeds sweep-width so it is fixed.  Faster speeds want less width, and slower ones want wider.  So, consider using 4M7 for R35, and a toggle to strap in one of two other fixed resistors in parallel to achieve other widths.  Try not to go any lower than 3M.
• The bias current coming through R7 (1M) sets the initial starting point of the sweep, and the LFO simply adds to that.  Varying the value of R7 adjusts the "offset" or range of the sweep, so that it can be gurgly to swirly.  Consider replacing it with a 680k fixed resistor in series with a 500k variable resistance.
• Higher feedback tends to results in cumulative noise.  So what many manufacturers do is include a cap in the feedback loop of one or more phase-shift stages.  So consider a .001uf-.0015uf cap in parallel with R22.

It becomes your choice as to whether things like feedback, width, mix, or offset/range should be set with pots or toggl;es (2 or 3 position).Prsonally, I find that 3 settings each for width and feedback are good enough for me, while a vibrato/phase switch is sufficient for mixing.  Offset, however, is really more of the sort of thing that wants a variable control, since it really changes the character of the phasing  a lot.

[PRR]

LEDs do reverse breakdown BUT they also "forward breakdown". So we "always" have a resistor in series to limit the forward current. In general, this will also limit the reverse current.

Some quibble. Forward breakdown is say 2V and reverse breakdown is something over 5V. If the forward current limit is "melt down", and the supply voltage is significantly higher than 5V, you can get to a point that heat in reverse breakdown (due to higher break voltage) causes melt-down.

I don't think this can happen with 9V or 12V supplies. So an LED plus its resistor can be safely and frequently "reversed" in car, PC, and pedal applications.

I did once "prove" it could be a problem for 110V AC supply. Though at that heroic level the resistor ratings get awkward first.

[Kipper4]

2) Is it ok to use my U1 as a buffer and 1st phase stage? My U3 as last phase stage and LFO? Should I be worried about weird noises leaking through or something?
« Last Edit: Yesterday at 06:16:54 AM by Ben Lyman »

Question 1
Buffer and first stage should be fine using a dual op amp.

Question 2
U3 as a dual oppy.
I would not.
Maybe use 2x single op amps.
UA741 TL022 see below
Having said that while you have it on the breadboard you could try it and see for yourself.

It's a good idea to use a low power op amp for the lfo.
Such as TL022.

Look at some other. Phaser schematic's for parts.
Also take a look see the bi phaser for those caps in the feedback loop.

http://experimentalistsanonymous.com/diy/Schematics/Phasers/Mu-Tron%20Bi-Phase%20Phase%20Stage.gif

Then take a look at the univibe cap values for some different ideas. I think I read some where they're chosen for octaves. Maybe I imagined that.

Congratulations on some good mods mate.
Some really good advice here from the guys.
Keep up the good work

[Ben Lyman]

Wow! Thanks guys, some great info here. I've been having a  lot of fun experimenting with all those ideas.
I have made some very slight changes but not too much, as I really like the sounds as it is but I got some parameters to go a little further out there.
Still not sure if I really want sonic ray gun noises but a 1k8 for R33 allows for crazy buzzy noise without stopping altogether, backing the knob down a tiny bit returns to more normal vibey speeds.
But then, as D'Astro says, stopping it is pretty cool when you can sweep the other knobs around and get manual settings. It seems I don't even need a switch if I just leave out R33.
My main problem now is with the layout, AARRGGHH!!! What a complicated circuit to do on a perf.
I hope I can figure out something because I really like the sounds and I really want to build this thing bad!

[Ben Lyman]

finished at last!

[Ben Lyman]

Gotta revive this for a minute, rather than start a new thread.
I love this pedal and use it constantly, however I want to simplify it a bit, maybe come up with something I can make for friends and family, maybe a smaller box.

I tried a PHASE 45 on the BB and I love it! I don't know why people bag on the P45, it sounds perfect.
I also tried one part of the JC Maillet Vibe Mod, just the 2x 50nF caps swapped out for a 10nF and 100nF.
I guess it's more vibier, I dunno, it was late, I'll check it again later.
Pulling the dry lead made a nice vibrato, not worth putting on a switch if you ask me.

Here's my question: Before I build this, is there any new developments I should be aware of? Any errors in the schematic?
Mark Hammer: I used your schematic (THANKS!) from Ampage: http://www.ampage.org/schematics/mxrphase45.pdf
I also tried to read the GGG schematic, it looks about the same but it was a little difficult for my rookie eyes to suss: http://www.generalguitargadgets.com/pdf/ggg_p45_sc.pdf
Here is JCM's: http://www.lynx.bc.ca/~jc/phase45modded.gif

I'm thinking of replacing the 1M (depth resistor?) with something bigger or a pot?

Other changes of mine include:
2x TL072
2x 2N5457 or 2N5458
Series 1n5817 polarity protection
Series 100r power filter and bypass caps 100uF & 100nF

I think thats it
Thanks!