Planning a new vintage EH Polyphase project

[Morocotopo]

I want to share with you people a new project I´m planning to make, after I finish some other things. It´s the vintage EH Polyphase.

Some background: I made, a looong time ago, this pedal using the Topopiccione project. Back then I was just starting in this FX business/hobby/obsession, so my knowledge of electronics was very small and this was my first complex project. It worked, it´s one of the nicest phasers I´ve heard. But, the envelope part never worked, what´s more, it emmited intermitently a very very loud high pitched whine that was really annoying. I tried to make it work but never succeeded, and one day, in a fit of anger after being assaulted once more by that horrible whine, I tore it apart, and that was the end of that.

Recently I found in my computer a sound file of the thing, and listening to it made me want to have this great pedal again, but fully working. I´m confident that with the (bit more of) knowledge I have now I´ll be able to make it work. Or at least I´ll give it a try...

Of course, with knowledge comes ambition... so I won´t just recreate the one I made before, I want MORE!!!

What I want is this:

- First, breadboard it and make it work.
- Replace all odd value/type parts to more common available ones. That includes the CLM 8200/2. I originally found those, and still have them salvaged from the first build, but in the interest of making this a project that anyone can make, it would be nice to make it work with more common optos. Of course that would require SIX single optos if no suitable double optos can be found... will be expensive...
I´ll also try to replace the 3140 and replace/eliminate the goddamed thermistor.
- The topopiccione project has some mods in the schem, some values replaced, I believe filtering added and some other things that I´ll have to investigate and eventually add to the design if they improve the pedal, wich judging by my first build they do mostly.
- I will design a new PCB, to try to make it as small as possible, I think the Topopiccione one was a bit too big, partly because it has all the pots lined at the top edge of the board, and that means more tracks and more space needed.
- I´ll also try to make it run on 9V with a charge pump or voltage doubler, so as to make it more pedalboard friendly.

Well, that´s my plan of action. The first step I took was to redraw/number the schem:





In the next days I´ll comment about the oddities/mistakes I see in the original schem, of wich there are a few, I believe.

Something I remember from my first build:
It didn´t phase with common 4558´s, I had to track down some BA4558´s, wich I believe are very old ones, to make it phase. Odd.

Well, that´s the general idea. surely I´ll ask for help here, and any comments/suggestions/etc are of course welcome. This is a great pedal, and it would be nice to have a project for an "easy polyphase".

P.D: Of course this project is possible because of the work of Paul Nelson and others I´m not aware of that made available the first incarnation of the pedal. To them, thanks!

[Mark Hammer]

Couldn't come at a better time!  I'm staring at one right now that I'm trying to get working again for a buddy.

[armdnrdy]

Hey Morocotopo,

My hat (if I were wearing one) goes off to you for being persistent! I don't like these silly little collections of electronic components to get the best of me either!

If I may make a few suggestions: Check out the VTL5C3/2 dual vactrol. I just compared the specs and they are very close. I think it's a winner.

Also, did you try different op amps beside the 4558 and BA4558? There's a post on here describing how a MXR 117 clone wouldn't flange with certain op amps and would with others. Try different models instead of trying to source the BA4558s.

[Govmnt_Lacky]

I'm a big fan of using LM358s as LFO op amps.

Check those too 

[armdnrdy]

Yeah, 358's and TL022's which Tayda is now stocking!

[Morocotopo]

Couldn't come at a better time!  I'm staring at one right now that I'm trying to get working again for a buddy.

Mark, glad the schem is useful to you.
You have a real one with you? Would it be too much to ask for some (or lots of!) photos of the thing to try to trace it and compare with the schem? Pretty pleaaaaase??



Armdnrdy, I´ll look into the VTL5C3/2, but I was kind of thinking to try the NSL32, haven´t really looked at the datasheet or anything, just that it seems to be the cheapest one, and easily available. But we´ll see. Of course I´ll try different opamps once I have the thing on the breadboard.

Govmnt, the 1458 is a common opamp, so to begin with I wasn´t thinking of replacing that one as the LFO opamp. But it´s an option to be considered.

[armdnrdy]

Morocotopo,

I learned on this site that many LFO designs utilize the LM358 and TLO22 because they offer very low power consumption. What that means in a flanger or phaser LFO is when the LFO is cycling up and down the power consumption varies which can put a drag on other parts of the circuit. It can also cause the infamous "LFO ticking". If a build calls for a dual op amp for the LFO, (most do) I always use one of the above just to be safe.   

As far as using the NSL32, it will work, your build will phase, but it won't be a Polyphase!

The decay time is too slow. The NSL32 is 500ms to 100K compared to the CLM8200/2 at 30ms to 100K. The VTL5C3/2 decay is 50ms to 100K. The reason why I suggested the VTL5C3/2 is because the overall specs are close enough to achieve a similar effect.

But then again, if you're going to breadboard the circuit, and don't mind experimenting, try the NSLs and compare them to the CLMs that you have. Let your own ears be the judge. Maybe you'll stumble upon a good sounding accident!

[Morocotopo]

There goes the NSL32... bye bye.



Thanks for the info Armdnrdy. I was also thinking about home made optos, but I have the suspicion I won´t find fast enough LDR´s. I´ll have to order some VTL5C3/2´s from the Bear, I guess. Or, or, go to the store were I bought the 8200/2´s a few years ago, to see if they still stock it.

[armdnrdy]

I wouldn't suggest "roll your own" vactrols for this build. I experimented with making my own but two issues arose. First, as you stated it's very difficult to find LDRs with a fast enough rise time. Secondly, they should be roughly matched, the closer the better.

Email Tayda and request that they bring in a line of Vactrols. I'm positive that their price will be more competitive. I've requested that they stock several items. I've outlined why they are essential to the DIY community and Tayda's brought them in.

[Morocotopo]

Just bought 24 CLM8200/2´s at a store, they were the last ones they had in stock. U$S 5,80 each!!!!!



So I have enough to make 8 Polyphases. Maybe make them and sell them at an exorbitant price, with the original parts mojo and all...

[Mark Hammer]

Finally got the Polyphase on my bench working. Problem was a fried 4558 and a fried 4013.  All is good now.

Interesting effect.  Nice phase sound, smooth envelope following, flexible control.  Capable of producing traditional phase-shifter sounds, and some different ones.  Probably a good candidate for true-bypassing.  I haven't attempted to use it in "split" mode yet (irect to one amp, phase-shifted to another).

The unit employs two entirely different LFOs, so you can have two different speeds for the normally-swept and envelope-swept settings.  The regular sweep has just the rate and feedback/regeneration control, with a pleasing range of feedback settings and sweep speeds.  Can produce a nice gargley sound.

The envelope function combines an LFO and pick-attack sensitivity.  Turn the modulation and rate down and the unit can simply sweep the phase shift higher as you pick harder.  I suppose from my experience with the Line 6 Tap Tremolo and M5, I was mistakenly expecting that the LFO rate would correspond to picking intensity, but that's not how it works.  The LFO rate is set via the speed control and does not change in response to pick attack.  What changes is how much the pick attack causes the LFO to be offset.  So, you get the same sweep, except it moves higher up in the audio range; a bit like yodelling with your mouth slightly open and gradually opening your mouth as you continue yodelling.

The few hand-drawn "factory" schems that are posted around conflict with each other.  Not so much because they reflect different issues of the pedal, but rather because they are both incomplete in their own way, whether it doesn't show the switching properly, or omits the entirety of the power supply.  I'm also not entirely certain the envelope control section is drawn properly.  The way some things are connected doesn't seem right.  On the other hand, it is drawn in such a convoluted way, it can get hard to recognize parts of the circuit in comparison to familiar mental templates for doing this and that.  So, it may well be more accurate than I think, and I just don't recognize it.

I look forward to Morocotopo's final "improved" drawing.  It's a complicated effect, with a lot of parts, but it does a good job at what it does.  Strikes me as territory for interesting mods, but they wouldn't be all that easy, and I certainly wouldnn't recommend doing them on a vintage unit and compromising its value.

[pinkjimiphoton]

here's an assload of gut shots.
i gotta get mine working too, so maybe together we can crack this hen's tooth and make it buildable!

http://www.diystompboxes.com/smfforum/index.php?topic=100570.new#new