EPFM phaser

[BlueFuzz]

Hi there,

I've just finished breadboarding by first pedal - the Phaser form Craig Anderton's book. Oddly, though, all I get is an effect where the volume kind of modulates slightly rather than sounding like a phaser as such. Even then, it's not very blatant and is only noticable with the resonance and intensity cranked up full. Sounds nothing like the demo CD, that's for sure. I've tried the suggestion in the book to check that the optoisolators' resistance is varying with a multimeter when the effect is on, but it looks fine. I'm using NSL32's instead of the CLM6000's, and switched the 4739 ICs to 5532s - I got the pin mapping for this from some link or other on this forum.

Any suggestions? Is anyone familiar with this one enough to know which components I might try varying, or does it just sound like there's some obvious flaw in my wiring, in spite of me checking it a dozen times...?

All ideas gratefully received...

[petemoore]

  All probably needs checked.
 They're probably typical phase stages like in the SS or the 90, Ross.
 Could be offbase with no schematic on this though.
 CA Phaser I would think would sound phase in any case, see if the phase stages are similar to a phase but with a vibe type LDR controlling the sweep.
  See the Debugging sticky.
 Audio probe and voltage measurements should tell you more, if not, post the results.

[BlueFuzz]

Thanks - i'll keep checking it over and report back.

By the way, the pin mapping I'm using on the relacement op amps is as follows:

4739 --> 5532
1            1
2            N/A
3            N/A
4            N/A
5            3
6            2
7            4
8            6
9            5
10          N/A
11          N/A
12          N/A
13          7
14          8

I should have posted that on the first one...

[Mark Hammer]

The location of the notches is a function of the LDR resistance and capacitor value.  If the combination of the two, using the Silonex units is too high up, or too low down, then you won't really hear the notches in action.  That will be because either: a) you don't have any guitar signal to filter in that range, b) your speakers can't reproduce those frequencies, or both.

The off resistance of the LDRs in the Silonex units (500k) seem to be the same as those in the Clairex units, but may produce a different resistance for any given degree of illumination of the LED.  The location of the resulting notch is a function of the joint resistance of the LDR and parallel resistor, plus the cap.  So, if the LDR is completely off (500k), the LDR+470k resistor and .01uf cap produces 90 degrees of phase shift at 66hz.....kinda low.  When the LDR is completely "on", it produces 90 degrees of phase shift at approximately 32khz...kinda high.  For an audible effect, you likely want the location of the notches to vary between around 150hz up to maybe 3khz or so.

It may be that you need to either: a) change the values of the 470k resistors in parallel with the LDRs to achieve a suitable resistance change, or b) change the value of the caps from .01uf to something higher or lower to have the range shift over appropriately (larger for lower notches, smaller for higher ones).  You will note, for example, that the Phase 90 uses .05 (.047uf) caps and a 22k fixed resistor in parallel with the JFET, to get the notch in a meaningful range to produce the desired "gargle".

[daverdave]

This might be a dumb response but make sure you're using a dual supply, +15V and -15V. The circuit won't run off a 9V battery.

[daverdave]

Mark, just out of curiosity, I've read that allpass filters produce a phase shift of up to 180 degrees, does this depend on the frequency of the signal, as i've read in most stuff on phaser pedals that it's 90 degrees.

[BlueFuzz]

Guys,

Thanks so much for the suggestions. Mark -  I will tinker with those values and see what I get. Re: the power supply, I'm running it off two 9V batteries to get a bipolar supply as per the spec - is this likely to be too low still?

I've been trying to debug, and got to the point where I started wondering if the op amps were damaged, so out they came for a checkup. I hooked up a little test circuit that I found here: http://tangentsoft.net/elec/opamp-tester.html except I changed the pin layout to account for the different arrangement of pins on the 4136 Quad op amp, and I didn't have the rail splitter so replaced that part with my two 9v batteries so I still had a virtual ground. I tried this on the 4136 and found something odd - it lit the LED on all channels, but it didn't matter which of the outputs was hooked up to the LED! I.e. when I was testing the channel B op amp section I inadvertently left the output of channel A connected to the LED. I only noticed when I went to change to test the next op amp section. I then tried various combinations and it seems as though all the op amp outs triggered the circuit regardless of which op amp was in use. I'm not an expert by any means, but does this mean my chip is technically fried?!

Thanks,
Jon

[Mark Hammer]

Mark, just out of curiosity, I've read that allpass filters produce a phase shift of up to 180 degrees, does this depend on the frequency of the signal, as i've read in most stuff on phaser pedals that it's 90 degrees.

Each stage can produce up to a maximum of 90 degrees of phase shift above some corner frequency, and decreasing amounts of phase shift on the other side of that "inflection point".  That's why 2 stages are required to produce a notch: 180 degrees is needed to have a completely anti-phase signal.

Allpass stages are roughly grouped into lead and lag types.  The one most of us are familiar with (cap in series, resistance to ground) produces increasing amounts of phase shift at higher frequencies.  The other configuration has the resistor in series and the cap going to ground.  That arrangement produces decreasing amounts of phase as frequency increases.  I always forget which is lead and which is lag.

Like any type of filter, allpass filters can be single-pole or multi-pole, with 2-pole and higher having steeper slopes (i.e., amount of phase shift drops off quickly beyond the corner frequency, rather than gradually).

[jimbeaux]

Projects in both the older & later EPFM books are made to run off of a 9 volt bipolar power supply.

- the opto-isolator projects can be modified to run off of +-15 vdc.

fresh 9 volt batteries should be fine.

[BlueFuzz]

I did some more work on the quad op-amp IC, since it behaved oddly in the test circuit. I measured the resistance between the outputs of the 4 stages, with nothing else connected, as about 9.5k

Does anyone know if this is normal or indicates a damaged chip? I wondered if it might be what's muddying the phasing effect when it's in the circuit?

Jon

[BlueFuzz]

Hi, just thought I'd add an update.

I did try a replacement 4136, but it has the same behaviour, so it looks like I didn't fry the last one after all - it still just sounded like some odd delay/rotary effect, with a juddering sound but no actual swelling effect. However, when I tried the suggestion of changing the values of the capacitor next to the optocoupler, and the resistor in parallel to it, that was a different story - I progressed up to 0.047uF and down to 2k respectively (from 0.01uF and 470k), and it's now a massively different sound, and one I feel justified in calling phasing! I would even go as far as to say I like it, so it can't be bad!

I will keep tweaking, but thanks for the tips - it's helped me to get there in the end.

Jon

[Nasse]

Usually, in a phaser, output from phase sift stages are mixed, some way, with non phased signal. Perhaps you have problems with mixing stage