Univibing a Ropez / Ross Phaser

[Mark Hammer]

One of my son's friends asked if I could build him a Univibe.  Rather than dealing with LDRs, and largely because Peter Snow was generous enough to give me a fistful of LM13600's, I figured I'd try building a Ross Phaser (again) with Francisco Pena's Tonepad layout, and doing the usual capacitor swap.

I can happily report that the Ross/Ropez delivers nice milky swirl with the standard cap change from all 3300pf caps to the .015, .22, 470pf, and .0047uf combo found in Univibes.

A few recommendations....

1) Though the Ross resonance control gives a nice robust edge to the sound of phase shifting, providing more flexibility than a Small Stone, in Univibe mode the resonance control does absolutely jack squat, so you can safely take it off, plus the save yourself the cost of the 1uf caps before and after it.  *Why* does it do jack squat?  Because equal value caps produce specific notches and peaks, which resonance emphasizes.  The discrepant cap values of the UNivibe spread phase shift around so there are not real obvious troughs and peaks to emphasize.

2) Sometimes you like a lot of sweep, and sometimes you don't.  The 10k current limiting resistor from pin 9 of the 13600 used as LFO sets the device for maximum (or close to it without contravening the input current limitations of the chip) sweep.  If you stick a 100k linear pot in series with the 10k resistor, you can limit the amount of sweep so that the effect is subtler.  Helpful for the phase shifter mode too.

3) Univibes don't really take much advantage of extra slow sweeps.  I also found that the stock LFO values for the Ross didn't cover off a useful range for the Univibe mode.  I changed the 3.3uf cap in the LFO (between pins 10/12 and V+) to a 1uf cap (now that I had two left over), and changed the speed pot from 500k to 100k linear.  This allocates a broader portion of the pot's rotation to small variations in faster speed so you can dial in something that sweeps with the rhythm of the song a little easier.

4) You can do it with the phase shifter as well, but anything that purports to be a Univibe should have a vibrato-mode switch.  In this case, you simply lift the 27k resistor between pin 1 and 6 of the dual op-amp to cancel the dry signal.  That resistor can be found adjacent to a 470k resistor between pins 4 and 5 of the chip.

5)  If you want to get deluxe, replace the 15k feedback resistor in the output stage of the op-amp to a 22k or even 33k for more output.  You can then stick a 100k volume control on the output instead of the 150k fixed resistor.

Is it the "ideal" Univibe?  I 'spose not, but it is dead cheap and dead easy and sounds nice, and it fits in a small box.

[Joep]

Hi Mark,

Great article!

I was about to build a Ross Phaser, so this comes in handy!

Bye,

Joep

[Fp-www.Tonepad.com]

I'm going to try this real soon!

Thanks Mark!

Fp

[Aharon]

Thanks MARK HAMMER.
This is the kind of practical advice and no nonsense DIY hands on approach that make this forum amazing.
We should have more people like you sir.
Thank you sincerelly
Aharon

[Chico]

I just built the Ropez phasor pedal and, based upon various suggestions, most notably from Mark H., of course, I started doing some experimenting.  

First off, Kudos for suggesting this pedal Mark.  I ended up tweaking the values per your post, but found for simplicity, ease of build and and cheapness, this pedal is hard to beat.

I found that for my taste, the stock vibe cap values did not work well however.  Something about the larger caps I think.  For example, the .22uf cap caused a slight loss of "distinctness" to the swirl.  That may be a good thing for some rigs, but not to my ears on my setup (Reverend Hellhound + G&L strat).  I tried the Shin-ei (spelling?) alternative values of .001, .0022, .003 and .0047.  This to my ears gave a slightly more "edgy" and bright sound.  Compared to the lightbulb based vibe I just built, I found that the Ropez with Shin-ei cap values had the same "characteristics" but not the same character if that makes sense.  In other words, it hit the sonic fingerprint but not with as much detail in the swirl compared to my vibe build.  

Still, the Ropez with Shin-ei mod took 1 night and a few bucks.  The vibe took me about 3 months and about 3 times the cost.  These factors alone are enough to put this project on your short list.

Next on the chopping block was the swamp vibrato mod.  In combination with the vibrato switch, I found this option more fun to play with than the vibe mod.  Here is the catch however.  Again, to my ears, grounding one or two of the stock 0.0033 caps made the swamp/tremolo/wah a little too dark.  I found that a much smaller cap value worked well here.  I found that the sound I liked, a fast tremolo lo-fi sound was best achieved with a single 470pf cap to ground in place of one of the stock 0.0033uf caps.  
Also, for me, the swamp vibrato thing really sounded good with the resonance cranked.  Thus, I left that control in.

To sum up my mods:

Added a vibrato switch
Added a switch to change between 1uf and 3.3 uf in lfo for that fast vibrato
added a 3pdt switch to switch between the stock 0.0033uf phasor caps and
vibe caps - hybrid Shin-ei based cap values  of 470pf, 0.002, 0.0033, 0.0047.  (note that one cap value is the same in phasor and vibe mode, so I only needed a 3 pole switch)

Added a switch to lift the 470pf cap to ground


Is this the ultimate?  Of course not, but in one pedal, I crammed a cool tremolo that sounds different enough from my EA to be really useful, a good sounding phasor, and a second flavor of swirly sound (vibe mode) that sounds different enough from my vibe build to be useful in its own right.

Has anyone else played around with this circuit?  If so, what did you conclude?

[Mark Hammer]

Between cap swaps, phase-filter option, and dry-cancel (vibrato mode), this thing has so much sonic flexibility for the time/money/parts investment its ridiculous.

Most existing implementations of the phase-filter thing assume same-value caps across all stages, but it doesn't need to be that way, so kudos to you for experimenting.  I'll keep some of your ideas in mind.

Most thanks go to "Huge" Tripps for originally providing the Ross schem and to Francisco Pena for making such a nice convenient layout.  I don't know that one can find so much modulation power in such an easy format anywhere else.

[Aharon]

One of my son's friends asked if I could build him a Univibe.  Rather than dealing with LDRs, and largely because Peter Snow was generous enough to give me a fistful of LM13600's, I figured I'd try building a Ross Phaser (again) with Francisco Pena's Tonepad layout, and doing the usual capacitor swap.

I can happily report that the Ross/Ropez delivers nice milky swirl with the standard cap change from all 3300pf caps to the .015, .22, 470pf, and .0047uf combo found in Univibes.

A few recommendations....

1) Though the Ross resonance control gives a nice robust edge to the sound of phase shifting, providing more flexibility than a Small Stone, in Univibe mode the resonance control does absolutely jack squat, so you can safely take it off, plus the save yourself the cost of the 1uf caps before and after it.  *Why* does it do jack squat?  Because equal value caps produce specific notches and peaks, which resonance emphasizes.  The discrepant cap values of the UNivibe spread phase shift around so there are not real obvious troughs and peaks to emphasize.

2) Sometimes you like a lot of sweep, and sometimes you don't.  The 10k current limiting resistor from pin 9 of the 13600 used as LFO sets the device for maximum (or close to it without contravening the input current limitations of the chip) sweep.  If you stick a 100k linear pot in series with the 10k resistor, you can limit the amount of sweep so that the effect is subtler.  Helpful for the phase shifter mode too.

3) Univibes don't really take much advantage of extra slow sweeps.  I also found that the stock LFO values for the Ross didn't cover off a useful range for the Univibe mode.  I changed the 3.3uf cap in the LFO (between pins 10/12 and V+) to a 1uf cap (now that I had two left over), and changed the speed pot from 500k to 100k linear.  This allocates a broader portion of the pot's rotation to small variations in faster speed so you can dial in something that sweeps with the rhythm of the song a little easier.

4) You can do it with the phase shifter as well, but anything that purports to be a Univibe should have a vibrato-mode switch.  In this case, you simply lift the 27k resistor between pin 1 and 6 of the dual op-amp to cancel the dry signal.  That resistor can be found adjacent to a 470k resistor between pins 4 and 5 of the chip.

5)  If you want to get deluxe, replace the 15k feedback resistor in the output stage of the op-amp to a 22k or even 33k for more output.  You can then stick a 100k volume control on the output instead of the 150k fixed resistor.

Is it the "ideal" Univibe?  I 'spose not, but it is dead cheap and dead easy and sounds nice, and it fits in a small box.

Hey MArk,mod 1,when you talk about the two resonance caps,are you  talking about the Regeneration control as it appears on the tonepad layout?
Thanks
Aharon

[gez]

2) Sometimes you like a lot of sweep, and sometimes you don't.  The 10k current limiting resistor from pin 9 of the 13600 used as LFO sets the device for maximum (or close to it without contravening the input current limitations of the chip) sweep.  If you stick a 100k linear pot in series with the 10k resistor, you can limit the amount of sweep so that the effect is subtler.  Helpful for the phase shifter mode too

When I had this phaser on the breadboard I did exactly this Mark, and I used the same value pot.  I don't know how it performs with the 'univibe' cap values, but I found it still sounded a bit odd at faster rates.  Yes, the amplitude of the LFO output is reduced by doing this, which gives less 'depth', but it seems that the very thing which makes this phaser sound rich at slow sweeps is the very thing which makes it sound a tad ridiculous with the rate cranked up, i.e. the hyperbolic shape of the LFO's output waveform.

What I did find satisfactory was sticking a huge value cap to ground in parallel with the Iabc inputs.  At the slowest sweeps the cap has virtually no effect on the LFO's waveform - there's a miniscule amount of distortion which is inaudible - but as the rate knob is turned up the amplitude decreases and there's subtle distortion of the wave form (it starts to resemble the more familiar non-linear triangle that so many LFOs churn out).

In practice, I found this gave a much better sound than the pot in series trick (though this is totally subjective of course).  Perhaps some will lament the loss of depth this mod causes, but for me it was a real improvement.

Cap values should be no greater than 220u and no smaller than 100u (parallel a couple to get something in between for best results).  I actually liked the 220u,  though it might be considered a bid too subtle by some.

[gez]

PS, If you've changed the stock cap value in the LFO which sets the rate, then I can't vouch that the above trick with the cap will work that well.  A different cap value might be needed (smaller if it's been speeded up)

[gez]

Yes, the amplitude of the LFO output is reduced by doing this, which gives less 'depth'

The waveform measured at any one of Iabc inputs that is.  Just thought I'd clarify that (I do realise these things have a current input so perhaps I didn't describe things too well).  :oops: