ICs for phaser

[carlozsulca]

I am looking at what ics I can use in the phaser that I am going to design, I will base myself on the beautiful sound of the small stone, I was looking at the different versions of the pedal and there is a point where the ca3094 is changed by the lm13600 and then by the lm13700, so that I read and heard in the comparisons of both versions of the small stone , apparently more depth is perceived in the pedal than in the one with the ca3094, then I saw the schematic, because I understood the small stone basically there are 4 high pass filters in series controlled by an exponential waveform oscillator that varies the cutoff frequency, now my question is, why can't the lm13600 or 700 cut more? What calculations determine under what voltage is the cutoff frequency? Since the formulas and high pass filter settings in the datasheet are not as in the small stone, what changes do I have to make to make an oscillator with ba6110 to save space and not vary the ICs? and most importantly, can the ba6110 give me more depth compared to the lm13x00?

[Kevin Mitchell]

The format and facts of your post are confusing - to me at least. Some points I certainly haven't heard of. References would be helpful.
Did you take a look at the datasheets?
LM13700
CA3094

[Mark Hammer]

NEVER confuse someone's posted experience and comments about a single pedal with how all other units from the same manufacturer with the same design behave.  Virtually all components have tolerances, and the older the pedal, the more likely it is to use 5% resistors rather than 1%.

I've made myself several Ross phaser clones (courtesy of the Tonepad layouts graciously provided by Francisco Pena).  They use LM13600s for the phase-shift/allpass stages, and a 10k current-limiting resistor to feed the LFO to the 4 stages.  I used 9k1 on one of mine, just to see, and the sweep is spectacular.  If you are familiar with the Electric Light Orchestra's song "Strange Magic", it sounds like that, sweeping waaaayyyyy up.  None of my others do that.  The other day, going through my bin of unboxed circuit boards, I found another populated Ross board.  I replaced the 10k resistor with 9k1 and a 20k pot in series to vary the sweep width.  Even with the resistance dropped to 9K1, I couldn't get the width of sweep achieved in the first unit I made.  I expect that some of this is a result of the LFO, rather than the phase-shift stages themselves.  Both Ross and EHX used transconductance amps for the LFO; 13600s for the former and 3094 for the latter.

So, I wouldn't attribute too much to the use of 3094 vs 13600 vs 13700, when it comes to sweep width.  If I were you,I would save my BA6110 chips for a nobler purpose, like a compressor.

[ElectricDruid]

+1 agree with Mark. The sweep width isn't down to the chips used. It's down to the LFO and how the control current for the OTAs is generated.

[StephenGiles]

Back in the days when I took advantage if the TI sample program, I remember trying each quad and dual opamp in breadboarded phasers - none of them gave any difference in sound !!

[Rob Strand]

+1 agree with Mark. The sweep width isn't down to the chips used. It's down to the LFO and how the control current for the OTAs is generated.

Using an OTA for the Ross circuit certainly simplifies the job of getting the quadratic LFO waveform.  The LFO output is fairly reproduceable as well.   Using JFET shapers is hit and miss.

As far as the all-pass filters go OTAs are probably easiest way to produce consistent phasing.   Low tolerances, no part sorting.

[Vivek]

Can an FV-1 easily replicate the whole circuit ?

I know it can emulate two BBD comfortably. And some old BBD can cost somewhat comparable to an FV-1

[ElectricDruid]

Can an FV-1 easily replicate the whole circuit ?

I know it can emulate two BBD comfortably. And some old BBD can cost somewhat comparable to an FV-1

You can make a phaser with the FV-1 easily enough, yes. Allpass filters are pretty easy digitally, and the FV-1 has LFOs built in.

I'm not sure I'd say it's a "replica" of something like the Ross circuit though. The response of the LFO is rather different, and OTAs have subtle distortion that would be missing. Perhaps with some extra work you could get pretty close.

[carlozsulca]

NEVER confuse someone's posted experience and comments about a single pedal with how all other units from the same manufacturer with the same design behave.  Virtually all components have tolerances, and the older the pedal, the more likely it is to use 5% resistors rather than 1%.

I've made myself several Ross phaser clones (courtesy of the Tonepad layouts graciously provided by Francisco Pena).  They use LM13600s for the phase-shift/allpass stages, and a 10k current-limiting resistor to feed the LFO to the 4 stages.  I used 9k1 on one of mine, just to see, and the sweep is spectacular.  If you are familiar with the Electric Light Orchestra's song "Strange Magic", it sounds like that, sweeping waaaayyyyy up.  None of my others do that.  The other day, going through my bin of unboxed circuit boards, I found another populated Ross board.  I replaced the 10k resistor with 9k1 and a 20k pot in series to vary the sweep width.  Even with the resistance dropped to 9K1, I couldn't get the width of sweep achieved in the first unit I made.  I expect that some of this is a result of the LFO, rather than the phase-shift stages themselves.  Both Ross and EHX used transconductance amps for the LFO; 13600s for the former and 3094 for the latter.

So, I wouldn't attribute too much to the use of 3094 vs 13600 vs 13700, when it comes to sweep width.  If I were you,I would save my BA6110 chips for a nobler purpose, like a compressor.

then it is a question of the oscillator and not of the configuration of the filter , I will make a pcb by hand with all the configuration and I will do tests with the oscillator, I managed to edit the code of a tap lfo so that it would give me the same waveform as the oscillator of small stone, I guess now I have to do some modification so that everything works correctly
thanks for the clarification

[carlozsulca]

Can an FV-1 easily replicate the whole circuit ?

I know it can emulate two BBD comfortably. And some old BBD can cost somewhat comparable to an FV-1

Of course, the use of the fv-1 solves many headaches and reduces everything to programming, but now it is difficult to get them (especially in my country that shipping is very expensive)

[iainpunk]

3094 vs 13600 vs 13700

`
hmmm, i thought the small stone uses the CA3060? attest the green Sovtek Small Stone i have here has CA3060's in it.

if you really want that control over the specs of the OTA, build a few discrete ones, using transistors and diodes. (just kidding, DIY OTA's arent that accurate if you don't spend time matching some of the transistors, and its definitely not worth the effort)

cheers

[Mark Hammer]

Never heard of the 3060.  Why haven't I?

[ElectricDruid]

Me neither. It looks like a very handy chip, so it's a pity that it's (a) rare, (b) obsolete, and (c) out of production!! 

https://pdf1.alldatasheet.com/datasheet-pdf/view/176969/HARRIS/CA3060.html

[iainpunk]

oops, my bad here,
sorry for the confusion i meant a CA3080, not the 3060.

ive been helping out a friend who builds synths and he as a VCF based on the CA3060, while my green Russian Small Stone runs CA3080's

sorry for the confusion...
and youre welcome if you liked finding out about this expensive/rare/obsolete chip

cheers