Gyrator for cry wah

[cordovaplata]

Hi. I made a crybaby but I missed the fact it uses an inductor (550mH). in my country it is hard to find inductors or audio transformers and ordering online the shiping cost is about 50% to 100% the price of the product. I want to know if I can use a gyrator instead of the 550mH inductor. and if it posible... which value should I use for the R and C and best opamp . thanks

[Mike Burgundy]

Nope. Gyrator circuits emulate inductors to ground.
Try a twin-T circuit (and read RG's "technology of the wah" at GEO as a starter!)

[earthtonesaudio]

There are gyrators which emulate "floating" inductors as well.

This one for example:
http://www.hollis.co.uk/john/phantomwah.jpg
[edit] Sorry, the above is not a "floating" inductor.  Rather it is basically the same circuit as this.
For true floating inductors, see "Generalized Impedance Converter" op-amp circuits.  They are tricky, often require component-matching and trimming.

[Mark Hammer]

For many years, producers have used parametric equalizers as "post-production wahs", adjusting the Q control appropriately, and moving the centre frequency control manually for a track.

While it works, my sense is that it will not perfectly mimic an inductor-based wah.  I may be misinformed, but my understanding is that the Q does not remain constant across the entire sweep range, becoming sharper at the toe-down portion.  But get confirmation of that from others.  Don't rely on me.

[cordovaplata]

Thanks for the replies. I tried the twin T circuit from the colorsound inductorless wah and add it to the feedback loop of a non inverting opamp buffer. it indeed had sweep, but it was very sharp the q and resonance (almost going into self oscillation) any ideas?

[joegagan]

lower the gain of whichever transistor is too hot.

not too familiar with the colorsound inductorless, but oscillation is often a sign of too much gain on one trans or the other.or both?

[R.G.]

You're right Joe. The Q of a Twin-T can be lowered by lowering the gain of the circuit for which the Twin-T is the feedback element.

Ditto, Mark. The inductor wah has a varying Q with sweep.

Good results can be had with either the Twin-T or the state variable filters. Actually, better wah tone is usually connected to using a modest-Q (5-10 or so) lowpass filter instead of a bandpass. Bandpasses necessarily cut out most of the bass, and leaving the frequency content below the resonance is what our ears have come to expect. The state variable does this naturally, as do all high-Q lowpasses.

[Mark Hammer]

.....which is why producers would use a parametric EQ, because it would leave the top and bottom intact and simply impose a noticeable moving resonance.

Hmmmm, did I just hear someone mumble Tychobrahe Parapedal?

[cordovaplata]

mmm parapedal... good idea! Joe. I can't lower the gain of the transistor since it is an opamp (maybe lowering the resistors in the T network may help reduce Q)

Any schem of a simple lowpass filter suitable for both guitar and synth?

-EDIT- circuits using opamps, since hi gain transistor and inductor are very expensive in my country)

[earthtonesaudio]

For what it's worth, if you want you can go to www.falstad.com/afilter and select file-import and import this code for Hollis' Phantom Wah circuit:

Code Select Expand

$ 1 5.0E-6 5 50 5.0 43
% 0 28853.998118144256
a 704 192 800 192 0 15.0 -15.0 1000000.0
a 832 208 928 208 0 15.0 -15.0 1000000.0
a 928 448 832 448 0 15.0 -15.0 1000000.0
a 928 352 1024 352 1 15.0 -15.0 1000000.0
174 800 192 832 256 0 1000.0 0.5 Resistance
g 800 256 800 272 0
w 832 192 832 160 0
w 832 160 928 160 0
w 928 160 928 208 0
w 800 192 800 128 0
c 800 128 976 128 0 1.0E-6 -0.013188010744820566
r 976 128 976 208 0 47000.0
g 976 208 976 224 0
r 656 336 656 416 0 47000.0
g 704 208 704 240 0
w 704 176 704 128 0
r 704 128 800 128 0 2200000.0
r 560 128 656 128 0 2200000.0
w 656 128 704 128 0
r 656 128 656 336 0 100000.0
w 928 432 928 400 0
w 928 400 928 368 0
w 1024 352 1024 400 0
w 832 448 832 400 0
r 832 400 928 400 0 10000.0
r 928 400 1024 400 0 10000.0
r 928 464 928 544 0 10000.0
r 832 336 832 400 0 10000.0
g 928 544 928 576 0
c 928 464 1024 464 0 4.7E-9 -0.06148075460360797
w 1024 464 1024 400 0
c 928 336 928 208 0 1.0E-8 -2.6026232113571797
w 928 336 832 336 0
w 832 336 656 336 0
g 656 416 656 432 0
c 560 128 560 192 0 1.0E-8 -0.06254506375437163
r 560 192 560 320 0 1.0E7
g 560 320 560 352 0
170 560 192 512 192 2 20.0 4000.0 5.0 0.1
O 976 128 1024 128 0


Compare to this version which uses a more standard gyrator:

Code Select Expand

$ 1 5.0E-6 5 50 5.0 43
% 0 28853.998118144256
a 608 192 704 192 0 15.0 -15.0 1000000.0
a 736 208 832 208 0 15.0 -15.0 1000000.0
a 736 448 640 448 0 15.0 -15.0 1000000.0
174 704 192 736 256 0 1000.0 0.6386000000000001 Resistance
g 704 256 704 272 0
w 736 192 736 160 0
w 736 160 832 160 0
w 832 160 832 208 0
w 704 192 704 128 0
c 704 128 880 128 0 1.0E-6 0.017371677345531855
r 880 128 880 208 0 47000.0
g 880 208 880 224 0
g 608 208 608 240 0
w 608 176 608 128 0
r 608 128 704 128 0 2200000.0
r 464 128 560 128 0 2200000.0
w 560 128 608 128 0
r 560 128 560 336 0 50000.0
w 736 432 736 400 0
w 832 400 832 336 0
w 640 448 640 400 0
r 736 400 832 400 0 100.0
r 736 464 736 544 0 200000.0
g 736 544 736 576 0
c 736 464 832 464 0 1.0E-8 -0.011016452031421986
w 832 464 832 400 0
c 832 336 832 208 0 1.0E-8 0.12603622833188088
w 832 336 560 336 0
c 464 128 464 192 0 1.0E-8 -0.01047419525897908
r 464 192 464 320 0 1.0E7
g 464 320 464 352 0
170 464 192 416 192 2 20.0 4000.0 5.0 0.1
O 880 128 928 128 0
w 640 400 736 400 0


[cordovaplata]

Jmmm interesting. I really like both designs; especially the second. which value of potenciometer you suggest me? and which opamp should I use. P.S How do I bias this? (I don't have  bipolar power supply :/ )

[earthtonesaudio]

The potentiometer value is not critical, it can be anywhere from 1k to 1M without changing the circuit.

Here's a way to bias for 9V operation:

Code Select Expand

$ 1 5.0E-6 5.023272298708815 75 5.0 43
a 704 272 800 272 0 8.0 0.0 1000000.0
a 832 288 928 288 0 9.0 0.0 1000000.0
a 832 528 736 528 0 9.0 0.0 1000000.0
174 800 272 832 336 0 10000.0 0.005 Resistance
w 832 272 832 240 0
w 832 240 928 240 0
w 928 240 928 288 0
w 800 272 800 208 0
c 800 208 976 208 0 1.0E-6 4.459108492675508
r 976 208 976 288 0 47000.0
g 976 288 976 304 0
w 704 256 704 208 0
r 704 208 800 208 0 2200000.0
r 560 208 656 208 0 2200000.0
w 656 208 704 208 0
r 656 208 656 416 0 50000.0
w 832 512 832 480 0
w 928 480 928 416 0
w 736 528 736 480 0
r 832 480 928 480 0 100.0
r 832 544 832 624 0 200000.0
c 832 544 928 544 0 1.0E-8 -5.635856952563856E-4
w 928 544 928 480 0
c 928 416 928 288 0 1.0E-8 -0.017056318610503496
w 928 416 656 416 0
c 560 208 560 272 0 1.0E-8 4.487403748923439
r 560 272 560 336 0 1.0E7
g 560 336 560 352 0
170 560 272 512 272 2 20.0 4000.0 1.0 0.1
O 976 208 1024 208 0
w 800 336 800 384 0
w 704 384 560 384 0
w 560 384 560 624 0
w 832 624 560 624 0
w 704 288 704 384 0
w 704 384 800 384 0
r 464 528 464 624 0 10000.0
w 464 624 560 624 0
r 464 624 464 704 0 10000.0
g 464 704 464 720 0
R 464 528 464 480 0 0 40.0 9.0 0.0 0.0 0.5
w 736 480 832 480 0
c 560 624 560 704 0 1.0E-5 4.491161487039965
w 560 704 464 704 0
o 29 64 0 34 10.0 9.765625E-5 0 -1


Basically you keep the input/output pull-downs tied to ground, but replace the other ground references with half of the 9V supply.

[cordovaplata]

Thanks! the opamp could be any? like a tl072 o 2 tl071 ?

[earthtonesaudio]

Yeah.  I should warn you I have not built this and on John Hollis' site the Phantom Wah is labeled as "experimental" due to excessive hiss.  So it may or may not work.  Simulation and real life aren't always the same.