Wah Inductors and magnetism

[joegagan]

no one has ever measured the resonant frequency.

or, if they did, they didn't put their findings on the web

[gmoon]

First, I agree completely with Paul.

There's no magical, perfect inductor--musical taste is just too subjective, and the historical Vox/Thomas circuit IS adversely effected by various issues (impedance), and there are too many variables. What sounds great with one setup...well, you get it.

However, a resistor in series with the inductor doesn't exactly alter the DC resistance in the way speculated. I've done experiments with my "variable inductor" wah (as well as lspice sims). Raising the DC resistance of the inductor lessens the 'Q', increasing the bandwidth. It's the parallel resistance we're talking about--which is why a resistor in parallel with the inductor is generally considered the 'Q' adjustment for the circuit.

But a series resistor raises the "floor" of the other frequencies, so those rejected frequencies are louder. It's pretty obvious on the simulations, too. The two effects aren't that easy to differentiate sonically; they both "open up" the sound, just in different ways.

Personally, I think a really narrow bandwidth high 'Q' sounds BAD; not vocal at all.


@GBlekas
Have you read RG's "tech of wah pedals" page? 'Cause the info he references on testing the old Fasel inductor for a magnetic offset is there. It's a very interesting theory.

Inductors are the least "perfect" of any components. That's why they come in so many shapes, core materials, gaps, etc. It's tougher to turn theory into practice.

I'm 100% certain that a strong magnet will alter the inductance. But so does the magnetic permeability of the core material--the "slug" in a variable inductor isn't magnetic. So it's not really the inductance per se that matters when introducing that magnetic offset (because the offset is creating a "less perfect" inductor).

It's also gonna be really difficult to test the effects of a magnetic offset on inductors, due to the different constructions, geometry, strength of the perm magnet, distance (inverse square law stuff), etc. It's worth doing, but will be hard to quantify.

(I'm a little skeptical of the pisotones.com page. Flexible fridge magnets are really weak, and have a pattern of alternating-polarity stripes to boot. But I can't definitively say it's nonsense either, of course...)

It's cool to see all those inductors lined up, so pretty-like. I've got a old Thomas Organ wah, with the "Stack 'o Dimes." I gotta say that I hated that thing for years...but now I love it.  At least one article lists the "Stack" as the "holy grail" (although most cite the Fasel), so why did I hate it? Probably because my amplifiers were really sub-par back then (those external variables.)

[Paul Marossy]

However, a resistor in series with the inductor doesn't exactly alter the DC resistance in the way speculated. I've done experiments with my "variable inductor" wah (as well as lspice sims). Raising the DC resistance of the inductor lessens the 'Q', increasing the bandwidth. It's the parallel resistance we're talking about--which is why a resistor in parallel with the inductor is generally considered the 'Q' adjustment for the circuit.

But a series resistor raises the "floor" of the other frequencies, so those rejected frequencies are louder. It's pretty obvious on the simulations, too. The two effects aren't that easy to differentiate sonically; they both "open up" the sound, just in different ways.

Personally, I think a really narrow bandwidth high 'Q' sounds BAD; not vocal at all.

Yeah, I was being a bit simplistic, I know. Interesting results to the tests you have done, though. I think that is why they mention in the patent documentation that it was found necessary to add a resistor in parallel with the inductor, because the Q was so high that it didn't sound good.

To quote the patent:

"The innate Q of the circuit, in fact, is so high that it was found desirable to insert the resistor 36 (what we now commonly call the Q resistor) in order to admit some of the non-favored (off-peak) frequencies into the amplifier, so that they could pass to the output. Without the resistor 36, virtually the only frequency appearing at the base 24 (what we commonly call Q1) would be the only selected frequency, because of the high Q of the circuit. The presence of resistor 36 allows all frequencies to be developed to a certain extent thereacross, and hence passed to the amplifier."

So adding a resistor parallel with or in series with the inductor both have the same end result - they both lower the "Q".

[GBlekas]

I posted the resonant frequency and bandwidth of the film can as well as a SOD and the text book ideal on page one of this thread.

What the number show was that the bandwidth of my particular film can was 1/2 of real world so adding the resistor should take care of that issue and bring it to where it needs to be. This is all being done by the numbers in attempt at getting at the root of why this wah doesn't sound right.

Also, if you were to take a great sounding real Clyde McCoy and reverse engineer each part actual value it takes the guess work and speculation out of things.

Also, playing guitar and wah for close to 40 years doesn't hurt..

In case you missed it here are today's numbers


Any reason raising the bandwidth would be a bad thing?

I heard an issue with the inductor and the numbers show this film can to have half the bandwidth of other wah Inductors.
So, my train of thought is this very well could be the issue.
So, I soldered in a 22 ohm in series and have it in the wah now.
Testing next....

[Paul Marossy]

GBlekas :

I see that you have resonant frequency numbers on your simulation program or whatever that is from, but have you actually determined what the real resonant frequency is in the real world by testing them? That's the only way you're going to get reliable real world data. I don't know what those charts are from. Are you measuring them with a computer based program or something?

If you are basing all of your work on what a sim program is telling you, I think it can possibly lead you to false conclusions. I am interested to hear what the result is of your latest test, though.

EDIT: If those numbers are correct, then the resonant freq doesn't even seem to be a factor for a guitar in standard tuning. An open low E string is 82.41Hz...

[GBlekas]

The internet statement that the Fasel is/was the holy grail is bullshit!

That article, of course, is most likely why Dunlop reissued them.

Some of what you read is intentional disinformation put out for obvious reasons if you see who is putting it out there.

You have to wonder how the Fasel could be the holy grail when the most sought after wah is the Clyde McCoy and that has a Halo in it and not a Fasel.

Think about it.......

@gmoon  I read RG's stuff a bunch of years ago and probably where the magnet idea came from. I used two different strength magnets in my testing and the agree that the frig magnet is probably very subtle in what it would do.  Over all the magnet smooths the highs of the wah....or at least that is what my ears are telling me.

@Paul M  the Film can types for the most part sound great so I think you surmise incorrectly. It just happens that this particular one doesn't sound great.

The best thing I think that I have ever did to understand the wah was to take a Clyde apart.
What I found is not written on the web or if it is I haven't seen it.
The other thing was to do repairs for people.
I have found some things that people did to their wahs were actually excellent ideas and once again not written on the web because I think some of these ideas were done by accident.

[GBlekas]

GBlekas :

I see that you have resonant frequency numbers on your simulation program or whatever that is from, but have you actually determined what the real resonant frequency is in the real world by testing them? That's the only way you're going to get reliable real world data. I don't know what those charts are from. Are you measuring them with a computer based program or something?

If you are basing all of your work on what a sim program is telling you, I think it can possibly lead you to false conclusions. I am interested to hear what the result is of your latest test, though.

EDIT: If those numbers are correct, then the resonant freq doesn't even seem to be a factor for a guitar in standard tuning. An open low E string is 82.41Hz...

This is just readings of the inductor out of the wah circuit.
Since the issue I was having I know to be an inductor issue I isolated to figure out what was wrong.

[Paul Marossy]

GBlekas :

I see that you have resonant frequency numbers on your simulation program or whatever that is from, but have you actually determined what the real resonant frequency is in the real world by testing them? That's the only way you're going to get reliable real world data. I don't know what those charts are from. Are you measuring them with a computer based program or something?

If you are basing all of your work on what a sim program is telling you, I think it can possibly lead you to false conclusions. I am interested to hear what the result is of your latest test, though.

EDIT: If those numbers are correct, then the resonant freq doesn't even seem to be a factor for a guitar in standard tuning. An open low E string is 82.41Hz...

This is just readings of the inductor out of the wah circuit.
Since the issue I was having I know to be an inductor issue I isolated to figure out what was wrong.

So then maybe it is all due to bandwidth. Or lack thereof. What program is that you are using? I've never seen it before.

@Paul M  the Film can types for the most part sound great so I think you surmise incorrectly. It just happens that this particular one doesn't sound great.

Yes, that is what I hear. I guess there's always at least one bad apple in the bunch, eh? 

[merlinb]

There is no way your inductors have a parasitic capacitance of over 4uF! You can't measure it just by hooking the inductor up to your cap meter.

[gmoon]

So adding a resistor parallel with or in series with the inductor both have the same end result - they both lower the "Q".

That's reasonable. But not in quite the same way, so alterations wouldn't sound quite the same...


Adding a parallel or series resistor to an inductor will alter it's response...but it's not the same as changing the parallel or series resistance of the inductor itself...

The "bandwidth" values here-- we're all a little unsure how these values (and others) are calculated. But they should equate with 'Q', yes? That would be consistent--a higher 'Q', a narrower bandwidth, and a lower DC resistance all fit.

But like I said, personally, I don't like a wah with a really high 'Q'. A really narrow peak with 10 or 15dB of boost, surrounded by a floor with 20dB of rejection is...nasty harsh. But practically, it's better to have a lower DC resistance because it's easy to lower the 'Q' with a resistor or two than it is to raise it. Regardless of whether such adjustments sound identical with an inductor with inherent DC resistance...

Resonance is a factor of the "tuned circuit," not of the inductor by itself. I'm sure the resonant frequency (in an identical circuit) of all these inductors is reasonably close (given their inductance). I'm sure from an EE POV (I'm NOT an EE), that 10-20% variation in the inductance are as acceptable as similar variations in caps and resistors (and equally unavoidable).

But given all those other practical limitations of (cheap) inductors, then it's not the inductance per se that imparts the subtle differences... That's assuming there is a "holy grail" to be found  , and the good wahs aren't just "tuned in" well with the other components. (yeah, Halo, Stack or Fasel, doesn't matter to me.)

So if you're looking to uncover the subtleties; i.e. the non-linear response cited by RG, you'd be better off with a signal generator and a 'scope. And test it throughout the audio spectrum.  A meter (with an unknown test frequency) can't show you the harmonics, etc. I've got a similar meter, and I love it, but it's pretty limited.

[Paul Marossy]

But like I said, personally, I don't like a wah with a really high 'Q'. A really narrow peak with 10 or 15dB of boost, surrounded by a floor with 20dB of rejection is...nasty harsh

Yeah, I HATE a wah that makes me cringe whenever I go to use it. I'd be modding that thing as soon as I got it, whatever it was, vintage or not.

But given all those other practical limitations of (cheap) inductors, then it's not the inductance per se that imparts the subtle differences... That's assuming there is a "holy grail" to be found  , and the good wahs aren't just "tuned in" well with the other components. (yeah, Halo, Stack or Fasel, doesn't matter to me.)

I don't really care about the actual inductor used, either. It's only one part of the bigger picture.

So if you're looking to uncover the subtleties; i.e. the non-linear response cited by RG, you'd be better off with a signal generator and a 'scope. And test it throughout the audio spectrum.  A meter (with an unknown test frequency) can't show you the harmonics, etc. I've got a similar meter, and I love it, but it's pretty limited.

I agree, there's got to be some better way of testing these inductors. I am still not convinced that it all just boils down to some certain DC resistance and mH value and then you get a magical inductor. There must be something else going on, harmonics or something.

I thought maybe I was onto something yesterday about resonant frequency, but if all types of indcutors average around 70Hz, it seems like a moot point. Anything above 70Hz will be diminished as it goes up in frequency, right? So if you had an inductor with that low of a resonant frequency, I would think that it wouldn't be a very lively sounding wah pedal, or would it? You wouldn't want the resonant peak too high because it would make it sound too harsh, but if it were too low, then it would be muddy and blah sounding. Then again, I keep coming back to you can't take the inductor out of context, because everything else in the circuit matters - because as you said, it's a tuned circuit.

[merlinb]

I thought maybe I was onto something yesterday about resonant frequency, but if all types of indcutors average around 70Hz, it seems like a moot point. Anything above 70Hz will be diminished as it goes up in frequency, right? So if you had an inductor with that low of a resonant frequency, I would think that it wouldn't be a very lively sounding wah pedal, or would it?

There is no 70Hz resonance; GBlekas measure his inductor's wrongly. They don't have ~5uF of self capacitance!

[Paul Marossy]

I thought maybe I was onto something yesterday about resonant frequency, but if all types of indcutors average around 70Hz, it seems like a moot point. Anything above 70Hz will be diminished as it goes up in frequency, right? So if you had an inductor with that low of a resonant frequency, I would think that it wouldn't be a very lively sounding wah pedal, or would it?

There is no 70Hz resonance; GBlekas measure his inductor's wrongly. They don't have ~5uF of self capacitance!

Ha ha, good point!

[joegagan]

There is no way your inductors have a parasitic capacitance of over 4uF! You can't measure it just by hooking the inductor up to your cap meter.

so, does this mean that the inductor does not have static capacitance? is it dependent on voltage applied?

[merlinb]

so, does this mean that the inductor does not have static capacitance? is it dependent on voltage applied?

It will have some parasitic capacitance in parallel with it, but it will be in the order of picofarads. You would need a signal generator and o'scope (or audio millivoltmeter) to measure it.

[joegagan]

thanks merlin. i was having trouble understanding what the calculator in george's post was looking for.

do george's findings still have some relevency ( not for freq range but general comparison) since his capacitance numbers are within a similar range?

gonna go play with the calculator to see if scaling the capacitance numbers down into pf ranges ( identical values for purpose of comparing) gives me any clues.

[merlinb]

george's findings still have some relevency ( not for freq range but general comparison) since his capacitance numbers are within a similar range?

To be honest, I don't think so. For example, to get a self resonant frequency down below 10kHz where it *might* make some difference would require more then 500pF of self capacitance, and I don't think that's likely.

[GBlekas]

GBlekas :

I see that you have resonant frequency numbers on your simulation program or whatever that is from, but have you actually determined what the real resonant frequency is in the real world by testing them? That's the only way you're going to get reliable real world data. I don't know what those charts are from. Are you measuring them with a computer based program or something?

If you are basing all of your work on what a sim program is telling you, I think it can possibly lead you to false conclusions. I am interested to hear what the result is of your latest test, though.

EDIT: If those numbers are correct, then the resonant freq doesn't even seem to be a factor for a guitar in standard tuning. An open low E string is 82.41Hz...

This is just readings of the inductor out of the wah circuit.
Since the issue I was having I know to be an inductor issue I isolated to figure out what was wrong.

So then maybe it is all due to bandwidth. Or lack thereof. What program is that you are using? I've never seen it before.

@Paul M  the Film can types for the most part sound great so I think you surmise incorrectly. It just happens that this particular one doesn't sound great.

Yes, that is what I hear. I guess there's always at least one bad apple in the bunch, eh?  

Yes sir! OK, ends up adding the 22 ohm resistor did it for this wah! Whoo hoo!
I will record it later and post a few sound clips of it.
Highs are where they should be and very sweet sounding, not harsh!

So, the band width theory totally nailed it for this inductor and wah.

The calculator tool was found here:
http://www.calctool.org/CALC/eng/electronics/RLC_circuit

[GBlekas]

There is no way your inductors have a parasitic capacitance of over 4uF! You can't measure it just by hooking the inductor up to your cap meter.

Well that's just great to know but if you could explain this to me it would be greatly appreciated.

I measured using the cap setting on this meter and went from inductor to inductor to inductor and then a 4.7 uf electrolytic cap.
Now, the 4.7 reads upwards of 5 so I think the meter is not all that accurate but it is reading capacitance.
I did not change meter settings and only changed item under test.



Right now, barring the meter is reading the temperature in the room,   what I am thinking is that the 4.7 uf cap to ground in the wah circuit is no accident.

[phector2004]

You're measuring capacitance from both ends of a long conducting wire, not two separated sheets of conductor.

I think if you put a cap in series with it, you can get a better reading, as it will block DC, though you'll have to do further calculations.