News:

SMF for DIYStompboxes.com!

Main Menu

Wah with fets?

Started by Dingleberry Electronics, January 25, 2007, 08:17:59 AM

Previous topic - Next topic

Dingleberry Electronics

Hi!

Has anyone tried to make a wah-wah pedal using J-fets instead normal npn-transistors.
Could there be any major tonal improvements or would the differences be inaudible?
How about mosfets?
Are we stuck on transistors  just for the tradition and because all "those good old recordings" were done with
vox or cry baby so that's why we're using transistors to get exactly the same sound...(which we never achieve, because it's in the playes fingers...)
(Now all the mojo-vintage-purists will start to hount me...)
Is there any electrical basis for using transistors? I've seen Tube-Wah schematics, but never a fet-wah.
Why?

If you have experiences be free to share.

R.G.

If you've read "The Technology of Wah Pedals" at GEO, you already know that any gain block will work. The fact that it's bipolar, JFET, MOSFET, opamp, whatever matters not at all.

What is critical to the inductor wah circuit is that the gain block must be inverting  and suitably stable under feedback to make the amplified-capacitor trick work.

Other than that, what you hear is the tone of the amplifying circuit. If you like JFETs, JFETs it is. If you like MOSFETs, go with MOSFETs. If you want tradition, go with bipolars.

If you want predictability, use opamps. If you want wide range, go opamps and state variable.
R.G.

In response to the questions in the forum - PCB Layout for Musical Effects is available from The Book Patch. Search "PCB Layout" and it ought to appear.

zachomega

Hmm...I'd imagine J-fets would improve the noise of the circuit too, right?

-Zach Omega

R.G.

QuoteHmm...I'd imagine J-fets would improve the noise of the circuit too, right?
Maybe, maybe not.

Noise is inextricably linked to impedance. Only resistive impedances make thermal or excess noise. Pure capacitance and pure inductance cannot make noise. But the electrons in resistors are subject to thermal agitation, and that's where the common hiss comes from. The bigger the resistance, the bigger the noise.

It happens that noise is an issue that is affected by impedance matching. If the source of your signal has a low impedance, you can minimize the noise you amplify by using a low impedance amplifier to amplify the signal. A high impedance signal needs a high impedance amplifier to keep noise down. For impedances from about 1K on down, bipolars, even common-base bipolars are better than FETs. For impedances over about 100K, you want the FETs. For guitar signals, you're firmly in the middle. So the exact situation of the circuit makes a difference. You can't make a blanket statement that one is better than the other. Well, you can, but it's not necessarily a true statement if you do.

R.G.

In response to the questions in the forum - PCB Layout for Musical Effects is available from The Book Patch. Search "PCB Layout" and it ought to appear.

Paul Marossy

Might be a worthwhile experiment for one of us wah freaks to try an FET wah. Might sound pretty good. Of course, we are talking about degrees of subtlety here. I wouldn't expect anything very noticeable.

rockgardenlove

I've got a wah made with a bunch of sockets.  I was messing with some transistors I had (including FETs) and they didn't make a huge difference, just like Paul said up there.



Paul Marossy

QuoteI've got a wah made with a bunch of sockets.  I was messing with some transistors I had (including FETs) and they didn't make a huge difference, just like Paul said up there.

I rest my case!  :icon_wink:

Goatmeal

Hi.I have had this on my bread board since christmas.I used a mini booster for the gain stage,and a jfet buffer. I have been getting some very cool tones, but seem to have trouble getting the sweep where  I like it. If you look at RG's articles on the tube wah,and the op amp wah,you can get the basic idea of where to insert your favorite gain stage and buffer.  Most of the typical values seem to work, .22's ,feedback resistor about 2k. The Q resistor can be quite high,it will wah without it,but about 470k is whats stuck in the bord right now.Just a crappy R.S. transformer in there,and decent tone. Go build it,or try and get the mosfet version done before me. Thanks RG!

Paul Marossy

Goatmeal -

The tone might improve quite a bit with a real wah inductor in it. That's what I've heard a few people say on this forum, but I have no personal experience with this.  :icon_cool:

Chico

Hi everyone.  It has been a long time since I have posted here.  I have been pretty busy, you know the excuses.  Hope to get back more regularly this year.
At any rate, I just build and boxed up my version of a FET Wah as my first project of the year.  I put it in an old late 80s crybaby pedal.  My version, is similar to what Goatmeal describes with just a few tweaks.

As typical, R.G. is the authority here.  Read his Foolin with Fets, History of the Wah and everything simply falls into place.  Now, I think of a wah effect as nothing more than an amplification stage that feeds a voltage follower and a corresponding feedback network that ties the output of the voltage follower back to the input of the amplification stage. 

With this simple model, I came up with a FET gain stage based upon the mu-amp.  This stage is nice because you can twiddle the value of a source resistor to ground (bottom FET) in order to set the gain such that there is no apparent volume change between effect on/off.

Also, you can twiddle the value of the cap between the gate and source of the top FET (in combination with the value of the feedback resistor in the feedback network) to alter the way that the lower frequencies respond.  In my tests, values around 10nf/1.5k feedback gave a response that "felt" similar to the traditional BJT wah.  Larger cap values, such as 100nf on up, esp. in combination with relatively larger feedback resistance, e.g., 2.7k-6.8k, resulted in a response where the low frequency felt "exaguraged" to my ears.  In my tests, as the cap value and/or feedback resistor value increased, the response soundes more like a lpf (low pass filter) with resonance compared to a morebandpass "ish" response.  I personally like a little extra low frequency response.   In my rig, this results in a more balanced transition from wah on/wah off.  YMMV

My feedback network is about like what you see in the Vox/Crybaby style, a simple RLC circuit.  I used an old red Fasel that I had.  I also prefer lower Q resistor value (33k) for my rig, which is a CIJ Tele into Dr. Z MAZ 18.  Again, YMMV.

Best regards