Marcos-Munky's Auto-Wah Build Report!

[Arn C.]

http://www.geocities.com/munkydiy/autowah.jpg

I built this circuit yesterday afternoon and brought it home last night and gave it a try.   I like it alot!  Quite easy to build and works great.  I will be using this out when I play.

I used transistors with an HFE of 206 for the first transistor and HFE of 208 for the second transistor.  I sort of tried to match them.  For what reason? Who knows.....Maybe one should be less gain than the other,  maybe they should be higher or lower gains.    I just put a little toggle in for the in/out switch on the left side of the schematic.  And as Marcos told me,  depending on where the (RATE) pot is set, the led flashes accordingly.  
The only thing that can cause a problem is the size of the pot (5 Megs!)  
Fortunately I just happened to have only one.  
Maybe someone could figure out how to change something in the circuit so we could used the more standardized 1 Meg pot for the rate.
Thanks Marcos!
Peace!
Arn C.

[Marcos - Munky]

Cool, Arn. Other mod that someone could try to add (Mark Hammer?) is a depth pot. In the magazine, there isn't any info about the transistors, so I think they needn't to be matched.

[petemoore]

Using a 1 meg for fine tune, and switched resistors for coarse tuning...can get anything from 0meg to 5meg...might take a few switches.
 Usually I use just one or two settings, [for the most part] you could use switches and fixed R's only and get that...er jest arder the dan thang

[petemoore]

What are those things on the schematicwith the numbers 1-14 around them, that look like wrapped hotdogs?
 I looked at NTE for crossref's using entry: "4001B".   ...nada
 I socket and thinking something like 3904's woul do for the tranny's...and that the 10n's are weeny caps...ten nanofarad's
 Everything else I got/can get.
 I would like see My Nurse [Quacky] compete with MONTAGEM ..!!!

[rx5]

Arn C.

have you read about R.G.'s article about POTs?

I think he mentioned you could get higher resistance fro pots by SANDING the carbon element with fine grit sandpaper..that way, it would thin out.....

anybody DONE that???

Regards,

Ralph

[C Bradley]

What are those things on the schematicwith the numbers 1-14 around them, that look like wrapped hotdogs?
 I looked at NTE for crossref's using entry: "4001B".   ...nada
 I socket and thinking something like 3904's woul do for the tranny's...and that the 10n's are weeny caps...ten nanofarad's
 Everything else I got/can get.
 I would like see My Nurse [Quacky] compete with MONTAGEM ..!!!

Those are either NAND gates or NOR gates, but they're drawn weird. If they're NOR gates they should have a bullet shape with the concave input, and if they're NAND gates they should have a flat input side. Strange. Maybe it's a 4011 CMOS NAND gate?

[eliktronik]

I believe they're NOR gates that with the input line going to both input pins. That's why they look wierd...

[smallbearelec]

It is possible to sand the element of a composition pot to increase the resistance IF the pot's construction lets you get at the "wafer". See the pics in my article on the Tremulous Bear.

I wound up being sorry that I chose this method when the 4.7 meg pot that I was using disappeared from U.S. shelves. A better way to get an odd value and/or taper is to use a photocoupler to "amplfy" the resistance of a pot. Keen describes this method in an article at GEO. I have worked out the component values needed for using the photocoupler work-around in my Trem, and I will re-work the board if I can ever get ahead of my order volume.

Regards
Steve D
www.smallbearelec.com

[Arn C.]

I am happy that people are interested.  It really is a nice simple pedal and works great!

Thanks Petemore:  Nte cross is NTE4001B   I used what I had, which was an RCA  CD4001AE.

As Eliktronic said:  Yes they are NOR Gates tied together, example:
5&6, 8&9, 12&13 are tied to each other.

10nf=.01uf

Ralph:  I really don't want to try that one! But thanks anyhow!

Mark Hammer:  Any suggestions how to tie in a "Depth Pot"?

Thanks!
Arn C>

[Arn C.]

Steve wrote:  
A better way to get an odd value and/or taper is to use a photocoupler to "amplfy" the resistance of a pot. Keen describes this method in an article at GEO.

Thanks Steve, I will check that out!
Arn C.

[Mark Hammer]

My guess is that a sweep depth control could be made from using a 250k pot wired as variable resistor to ground, in parallel with the BC549 in the middle of the schematic.  The range is obviously set by the two 10n caps, and the centre frequency set by the effective resistance to ground of the transistor, plus the caps.  So by shifting the range with the caps and varying how much impact on that effective resistance-to-ground the transistor has by using a parallel resistance, you should be able to get different amounts of sweep/modulation in different frequency ranges.

At first I thought this schem was a poor man's MXR Envelope Filter, with the 4001 pulse-width modulating the transistor, but then I ealized this is probably the *otehr* kind of auto-wah where the 4001 forms an LFO.  Am I correct?

[gez]

Looks like a simple relaxation oscillator with a square wave output - is it choppy sounding?

Edit: And does it tick? (these type often do)

[gez]

The 100k and 47u cap would probably smooth the output from the inverter 'buffer', so maybe it isn't that choppy.  

Reminds me of one of Chatwin's circuit for a tremolo, he used a similar set up for the LFO.

[gez]

The two NOR gates facing downwards are wired as inverters and act as buffers.  If you disconnect the one driving the LED and reconnect it via a reverse bias diode, then connect the anode of the LED to the positive rail and it's cathode to the output of the inverter (via the 1k resistor - or a larger value if you use a low-current job), you end up with a 'Flashing Millennium'.  

Stick a pull-down resistor at the output and wire the switch up so that pins 8&9 connect up to the output when the effect is kicked out.  The inverter's input will be grounded and it's output will go high so no voltage occurs accross the LED and it turns off.

I've done this with CMOS LFOs in the past and it works a treat.

[rx5]

Steve,

nice idea about the LDR

that would be the best pot if your about to go higher than 1M....

only problem is my DMM has a max reading range of only 2M...

how would i???

[Arn C.]

Gez wrote: is it choppy sounding? no

Edit: And does it tick? (these type often do) no

I am telling you, it is great!  Not sure how to make that depth control tho.  I almost need to see a diagram or to have it word for word what to do to achieve this.  Maybe someday it will all sink in!
Arn C.

[gez]

If any of you are having problems with finding a 4M7 pot, then try a 1M pot, reduce the 470k to 100k and increase the 330n cap to 470n.  This 'should' be ok.  making the resistors smaller speeds it up, which is why I suggest using the 470n cap.  You can fine tune the thing by increasing/decreasing  the 'stop resistor' or changing the cap value.

I don't have the formula handy to calculate the values, but the ones I've given should be o.k. (if not then apologies in advance)

[Arn C.]

Thanks for the info Gez!

I do recommend using shielded wire for the input and output of the circuit to the jacks with only one end of the wire shielding to ground.  Real quiet!

Arn C.

[Arn C.]

One more thing:  When I tried this at home with guitar and amp,  I just had the guts of this effect(not enclosed) and it was quiet,  I will be installing this one in a plastic box.

Arn C.

[gez]

If someone could verify the values I gave (or 'tweaked' values), get back to me as in the back of my mind I 'think' I added a 'swing' pot to one of these LFOs once.  It's dead simple to do, but I'd need some values to go on.  

Basically it changes the mark/space ratio allowing you to get a triplet feel, which sounds great with autowahs, especially on rock shuffles/psychedelic blues (is there much call for this?).  You also get a choice of direction, so it can either accent the beat as it sweeps down OR as it sweeps up.  

Only draw back is that it limits the range, but I think I got round this by using two cap values and switching between the two.  I'll have to dig out my note books and see if I can find the bugger (It was quite a while ago and my memory is a little hazy).