Let's talk wahs

[ayayay!]

Thanks gmoon.  I'll probably buy it.  I was just about to drop $30 on a capacitance meter, so this is fine with me. 

[Dai H.]

So are you measuring Henries with a meter guys?  Or are you doing the old, "run a 1khz sine through and test resistance and whatever else formula" to determine mH?  I'm looking for a *cheap* meter to measure inductance.  Any links are appreciated.

I bought an LCR meter from Hong Kong on ebay, total of ~$25 (shipping included):
http://stores.ebay.com/worldzon

Not affiliated with them. You'll probably find something similar (better, actually) in the $40-$75 range (USD.)

So far, it's worked well. I'm not wild about buying from Hong Kong, but I do have a budget. Tough to beat the price.

4 Inductance Measuring Ranges (not great for RF work, but excellent for audio):
20mH / 200mH / 2H / 20H, +/- 2.5%, + 5 digits;

(measures capacitance, too!)

holy cow that's amazingly cheap for an LCR meter

[Sauerkret]

could you turn anny wah in to a inductorless wah?

[yeeshkul]

I been marking up schematic of whas

Some thoughts.  Mostly about transistor operation points in three different whas

Crybaby like first
Q2 in the two transistor crybabay like circuit should not matter all that much with high enough hfe.
  Why?
  One it is an emitter follower input R ~= hfe x emitter R so lets take a hfe of say 200 x 10,000 = 2,000,000 or about 2 meg, if hfe is higher input resistance is higher.
  Two the transistor is biased via the 470K from the first transistors collector if the higher the hfe the lower the voltage drop across the 470K because of less base current needed.  R.G. writeup has this stuff in it.
  Three less loading on the pot and Q1 collector the higher the hfe of the transistor
  I would use a high hfe at Q2 and not worry to much

Q1  in a crybaby should not make much difference again if hfe high enough (please keep reading) a 22K collector resistor and 390 ohm to 510 emitter set most of the open loop gain BUT the bias is set by the voltage divider bias network of the 470K from the collector to the "top" of the cap and 82K etc resistor, then you have the DC resistance of the inductor and parallel damping resistor and 1.5k resistor to the base.  The voltage at the node of the 470K, 6uf on up electro, inductor and 82K and the series resistance from that node to the base is what is setting the bias for BOTH transistors.  SO adjust the voltage divider for the TWO TRANSISTOR OPERATING POINTS.  I would adjust the 82K because it is bypassed but the electro cap and is at a DC voltage but an "AC ground".


So for a crybaby build I would use >200 hfe and adjust the one 82K resistor value to bias the two transistors to taste.  I would use a high hfe for Q2.

Now when you look at the Boomerang it looks different.

Boomerang

The bias for Q1 is different a voltage divider is not used.  Open loop Q1 gain partly controlled by the 10K C and 120 ohm E resistors.  Bias set by the series resistance of the 1.5meg and (inductor and damping resistor 47K in parallel) and 4.7K from collector to base of Q1.  Again the end of the inductor and 47K and 1.5meg are at "AC ground" if the cap has a high enough value.  In this case adjust the 1.5meg,  Q1 hfe will be more of an issue than the crybaby bias IMO need to think about this some more.
A bigger difference is the bias of Q2 it is not biased via the collector of Q1 and a resistor.  It is biased via a resistor C to B and the input resistance is hfe x 8.2K in parallel with the collector to base bias resistor.  Different C to B resistor value should give different emitter to ground operation points.

Q2 will matter because of the bias used.

  If one was the use the crybaby Q2 with the the 470K another thing changes the 470K lets more signal pass via the 470K and there is the signal from the wiper of the wha pot.  In the boomerang the signal comes only from the wiper of the wha pot to Q2

Q1 will matter a bit BUT if you get in a hfe range you can control the bias with changing the 1.5 meg resistor value.

Colorsound

Colorsound more like the crybaby except for the use of the grounded emitter open loop gain stage at Q1 and no damping resistor across the inductor.  I would use a high hfe at Q2 and Q1 I need to think about some more.  Q1 bias and Q2 bias again can be controlled via adjusting the 100K resistor value

Thanks Gus! these things are definitely good to know 

[Exactopposite]

Colorsound

Colorsound more like the crybaby except for the use of the grounded emitter open loop gain stage at Q1 and no damping resistor across the inductor.  I would use a high hfe at Q2 and Q1 I need to think about some more.  Q1 bias and Q2 bias again can be controlled via adjusting the 100K resistor value

Be careful with going with too high of an HFE in the colorsound. I tried bc109C's in mine before and they caused the wah to have feedback. Bc108b's cured the problem.

[Paul Marossy]

Colorsound

Colorsound more like the crybaby except for the use of the grounded emitter open loop gain stage at Q1 and no damping resistor across the inductor.  I would use a high hfe at Q2 and Q1 I need to think about some more.  Q1 bias and Q2 bias again can be controlled via adjusting the 100K resistor value

Be careful with going with too high of an HFE in the colorsound. I tried bc109C's in mine before and they caused the wah to have feedback. Bc108b's cured the problem.

I used 2N5089s in my clone, and they work fine, no feedback problems even with heavy distortion. They both have an Hfe of over 800. But I also changed the voltage divider arrangement on the output to be more like a volume control (100K trimpot in series with the 100K resistor to ground), so that may be why.

Adding that trimpot helps a lot when you have the effected signal level lower than the bypassed level, which happens on some of the original Colorsound wahs.

[kierc]

Just wondering if anybody had saw my problem on page 4? got pushed away with this new page 

Cheers,

[Gus]

Just drew up a schematic inspired by the different whas.  I will post it when I get the scanner working.

It is a little different and hopefully a fun build.

[Paul Marossy]

Just drew up a schematic inspired by the different whas.  I will post it when I get the scanner working.

It is a little different and hopefully a fun build.

Sounds interesting...

[MohiZ]

Just wondering if anybody had saw my problem on page 4? got pushed away with this new page  Sad

Cheers,

I'm just guessing but could it be that the wah pot has dead spots in it, causing it not to conduct at all in certain positions. The sudden change in resistance MIGHT cause the clicking.

[kierc]

hmm sounds interesting... in a bad way lol

The pot was working great with the previous circuit (had no added resistor) - it's the ICAR pot from the Whipple guy...

I'm gonna try removing the resistor, and then try with a 33k between 1 and 2 and between 2 and 3 and see what happens...

[Gus]

http://www.aronnelson.com/gallery/main.php/v/gus/wha1.gif.html?g2_imageViewsIndex=1

Q2 is bootstrapped for a controlled operating point and min loading of the wha pot and the use of a 8.2K emitter resistor, also this was done to not use the 470K to base for bias and to remove the mixing it causes( like the boomerang)
Q1 is setup as like the colorsound with the grounded emitter.  Note the three rough calculations assuming .6VDC B to E the collector will be about(((.6VDC/100k) x 470K)+.6VDC) or (((.6VDC/82K) x 470K)+.6VDC) or ... from ground.
I like the voltage diver bias for Q1 better than the stock boomerang bias
I would try higher Hfe for Q2 and different Hfes(>200) for Q1 and a 22K Q1 collector resistor to start.  Maybe a 100K in the voltage divider to start

One can adjust the bias of both transistors independently

One of many wha variations I have been drawing.  this was because Paul liked both the colorsound and boomerang and I think this might be the parts of each that are a little different than the crybaby.

Have fun

Gus

Not built yet.

[Paul Marossy]

http://www.aronnelson.com/gallery/main.php/v/gus/wha1.gif.html?g2_imageViewsIndex=1

Q2 is bootstrapped for a controlled operating point and min loading of the wha pot and the use of a 8.2K emitter resistor, also this was done to not use the 470K to base for bias and to remove the mixing it causes( like the boomerang)
Q1 is setup as like the colorsound with the grounded emitter.  Note the three rough calculations assuming .6VDC B to E the collector will be about(((.6VDC/100k) x 470K)+.6VDC) or (((.6VDC/82K) x 470K)+.6VDC) or ... from ground.
I like the voltage diver bias for Q1 better than the stock boomerang bias
I would try higher Hfe for Q2 and different Hfes(>200) for Q1 and a 22K Q1 collector resistor to start.  Maybe a 100K in the voltage divider to start

One can adjust the bias of both transistors independently

One of many wha variations I have been drawing.  this was because Paul liked both the colorsound and boomerang and I think this might be the parts of each that are a little different than the crybaby.

Have fun

Gus

Not built yet.

Interesting idea. Planning on building it anytime soon?

[Gus]

I hope to change add and change  a few parts in a Brownsville wha I bought some time ago, it looks to be a crybaby like circuit.

I left some hints for bias of Q1 in the drawing.  One thing I don't like about the crybaby circuit is that the Q2 EF is biased from the collector of Q1, this causes the emitter of Q2 to be .6VDC or so less than Q1.  For better Q2 drive you want Q2 emitter to be higher than 1/2 the supply voltage.  One might want to adjust the collector resistor for the static Ic (collector current) for the load Q1 is driving, lower resistance value for better current drive.

The schematic allows one to set the operating points of both transistors independent of each other.  Maybe you want Q1s collector at 4VDC or maybe 2.5VDC or maybe? and Q2 emitter at 4.8VDC.  To adjust Q2 change the ration of the 100K and 150K voltage divider it is now set at 3/5 of the supply for the base voltage.

For a stock crybaby like circuit maybe cap bypass the emitter resistor of Q1 and use a 100K series input resistor for the 68K.  Another thing to reduce the mixing maybe a high hfe Q2 and >470K mixing/Q2 bias resistor.  The mix/bias resistor value is too big when you get a larger voltage drop across it

Have fun

[Gus]

Did anyone else look at the schematic?
  Yes it a hand drawn download it and view it with a picture and fax or something that lets you see it full screen and rotate

   I post a different take on a wha and nothing?

Even some simple math 1st approximation for the Q1 bias.  Same kind of DC bias as a BMP in the BMP there is AC feedback in the wha the AC is shunted to ground via the cap.
First you figure out the base to ground voltage. Take that voltage and calculate the current in the base to ground resistor.  Using that current calculate the voltage drop across the collector to base resistor.  Add the two voltages and you should be close to the collector to ground voltage(if the hfe is high enough).

[ayayay!]

The voltage at the node of the 470K, 6uf on up electro, inductor and 82K and the series resistance from that node to the base is what is setting the bias for BOTH transistors.  SO adjust the voltage divider for the TWO TRANSISTOR OPERATING POINTS.  I would adjust the 82K because it is bypassed but the electro cap and is at a DC voltage but an "AC ground".

So for a crybaby build I would use >200 hfe and adjust the one 82K resistor value to bias the two transistors to taste.  I would use a high hfe for Q2.

Gus, can you help me wrap my head around this?  I understand that the entire network ([470k, electro cap, inductor, 82k] plus series resistance to that node) sets the bias for both.  It's the tiny part inside that network (the 82k) that throws me.  Yourself and R.G. say this is a secondary bias resistor for Q1.  While I take your word(s) for it, what makes it the bias resistor, and not, say, the... wait a minute.  I just figured it out.  It's the one going to ground,... duh. 

So in that case, what effect does adjusting the 82k have on the main biasing resistor (470k) for Q2?  Is it gonna be OK to still leave that one alone?  I'd guess yes.  Sorry, just kinda thinking out loud here... 

[Gus]

http://www.aronnelson.com/gallery/main.php/v/gus/wha1.gif.html?g2_imageViewsIndex=1

  Note the three rough calculations assuming .6VDC B to E the collector will be about(((.6VDC/100k) x 470K)+.6VDC) or (((.6VDC/82K) x 470K)+.6VDC) or ... from ground.

Lets take the colorsound for simplicity 100K.  Grounded emitter like a FF Q1 so the base will be about .6VDC above ground.  Lets ignore the 1.5K and resistance of the inductor from the node to the base because they are in series and a low total value.  Think DC crude simple 1st approximation of Q1 operating points.  .6VDC base to ground and 100K resistor base to ground so .6VDC/100K+ .006ma.  Both the 100K and 470K are in series so the current will be about the same(ignoring base current)  so .006ma X 470K =2.82VDC.  Add both voltages together 2.82VDC and .6VDC =3.42VDC at the collector.  This is an approximation ignoring base current, less base current with higher hfe transistors.  IIRC Paul posted around 3.4 at Q1 collector in his colorsound build.

Now make the 100K a 82K you will have more current in both resistors and a greater voltage drop across the 470K so the collector will be at a higher voltage.  Note the crybaby has a 82K this might be for the addition of the emitter resistor.  The cap across the resistor is to remove AC feedback from the collector of Q1 by shunting it to ground.

Shifting operating points of transistor can change the sound, gain and output signal symmetry , rangemasters types are adjusted for taste FF types for taste etc.  Why not adjust the whas Q1?

I would QUESS the 100k/ 82K was adjusted by the original wha designers because the 470K is loading the collector of Q1 the other end of the resistor is at "AC ground" because of the cap across the 100K/82K resistor so maybe pick a value that fit their design rules and adjust the one that changes the loading the least when using different emitter gain control resistor values at Q1.

Is this helpful?

EDIT look at page 83 of The Art Of Electronics 2nd ed. for Q1 distortion with a grounded emitter

Also most of this is in R.G.s writeup.

[Paul Marossy]

I post a different take on a wha and nothing?

If I wasn't working three different jobs and also being Mr. Mom, I might have some time to mess with it...

[Gus]

Paul you posted it looked interesting.  Your post got me thinking about whas again.  I think the grounded emitter Q1 could be a cool change to the stock crybaby with a few other adjustments to the circuit.

[Paul Marossy]

Paul you posted it looked interesting.  Your post got me thinking about whas again.  I think the grounded emitter Q1 could be a cool change to the stock crybaby with a few other adjustments to the circuit.

Yeah, that might be a fun mod to try on a CryBaby...