Boomerang wah pot reversal

[joegagan]

here is the other one mick built (mick, which one was first?)

[Gus]

Q2 in the boomerang is just a simple base to + resistor bias that is why that matters


Q2 in the Dearmond is biased more predictably however if the hfe is low it might have a loading tapering effect on the wha potentiometer.
Lets do some math 810K||810K= 405K

 Then you have the( hfe x (the combination of 33K in || (with the 5.6K and .1uf cap in wha mode)))  in || with 405K

You could work this out for the transistors you like then rework the input bias resistors or just try a high hfe for Q2 and a 680K B to +9VDC and maybe a 810K B to ground

[Paul Marossy]

Q2 in the boomerang is just a simple base to + resistor bias that is why that matters

I remember someone once suggested to me that is part of what makes the Boomerang sound like it does, and it's because of some small amount of distortion created at Q2. Apparently that distortion grows to much larger levels with higher gain transistors. I can certainly hear it.

Anyway, the Boomerang to me has a more interesting sound than the Crybaby does. The Crybaby just seems bland and over focused to my ears. Crybabys work great in certain settings, but it sure ain't my go to wah pedal.

[joegagan]

gus- i see where you are going with that. thanks for reminding us that we need to do the math 

your suggestions helped when i built an 1800 dearmond clone ( this was after modding the first one you and i worked on last summer), i found the 1800 was more tunable than a boomerang due to Q2's biasing, like you said. i had trimmers all over the place. the 33k of the 1800, as compared to the  8.2 k to ground on q2 emitter- sounded best at the stock 33K. this was with a 100K hotpotz1 with no limiting resistors.   the 47k R parallel to inductor sounded best at under 25k in this config.
inductor had a an overall Mh of 700 ( put a second 100 mh inductor in series, i believe two inductors gives a richer tone, might just be superstition, ha). this required lowering the sweep cap to .0089 from .01 to compensate since the customer didn't like the low voicing of my original build. when you increase the Mh, you lower the center frequency of the sweep. this is easily remedied by lowering the sweep cap.

built  in a vox shell in standard CB rotation. the weird thing was, the wah was WAY louder than bypassed signal, when i tried to lower the volume at the emitter/GD R at Q1, th tone of the wah suffered, so my final solution was to add an external input vol control in place of the input resistor (100k  pot with 10k in series to ground so it wouldn't turn off when turned all the way down)

unfortunately, i did not take hfe readings of Q1 or Q2, i think i used 2n2222s or 2n3904s.
the wah quickly went to the customer in LA who has been using it on sessions and is very happy with it.

you can hear the final result here: ( first wah {left of 3} in line the one in the stock vox shell ). a good comparo to a standard stack of dimes crybaby ( far right of 3 ) in this vid.

[joegagan]

paul, i agree. the boomer is more interesting in the mids than a vox/cb. there are two comparos that corroborate this claim. brian temblay made a comparison of his 'std' wah to a noomerang  (boomerang clone from runofgroove), and the recent vid from mrbillyhill referenced above , all comparisons all show the same difference. the boomerang type has more interesting, pronounced mids.

3rd example, the comparo made in summer of 2010 filmed by zac zagar , a modded vox vs a stock 67 boomerang BG1

[Paul Marossy]

paul, i agree. the boomer is more interesting in the mids than a vox/cb.

Agreed! 

[Paul Marossy]

OK so tonight I put a 100K reverse log pot in my Boomerang wah clone in a Crybaby shell and changed the transistors to some metal can 2N2222s and the results are actually pretty nice. It has a very similar feel to the real Boomerang and it sweeps in a very similar way except that to my ears it sounds a little more lively than the real Boomerang does. Hmm....this pedal may go on my pedalboard now!

So I guess the OP is onto something. I never really noticed that the Crybaby and Boomerang pot mountings were opposite of eachother. There is something to that. (I know, duh!)

[Mick Bailey]

here is the other one mick built (mick, which one was first?)

The reversed pot one was the first. I was playing around with it before the reversal mod and decided to try it reverse operation (heel-down-treble) by reversing the pot leads. This made a big difference in sweep, but I didn't like the backwards operation of the wah. I then came across your clip showing the modified Dunlop pot action and rebuilt the pedal as per the original Boomerang.

I like the idea of a crank arrangement, though. The shape of the slot could be changed to give different sweep patterns. I'm looking for a milling machine at the moment - this would be ideal for such work.

[Paul Marossy]

The Colorsound wah has an interesting cam & slot arrangement which I have heard makes a linear pot act like a log pot. In any case it's kinda cool engineering.

[joegagan]

wow, paul, that is great news. glad you found a combo you liked. did you happen to measure the hfe of those 2n2222s?

mick, thanks for the clarification. you can imagine how it felt to stumble across mrbilly's video and then later that day find out i had something to do with the wah being built. very cool feeling indeed.

because i had a large stock of hotpotz1s, i was trying very hard to come up with a way to make them work in boomers. i tried all sorts of slotted lever arrangements , based on the colorsound rack and rod idea. never could get a sweep i liked that would not have involved an extra spring to pull the lever past the midpoint due to the curve i had to put in the slotted lever.  this was mostly due to the quick ramp taper of the hotpotz1, it is very hard to mechanically overcome.

the main problem i see with the colorsound arrangement is its limited travel across the pot. even 150 degrees of the pot's rotation is hard to achieve, whereas the gear  of crybaby type can get more than 165 if i recall correctly.

[Paul Marossy]

Joe: the Hfe on those 2N2222s is about 175

[joegagan]

cool, like that range.

[Paul Marossy]

cool, like that range.

Yeah in the Boomerang at least that seems to work pretty well. 

[Gus]

The following screen shot might help to understand why the transistors need to be selected for the Boomerang wha.

Using LT spice I picked two different transistors one with lower hfe and one with higher hfe.
Note Q1 collector voltages and Q2 emitter voltages

The biasing of the first stage Q1 is a Resistor C to B, R5 1.5meg in series with (L1 || R10 47K basically the resistance of the inductor) in series with the R6 4.7K
This type of biasing is a little more stable than the output biasing of Q2 however it is sensitive to R5 and the transistor used

The biasing of Q2 is a resistor B to +9VC sensitive to R2 resistor value and transistor used

Another transistor number to try would be a 2n4401

R8/R9 is the wha potentiometer more to the bass end.  This screenshot is to show more about the biasing of the transistors.

If you adjust R5 and R2 for the transistors you are changing other things.  If you adjust R2 are changing the loading of the potentiometer and the collector of Q1( more at max bass).

[joegagan]

looks great gus. this really  shows how much things change simply going to a higher gain trans. this explains a lot, ppl swap trans in and out and the whole dynamic would change. one note from past xperence- if the gains of Q1 and Q2 are not within 50 or so hfe of each other ( and no R bias adjustments made ) , the wah gets boring. seems these circs like a fairly nice balance of gains on both trans. makes sense when you consider the pot is a balance act between the two.

the 2n2222 is probably close to the stock hfe of boomer transistors-they usually hit between 170 to 250.

it is very interesting how sensitive the higher gain trans are, you made r2  and r5 4X larger to get the bias similar to lower gain trans. is it the interaction between the inductor/pot that makes this so wide a swing?  the swing is more extreme in Q1, i assume this is because of the parallel inductor circ ?

i don't think i have ever seen anyone suggesting that gains and biasing of Q1/Q2 can affect the taper of the pot before you brought this up.

just a reminder, the widely seen schems online have an error R2 is actually 820k, although i can see after looking at the bias voltages, the 200k difference between would not change bias much. i recall paul marossy got a better result when he removed the 620K and swapped in the correct 820K in his early boomer build.

one other thing i failed to mention as a dif between dearmond 1800 and boomer is that on Q1 for dearmond, there is a 100k bias resistor B to ground, which the boomer does not have.

[joegagan]

besides the 620K Q2 resistor error on many boomer schems ( should be 820k, some schems show this corrrection),

there is one other error on nearly every BG2 schem i have ever seen , including fuzz central's. it relates to the vol pedal function, so many people didn't even use this part, but the 56k right after the 'vol' switch is supposed to be 5.6K.

[Paul Marossy]

I wouldn't put 2N5089s in the Boomerang, it would sound distorted. I intuitively tried the 2N2222s and they work really great. I'm guessing they are close to what was used originally as far as Hfe range is concerned.

[joegagan]

i think what gus was illustrating is that by manipulating the bias, you can use higher hfe transistors, and that the boomer circuit is more adaptable than cb vox due to the independence of q1/q2 in boomers.

[Gus]

Paul Marossy

I wouldn't put 2N5089s in the Boomerang, it would sound distorted. I intuitively tried the 2N2222s and they work really great. I'm guessing they are close to what was used originally as far as Hfe range is concerned.

If that was a reply to the screenshot.  I was not posting to use 5089s.  I was trying to show how the operation points of the boomerang transistors are sensitive to Hfe.  In the blue notes you will see different collector and emitter voltages for the different transistors and with changed R2 and R5 values.

With stock resistor values a 2n2222 at Q1 sims at about 4.2VDC at the collector and a 2n2222 at Q2 sims at about 6.09 at the emitter with the wrong 620K value however it will sim a little lower with a 820K so both transistors are in a good range for headroom.

 MAYBE with an even lower Q2 transistor Hfe 620K would work better so maybe 620K might have been used in some of the builds this is just a guess.

Note with the higher Hfe transistor 2n5089 and the stock resistor values Q1 C sims at about 2.7VDC and Q2 E about 7.6VDC.

From the sim I would think you could change Q2 with maybe a small change in sound by adjusting R2 value.  Q2 is a emitter follower (EF) so it's "sound" is often more if it is biased for good input output voltage/current swing.

I am still thinking about what is going on around Q1 if you change R5 value to bias Q1.  R5, C5, L1, R10 node

A 2n4401 might be fun to try at Q1

[Paul Marossy]

If that was a reply to the screenshot.  I was not posting to use 5089s.  I was trying to show how the operation points of the boomerang transistors are sensitive to Hfe. 

I know, I was just sayin. 

That is an interesting point, though.