Boneyard shenannigans -SOLVED!!!!

[Willybomb]

Well, I didn't think it'd happen to me...  Just rocked and boxed a boneyard, and got the dreaded squeal.

This is a v1 haberdasher etch.  I rocked it, and was getting squeal on the boost and volume pots, and also on the crunch/hi-gain pots.  I put a klone buffer inline and the squeal left, or so I thought.  I boxed it up, and while there's no squeal on the boost or volume any more, there's some at the top end of "crunch" pot but goes if I back it off a touch, but the hi-gain pot squeals the whole time.

Any ideas?  Putting a cap across pins 1 and 9 of the chargepump didn't help.  I'm running off a battery if that helps.

[Willybomb]

Just tried it with a DS-1 in front of the boneyard - no squeal on any setting.  The DS-1 is just in the chain and powered from the same source...

*edit - just swapped the DS-1 for a Crossfire tuner pedal.  Still some squeal on the hi-gain, none on the crunch.

[mgwhit]

That was kinda what happened to me when the Boneyard squeal issue originally came up.  I breadboarded it and then built a PCB, but apparently had a Boss TU-2 inline when I rocked it.  (I can't remember if I had the tuner inline with the breadboard, which complicates things.)  I eventually built a buffer on PCB, added it to my box, and, yes, found that it still squeals a bit on extreme settings....  I would be more upset, but I only ever use it with the Boss tuner in front of it.

I didn't really think that the V1 etched board was susceptible to the dreaded squeal.  I certainly don't remember anyone talking about it until the first fabbed board came out.  I would recommend looking for spots on the board where the input traces and components run physically near the high gain op-amp stage.  The only spot I see is where C2 come close to D3.  Maybe reorienting C2 so it connects directly to the input pad might help.  Good luck!

[Willybomb]

Thanks, I'll have a look. 

So, what is it about the Boss buffers (presumably) that make the difference here? I've got a klon buffer in there already to some (no) avail.

I also suspect that the charge pump IC may have something to do with it, so I'm going to try some others.  Currently I have a TC1044SCPA in there, so I'm going to try a ICL7660CPAZ and a MAX1044CPA to see if either of those help at all.

[mgwhit]

So, what is it about the Boss buffers (presumably) that make the difference here? I've got a klon buffer in there already to some (no) avail.

I wish I knew.  Technically, buffers should only do three things (only one of which is typically desireable): provide a high input impedance and a low output impedance,  lower the gain by a minute amount (less than 1%), lose some highs and lows depending on the design and capacitance of the circuit.  I don't think it would be going out on a limb to suspect that the Klon buffer lets more highs through than the Boss buffer, but I don't really know.

I also suspect that the charge pump IC may have something to do with it, so I'm going to try some others.  Currently I have a TC1044SCPA in there, so I'm going to try a ICL7660CPAZ and a MAX1044CPA to see if either of those help at all.

You want an "SCPA" chip in there.  Those are the ones that don't whine.  And there's a difference between the whine and the squeal.  Charge pump whine is just a constant high pitch.

[Willybomb]

Yeah, it's a constant high pitch.  The actual pitch does drop when I turn the high-gain down, but probably only a semitone or two.  The sound is a constant "eeeeeeeeeeeeeeeeeeeeeeeeeeeeeeeeee" noise if that helps.

[Willybomb]

Well, tried a TL074 and a TL084, still got the whine/squeal.  Anyone see any issues here? I'm thinking i need some power filtering.

[wgc]

try changing the crunch pot (I think) per bean's build doc v1

also, this is great stuff:

http://e2e.ti.com/blogs_/archives/b/thesignal/archive/2012/05/30/taming-the-oscillating-op-amp.aspx

http://e2e.ti.com/cfs-file.ashx/__key/telligent-evolution-components-attachments/01-864-00-00-00-66-32-73/Solving_5F00_Op_2D00_Amp_5F00_Stability_5F00_Wells_5F00_6_2D00_5_2D00_12-_2800_2_2900_.pdf

[Willybomb]

Thanks, but I'm already using the V1 modified pot values, 50k and 250k.

Someone mentioned a cap across pins 1-8 on the charge pump.  I might have a play with that too.

[Willybomb]

Ok, so someone posted that they'd fixed it by putting a 1k across the input and ground, but it resulted in a loss of volume.  I tried it and... it worked, but the dirt was nearly non existent.  Crunch was... bluesy on max, and hi-gain was about not quite where crunch would have normally finished.

I figured too much signal was getting shunted to ground, so I put a 1meg there instead, and the whine was back.  I put a 22k in, and things are good, although I think the gain is still down a significant/little amount, a bit less than what I would like.  So anyway, now that I know we have a solution, I think I'll put a pot in there and test to see what the max R is I can put in so I can get maximum gain and still not get the whine.

BTW, I feel it's a little on the dark side, what can I do to brighten it up?

- EDIT: Can't get a good read off the pot with my multimeter - dead battery I think - so I worked my way up - 22k (good), 33k (good), 47k (good), 270k (bad), 150k (seems close enough).  I'll replace the battery and see if I can get a better exact reading, but 150k seems pretty close.

[Willybomb]

Have I mentioned how freaking happy I am to get this fixed?!! 

[alanp]

Always a good feeling, that

(and this kind of carry on is why I'm hugely leery of bothering at all with charge pumps these days, beyond FX where I really, really want to build them.)

[Willybomb]

Yep, I've been walking and driving around today just going "F**K YEAH! BONEYARD! MOTHERF**KER!! F**K YEAH!"

The multimeter measured the pot as 167k, so 150k is about as good as I'm going to bother with..... although... I wonder if a trimpot wouldn't work just a touch more efficiently.

[wgc]

Man, that's awesome!  I know just the feelings from my blacker forest build. Both when it was squealing and when I got it sussed.

I'm not sure yours was charge pump whine though. Mine was mismatched pots to the tone stack caps.

But in both instances, there is at least one thing in common, I think:  MB's std 3pdt wiring.

I love this wiring, but if you step back and think about isolating input/output, what do you see?  They're right next to each other ON THE SWITCH! 

I like this wiring so much, I used it for board mounting my switch. So I'm not knocking it. 

But I think THAT pinout made my mismatched pots less willing to play along nicely. If I put my thumb onto the switch contacts the squeal went away.

My next high gainer will probably have gauss m's wiring.

http://gaussmarkov.net/wordpress/thoughts/wiring-up-a-1590b/

Unfortunately, I've already got a different high gainer circuit ordered, with my suspected squeal inducing 3pdt wiring included. 

[Willybomb]

I usually use Gauss' wiring, but didn't this time... not sold on the idea that it's the cause of the whine though.  If I could be bothered I might rewire it, but it's working right now so who knows.  I'd still like to get the design's maximum gain out of it, can't be doing that if I'm shunting some signal to ground before it gets to the effect.