oscillation with boss builds

[Surgeon]

Hi guys,

2 of my 3 boss builds (BD-2 and HM-2) have this high-pitch "over-note" when I play. My OD-1 doesn'T but these 2 do. I don't know if it happens to some other guys out there but when I get close to max gain and a good amount of treble (not dimed though) it happens. Kind of annoying as well. Both my builds are exactly as per the layout and work "well" otherwise.

I'M wondering if you have any suggestions to fix this (if some of you encountered it).

Thanks!

[Frank_NH]

Look at JaviCAP's thread.  Try placing a buffer (or buffered pedal like a BOSS) in front of your effect and see if the squealing goes away.  I'd be interested to know if this works.

[Surgeon]

Thanks Frank, I guess I missed Javi's thread... I'll give this a try and let you know.

[Surgeon]

Well I'll be damned... if I put the buffered pedal before (after doesn't work as would be expected) none of the pedals squeal... wow. For the record, no high-gainer ever squealed (Dr. boogey, triwreck, tightmetal, etc...) but these two did. I'll report back in Javi's thread too...

[Frank_NH]

Hmmm...Now I'll have to break out my squealing OCD build again!  Maybe a buffer will do it good too.  By the way, if you want to make a buffer to tack onto your circuit, check out some of Jack Orman's circuits...

Basic Buffers

I can see making an op amp buffer on a small vero board and wiring it in front of the circuit input...maybe IvIark can comment on this.

[IvIark]

When we do layouts for Boss effects we always remove the bypass buffer and so it would make sense I suppose that these circuit would be optimised for having that in place even if it's something that we don't necessarily want to use for our bypass method.  

But it means it could make sense when you do have instances of a build not reacting as intended to include a buffer.  I'd add a Klon buffer or the JHS Little Black Buffer basically because they are both simple generic text book buffers that you might see in IC datasheets.  If you wanted to keep things smaller you could add a BJT or JFET buffer but personally I'd stick with the IC for its lower output impedance.  I could add a buffer post to the blog if people would like to see it and include the smallest possible layouts I can for IC, BJT and JFET buffers so you have a choice.  Let me know if you'd like to see it.

[JaviCAP]

Yeah, Mark, it would be great :) Anyway, I used a Klon Buffer board with the Dominator, as is maybe the most transparent and the opamp assures a really low output impedance.

If you don use a socket for the IC, and use too multilayer caps + a mini electro for the 47, the size is really small.

J.

[Surgeon]

I have to agree that a buffer post witht the smallest layouts possible would be great!

[IvIark]

I've added a few choices to the blog

[Surgeon]

You da môn! Thanks!

[Frank_NH]

Quick note.  I built the op amp and jfet buffers but they didn't kill the squeal in my OCD build.  Nonetheless, I can see experimenting with them at the input and output of some of my effects.

The TU-2 schematic can be found here:

TU-2 Service Manual

Note that the bypass buffer consists of a jfet buffer at the input and a BJT buffer at the output.  They both look pretty simple and it may be worth building this buffer just for fun.  However, I wonder if the Klon buffer would be just as good.

[Kinski]

Its weird, I built a boss OC-2 with a one knob fuzz after it in one box. Squeals when i turn the fuzz on, unless i turn the clean and sub down below unity. When I ran any buffered pedal before it, the squeal went away. So, I put a cornish buffer in the box, with the oc-2 and fuzz, but did not help. I don't get it.

[JaviCAP]

Klon BUffer is the one i use to get rid of the squeals, and the one I integrated at the input of the Dominator, so I was quite sure that the first one would work too, but it seems that it's not so... Please, try it with the Klon Buffer, and if it works, we'll be closer to find the root cause of the probem

So I've been checking the differences between Buffer 1 and Klon Buffer: They're very close, so obviously, the Klon is taken from the first 1 of the post.

Differences: In the Klon, the voltage divisor is made with 2x 100k resistors, on Buffer 1, with 2x 10k.
On Klon buffer, we find, besides the 1M resistor from input to Vbias, a 47uf from too.

After IC output, the Buffer 1 goes directly to a 10uf cap, and it's done. The Klon goes to a 1uf cap, with a 100k resistor to ground and a final 560r resistor to output.

I know that the voletage divididers resistors value, affect the input impedance but I can't understand why the Klon Buffer stops squealing, and the Buffer 1, cannot.

I'll think about it

[Kinski]

Cool, I'll try the klon buffer and report back.

[Frank_NH]

Hey JaviCAP - you and I have been thinking along the same lines.

Here's the Klon buffer:



In my experiments, one thing I found that would kill the squeals is to put a low pass filter before the input of the effect, essentially to remove the high frequencies which appear to feedback and squeal.  The only low pass filter in the Klon is the 47 uF cap in the bias voltage network.  Would that make a difference?

Another thing I found was that the BOSS TU-2 input buffer has a 10 uF electro cap connected from the output of the op amp (before the series 1 uF cap in the Klon buffer) to the 4.5 volt bias with a 100k resistor between the - side of the cap and the 4.5 volt bias.  This would act as a low pass filter, but I'm not sure what the effective roll off frequency would be.  I may try that trick with the Orman op amp buffer.

[Frank_NH]

I'm going to on vacation for a week, so I won't be able to do any more tests for a while.

Would someone be up for simulating the Orman and Klon buffers in LTspice??

[induction]

Frank_NH wrote

The only low pass filter in the Klon is the 47 uF cap in the bias voltage network.  Would that make a difference?

Doubt it. That's cap is used for 'noiseless' biasing. A common type of biasing would be two 2M resistors used in the biasing divider, which connects directly to the non-inverting input. That method is noisier because it allows voltage ripple from the power supply and thermal noise from the biasing resistors into the op-amp input. Noiseless biasing is quieter because it shunts the power supply noise to ground through the 47u cap, and the biasing current goes through smaller resistors (I would use 10k, not 100k, personally), so there's less thermal noise, and that noise also gets shunted to ground through the cap anyway. The 1M resistor between the non-inverting input and vBias maintains the same input impedance of the other method. In other words, the 1M resistor is there to prevent the signal from dumping high frequency content through the 47u before hitting the op-amp input. So that cap is part of a low-pass filter for vBias, not for the input signal, which actually sees a high-pass filter (with a very low corner frequency) at the non-inverting input.

I can't comment on the Boss buffer because I haven't seen the schematic, but it might be another example of noiseless biasing.

[JaviCAP]

So, the only difference left, is the 100k resistor from output cap to ground. NO really sure about of the role that this resistor play in the buffer

[Kinski]

So my memory was wrong. After opening the pedal up, I realized I used the KLON buffer, not the Cornish.

[Kinski]

I put the buffer after the 3pdt so the buffer is bypassed as well when the pedal is switched off. Is that where you place the buffers, Javi?