Hot Rod Ruby....almost there....oscillation problem...

[momo]

So I decided to go with a 3 circuit Ruby, if it was not for the oscillation, it sound awsome.
I made a Ruby with the bassman mods, added a hearthrob tremolo and an Jack Orman's AMZ mosfet boost..
I'm getting a nice fat tone and great volume as well as great tremolo......
the only problem now and I know it can be fixed is that its all oscillating.
the booster was biased at 5v using 9v supply on the money.
I tried biasing a bit lower but it did not help.
As you see, the wires are quite long everywhere right now as I was just putting everything together to test.
I will shorten everything and it will be short wires as the 3 circuits will be one beside each other.
Star grounding to the battery.

SO
Is star grounding best for this, or should I just daisy chain the ground?
I saw this explanation, would this do the job? ; http://www.diystompboxes.com/smfforum/index.php?topic=105609.msg951848#msg951848
I am aware that this is for a negative ground circuit, so I would flip the caps.
Also I'm reading R.G's response about all this and trying to figure it out!
http://www.diystompboxes.com/smfforum/index.php?topic=106524.0

See how it was when I tested, signal flow goes from right to left: Tremolo, booster, ruby.
thanks for reading,

[PRR]

I would expect long hay-wires around a high-gain amplifier to oscillate.

I bet if you touch or move wires, the oscillation changes.

Re-try with short wires, shielded for the audio wires.

[momo]

OK, thanks for that.
At least I know the circuits are fine, so I will wire the thing as final setup and hope its all good.
Will use sheilded wire for audio running the ground from circuit to circuit.
The amp box is wood and nothing is shielded in there, will see if I get radio interference or anything else related to that.
If so, I will try to put the whole circuit in some enclosure inside the box.
thanks.

[B Tremblay]

I have encountered instability with a Ruby sharing the same positive power rail with a booster. Try powering the booster separately or bypassing it with the transistor pulled.

[R.G.]

There is a reason all those old tube amps used a resistor between sections and an electrolytic cap to ground for every couple of tube stages. Coupling through the power supply could cause oscillation of all kinds.

DE...coupling with resistors and capacitors kept this from happening.

[momo]

You two are just on time..
I just finished wiring the 3 effects with short cables, sheilded audio and was getting to the power section....
I never decoupled the power supply for many effets together, I know its simple, but can anyone point me to some layout?

this is what I have so far and it an't be cleaner, the rest of the solution is in what you mention....

Voltage runs on top, grounding on bottom

[PRR]

>  it an't be cleaner

I dunno, I see sneak-back paths everywhere:

[momo]

thanks for your help Paul, so pardon my ignorance here, but I don't have a clue.
I never shielded wires to pots...
Are you saying I should ground all of those?
An open end at the pot and to circuit ground?

As I look at your arrows, you seem to indicate that voltages from one cable could induce voltage on another cable that is close by?
Sort of like by induction?, probably not the right word...
sorry, I don't know what to do for those!

[GibsonGM]

I think he is indicating that you have signal wires passing close to other signal and power wires, in parallel...not good.   Now, sometimes you can't help it, and so shielding them is the answer...use braided wire with only ONE END grounded; makes like a metal tube the signal wire rides in. 

Yes, you're correct about inductance...you have an AC signal that is "pulsating" (ok, not the best word), which can induce currents in nearby wires. That means power/ground wires, too, which will then couple this crap into OTHER audio sections in the same way.    When crossing wire runs, try to do so at a right angle...keep the wires tight and short as possible, shield the audio path....I'd bring the left and middle pot wires out straight, then bend them like an "L" once clear of the board....and if they're a problem, I'd shield THEM, too!    The chopstick will tell you if they are problematic (noise when moving with a nonconductor (chopstick)).    Only experiment will tell you when "enough is good enough".  High gain = picky....

If using a wooden box, you certainly can 'foil' the inside with aluminum foil that is grounded - this will make a Faraday cage to shield your board.  I do this inside guitars all the time, works great. 

[momo]

Great, thanks for the confirmation.
I have been studying on RC filter circuits now, read up on some great R.G articles and others.
This website is an amazing encyclopedia of info....high learning curve for sure!

Now I won't be using the RC circuit on the audio part so no filtering, I like the frequency response I have so far.
So as for power supply decoupling, I place the cap and resistor closer to the circuit if I read right as wires are resistors and can develop current, so if they are closer to circuit it will limit this effect?
Then I see that a 100nf cap is popular, but I have no clue on resistor value.

I now will just try the amp as is with rearanging the pot wires and shorter if possible.
I guess I could breadboard the RC circuit, but then it won't be close to my circuits....
So can anyone recommend cap and resistor value?
As I understand, it would be :  voltage wire----cap to ground----resistor, on to power the circuit?

At least after that I will have learned a bit more and won't have to ask around for that..

[tca]

I have encountered instability with a Ruby sharing the same positive power rail with a booster. Try powering the booster separately or bypassing it with the transistor pulled.

The LM386 is a strange beast. You should power the booster separately! There was/is a famous 1/2W amplifier that did exactly that, two 9V batteries one for the amp other for the booster.

[duck_arse]

sometimes the answer is not geofex, but amz .....

http://www.muzique.com/lab/filter.htm
and
http://www.muzique.com/lab/main.htm

[karbomusic]

The LM386 is a strange beast.

I was about to post the exact same thing. I essentially built a ruby but dropped back and designed my own buffer + opamp using only the spec sheets. I did it this way because I built the ruby or it's counterpart (don't remember which) and it had horrible oscillation issues. After dropping back and learning the circuit via spec sheets and designing my own from the ground up, the LM386 appears to be quite picky and careful addition of the right cap across the power rails and one or two other things of which I don't remember fixed the issue entirely.

Is it a high pitched whine oscillation or low 60 Hz sounding ?

[karbomusic]

Duplicate 

[tca]

... the LM386 appears to be quite picky and careful addition of the right cap across the power rails and one or two other things of which I don't remember fixed the issue entirely.

The relative position of the ground connections of the input/output, the LM386 ground and the bypass (pin 7) capacitor and also pin 3 (+ input).

[R.G.]

sometimes the answer is not geofex, but amz .....
http://www.muzique.com/lab/filter.htm

Introducing an inductor also means you then have to worry about resonances with the inductor, where they are and damping them down so the power supply system itself doesn't ring.

Sometimes the answer is general electronics knowledge - which as you know I'm a big fan of.   

[karbomusic]

The relative position of the ground connections of the input/output, the LM386 ground and the bypass (pin 7) capacitor and also pin 3 (+ input).

That sounds right. If I can find it I can post my schematic and layout. It used a 2N7000 input buffer if I remember correctly and it had an effects loop built in to it; getting those two to play nice took a little time but appears to be great now.

[momo]

wow, thanks for all the info....looks like I will have the monk lifestyle for the next few days, weeks!

I love it!
Yes its frustrating when things don't work, but boy, when you plug that guitar into some box with wires and things that you created and nice sounds come out....there is nothing better.......almost..hehe

I will get it done, I have put way too much time on this so far to let it go and sit in a box unfinished.
I realize now its a bit overkill for what it is, but its fun!

I have to get this done as I have lots of other circuits to do, including debugging a vibratone that is all populated but not working..

This is the amp, ain't she pretty!

[duck_arse]

I've been to that index page a million times looking for the power-line filter page, and the hum page sucks me in every time.

this:
http://www.muzique.com/lab/hum.htm

so even if the answer is amz, it sometimes still comes from geo. thanks rg.

[momo]

The LM386 is a strange beast.

I was about to post the exact same thing. I essentially built a ruby but dropped back and designed my own buffer + opamp using only the spec sheets. I did it this way because I built the ruby or it's counterpart (don't remember which) and it had horrible oscillation issues. After dropping back and learning the circuit via spec sheets and designing my own from the ground up, the LM386 appears to be quite picky and careful addition of the right cap across the power rails and one or two other things of which I don't remember fixed the issue entirely.

Is it a high pitched whine oscillation or low 60 Hz sounding ?

Hi pitched which changes frequency as I add or remove boost power..