Troubleshooting Electra-esque distortion

[GreenTeaZ]

So last night I put together a simple transistor based distortion. Now, the funny thing is, it picks up a radio signal. (And not even the good stuff!) Now, my reasoning has led me to believe that the dipped tantalum cap I'm using is wired in wrong. This is supposed to be soldered to the ends of the positive & negative wires, correct?

[mth5044]

Do you have a schematic or layout or anything?

EDIT: just saw the title! This is your basic electra. Where did you put the cap?



Also, is it boxed up or are you testing on breadboard?

[GreenTeaZ]

The cap I describe is in place of the electrolytic cap.  And this is when it's boxed up.

[GibsonGM]

What electrolytic cap?  The 2 on the schematic are input and output caps, are not *usually* going to be found in electrolytic (rarely), and are in series with your signal, not across the power supply...is there a different drawing you're working from?

If it works well otherwise, and you get radio on it, that indicates RF, and is often solved by trying a very small value cap across the input to ground...something like 47 to 100 pF.    Let's make sure we're on the same page first, tho

[GreenTeaZ]

What electrolytic cap?  The 2 on the schematic are input and output caps, are not *usually* going to be found in electrolytic (rarely), and are in series with your signal, not across the power supply...is there a different drawing you're working from?

If it works well otherwise, and you get radio on it, that indicates RF, and is often solved by trying a very small value cap across the input to ground...something like 47 to 100 pF.    Let's make sure we're on the same page first, tho

The cap you refer to in your solution is the one I'm talking about. It's a 22mf dipped tantalum cap I'm using.
It's the little yellow bump in that picture. Should it be wired to the very ends of the positive and negative wires on the sides of the board?

[GibsonGM]

If you're working from the Beavis schematic above, it's not connected to anything, Teaz...it's not part of the circuit!  

9V+ would come in at the 4.7K resistor, the 'wire' of which I think I see in your pic to the left of that blue one.   The leg you can see of the transistor is the collector, if that's a 2N3904...the 4.7K has 9V on one end, and goes to the collector.  That's your only power connection.   If you're troubleshooting, power it with a battery.  Still get RF?   Then it's not the power supply (probably isn't, anyway, but that removes PS from the issue)

If you're trying to put that cap across the input to snub RF, it would go from input to ground.   You could just put it across your input jack, really, altho many place it from base to ground (middle transistor leg) on the board...BUT - 22mF is WAY too high a value, it will kill your high frequencies (it's so big that its effect would be audible).    You'd want a small ceramic cap, 47pF-100pF or so.  They are all over the place, even Radio Shack has packs of them.  You can hold it there to see what happens, tho - you'll get rid of the radio for sure!!  And learn something, too (won't harm anything if you do it at the jack and don't short anything).

In case your thinking was that you should place it from 9V+ to gnd...yes, it's OK to do that for 'power supply filtering' provided the cap is rated higher than what you'll power it with, and it would go ideally from where the "+" hits the circuit board to a convenient (nearby) ground.    This is a reasonable thing to do for powering with 'wall warts' etc, as they are not always that great at regulation.    Limited usefulness for battery powered stuff, and not really a major factor in RF problems.      Wires run wrong and poor shielding are usually at fault for that.


Often, circuits like this pick up interference until they're boxed *properly*....that means the box gets grounded by the input jack (at least, most also have output jack connected to gnd).   If it's open, not grounded and no back cover, still could be susceptible.  You SURE you got it in there right?
Does it sound OK when you play, other than the RF issue?

[GreenTeaZ]

If you're working from the Beavis schematic above, it's not connected to anything, Teaz...it's not part of the circuit!  

9V+ would come in at the 4.7K resistor, the 'wire' of which I think I see in your pic to the left of that blue one.   The leg you can see of the transistor is the collector, if that's a 2N3904...the 4.7K has 9V on one end, and goes to the collector.  That's your only power connection.   If you're troubleshooting, power it with a battery.  Still get RF?   Then it's not the power supply (probably isn't, anyway, but that removes PS from the issue)

If you're trying to put that cap across the input to snub RF, it would go from input to ground.   You could just put it across your input jack, really, altho many place it from base to ground (middle transistor leg) on the board...BUT - 22mF is WAY too high a value, it will kill your high frequencies (it's so big that its effect would be audible).    You'd want a small ceramic cap, 47pF-100pF or so.  They are all over the place, even Radio Shack has packs of them.  You can hold it there to see what happens, tho - you'll get rid of the radio for sure!!  And learn something, too (won't harm anything if you do it at the jack and don't short anything).

In case your thinking was that you should place it from 9V+ to gnd...yes, it's OK to do that for 'power supply filtering' provided the cap is rated higher than what you'll power it with, and it would go ideally from where the "+" hits the circuit board to a convenient (nearby) ground.    This is a reasonable thing to do for powering with 'wall warts' etc, as they are not always that great at regulation.    Limited usefulness for battery powered stuff, and not really a major factor in RF problems.      Wires run wrong and poor shielding are usually at fault for that.


Often, circuits like this pick up interference until they're boxed *properly*....that means the box gets grounded by the input jack (at least, most also have output jack connected to gnd).   If it's open, not grounded and no back cover, still could be susceptible.  You SURE you got it in there right?
Does it sound OK when you play, other than the RF issue?

At this point it sounds delightful besides the radio. Okay, so I'm getting that the cap should go from the input to the ground to get rid of the radio, correct?


EDIT: And I need to ground the box itself? How would I go about that?

[mth5044]

If you are using open frame jacks (not the black, bulky ones), and you didn't paint your jack holes ( :lol: ), the enclosure will ground through the jacks. You can test by using your multimeter by measuring the resistance between the sleeve and the enclosure, should be very small.

[GibsonGM]

^  Yes, but they also need the wire from gnd to the jack, lol   Exactly, check for continuity.       Yes, cap from input > gnd will shunt RF ('very high') frequencies to ground.    But be sure it's a small cap....try 47p, 100p....you should have a bunch of them for this hobby, they come in handy.   A little cap goes a long way in this situation.

One other good way to avoid this is to use shielded wire for your input line....ground the shield only at one end.  This may make the cap unnecessary.   In lieu of a shielded wire, you can try twisting your signal and ground wires leading to the input jack together.    Often, just a small 'tweak' will get rid of the RF thing.   Even the way wires are run can allow strange junk to couple in.    We want to get it tweaked enough so it won't come back   

"* DID YOU KNOW?  Caps "like" high frequencies...they present a low AC resistance, known as Reactance, to them.  They DON'T like lower frequencies...as the F gets lower, their reactance increases, helping reject the passage/transfer of these lower F's.    RF carriers are way high frequencies...as long as we shunt F's above 20kHz to ground, we won't hear the difference.  Those small pF value caps, 10s to a few 100s, do just this - make no audible difference to us, but lower the bandwidth enough to get rid of what we don't want.   They present low reactance to high F's, allowing them to pass to ground, but prevent us from losing our signal down the same path. "