Help a little with orange squeezer?

[aloupos]

I've built the orange squeezer using markm's layout from his layout gallery.  I used a small silicon diode while waiting for the germanium diode on order from smallbear.  Sorry I can't be more precise, but I ordered a value pack of diodes that had no labeling, and I really have no way to identify it. 

At this point, the circuit doesn't pass audio, and produces low frequency oscillations.  The rate of oscillation changes when adjusting the trim.   (nice noisemaker, but not what I wanted)

I'll do some further de-bugging, but before I do that, I'd like to know if the diode sub could be causing my problem.  I read another thread where this circuit was explained that said D1 was an important part of the circuit, and it needed to be Germanium.  Would a silicon diode cause the circuit to oscillate?

Appreciate any comments ..

Anthony

[96ecss]

This is from the info at GGG concerning the diode in the Orange Squeezer: "D1 is a 1N100 germanium diode. You can substitute other germanium diodes for the 1N100, but not silicon, it must be germanium unless you make some alterations to the circuit."

I don't know if it really makes that much difference but J D Sleep seems to think so. I've built a couple of these from his layouts and always used 1N34A Germs.

I hope this helps.

Dave

[Paul Perry (Frostwave)]

I havn't built it, but looking at the ciorcuit, havign the 'wrong' diode won't cause oscallation, in my opinion.
Maybe if the 4.7uF cap from the diode to ground is dried out, that could do it.
But the thing is, a LOT of people have built squeezers with success, so it's fair to say the basic circuit is OK.

[Mark Hammer]

With the Orange squeezer, if it isn't solder bridges on the board, most often it's the difference between the true FET pinout and what you thought the FET pinout was.

Essentially ANY diode will "work", including LEDs, given the simplicity of the rectification its doing.  The problem is that the diode will always subtract its own forward voltage from the resulting signal being fed to the FET's gate.  If it's a Ge diode, it will feed the slightly smoothed (via the 4.7uf cap to ground) version of one half cycle of the audio input minus roughly 230-250mv.  Moreover, the diode won't conduct until the signal reaches that minimum amplitude.  If it's a silicon diode, that minimum and minus gets bumped up to roughly 500mv, and if it's a red LED that would get bumped up again to around 1.5-1.7v.  IF the op-amp goosed the input signal enough to get past each of those respective thresholds, the circuit would work.  In the stock circuit, though, without changing any component values associated with the gain of the circuit, a GE diode will provide the most consistently pleasing and glitch-free performance (though maybe a schottky diode would be even better).

[aloupos]

Thanks for the responses so far.  I'll go over joints and look for bridges tonight. 

I'm using markm's layout http://aronnelson.com/gallery/album76/Orange_Squeezer_LAYOUT

Just as a sanity check, can you please verify that I understand this layout correctly.  I compare it against the schematic, and it looks like Q1 is Q2, and that the FET is oriented like this: 

[markm]

Comparing it to the GGG schem, I would say you are correct however, this was created from a different schematic.
Like I said though, looking at the GGG schem, yes Q1 is Q2 and Q2 is Q1.

[aloupos]

awesome, thanks mark!  did I get the FET orientation correct?  Do you have the schem you used for this layout so I can compare correctly? 

I'm also using a TL082 isntead of the JRC4558.  I'll socket the 4558 tonight after I pick up my smallbear order. 

MarkH, I did look up the pins of the FET (I've built about 10 boards before, and learned the hard way), but maybe different manufactures use different pinouts (for extra confusion)? 

[markm]

awesome, thanks mark!  did I get the FET orientation correct?  Do you have the schem you used for this layout so I can compare correctly? 

It's on a website that had a host of different schems but, right now I can't remember for the life of me where.....
You should be okay with using either the TP schem or GGG. I think the Tonepad schem may be a bit easier to read but, that just an opinion.
The orientation you have of the 2N5457's looks to be like every 5457 I've encountered but, your pinout could still be different however, what you have written there looks good to me.
I would search out a Ge diode though for better performance.

[aloupos]

I'll get the GE diode this eve, allong with the 4558. 

I indicated in your layout where I though G S and D where supposed to go, according to the schem.  Did I get it right? 

[markm]

I indicated in your layout where I though G S and D where supposed to go, according to the schem.  Did I get it right? 

Yes.
According to every 2N5457 I've seen.