Paranormal Compressor Schematic - Guitar Magazine

[iefes]

great stuff, thanks!

[rankot]

I believe pots shall be wired like this:

[Rob Strand]

The missing track definitely goes to the gate of Q1.

I think rankot's schematic is pretty close.

I also did a rough trace:

- I drew Q1 as a JFET but it's probably a MOSFET but in one of the pics I can see 2N ? ? ? ? so I thought maybe JFET.

- I got a few different values.

- My connections around Q4 and Q5 are different.   When I checked it, mine looked correct.
  You can see Q4 and Q5 have different bias networks on the bases. 
  [EDIT:  OK I can see rankot's connection is correct.  There's two bias networks (22k & 10k) but it's connected like rankot's.  At the diodes 22k goes to +ve and the *10k* goes to ground.]

- As drawn (either rankot's or my schematic) the 1ohm across the input to Q2 just looks wrong.
   I don't see how the circuit can ever put enough drive into the gate.
  I have this feeling the resistor values in the pics are wrong.  Maybe the guy that built it made
  some mistakes!    The rectifier looks weird to me.

- My guess is the Mix pot goes to the fixed side of the Squeeze pot.  It make more sense
  that the dry signal passes through without being affected by the Squeeze pot position.
  A few pics in the PDF show the wiring.

Anyway I can't spend anymore time on it today.

[rankot]

Rob, I believe both FETs are 2N7000, and it can be seen on one of the photos, indeed.

It would be nice if someone could pull out that schematic traced by Col as mentioned before in the thread.

[highwater]

There are two pads marked ground... I wonder if this was (incompletely, in the pictured form) adapted from a design for a bipolar supply.

I could be shooting above my pay-grade here (and it's definitely past my bedtime), but it doesn't seem like Q2 could do anything but conduct it's Idss with it's source grounded... unless Q4/5's tail were connected to a negative voltage.

If it's a MOSFET rather than a JFET... I haven't the slightest clue about Q2... but I still think this was a bad first attempt at a single-supply conversion.

[rankot]

I am pretty sure both FET transistors are 2N7000 because if you zoom page 3, picture 25, you can see the label. Also, AFAIK, only MOSFETs have gate in the middle, JFETs usually have gate as pin 1 (with excuse of Japanese JFETs which have gate in the middle too).

Rob, if you could share that ltSpice file, it would be of a great help

[FlyingWild]

Hi Rankot,

Yep, both FETS are 2N7000.

Here is a pic of mine, I would have been following the instructions that came with the kit when I made it, and have done no modifications save for putting in a 9V dc jack.

It looks like I had to put in a jumper and R14 was not installed.

Mine was working but something is now causing Q1 (FET 2N7000)  to fail, I've replaced it twice it works for a short while then no compression and the bypass tone gets the life sucked out of it.

[FlyingWild]

Q1: 2N7000
Q2: 2N7000
Q3: C549C
Q4: C549C
Q5: C549C
Q6: looks like: BC56()429 The silkscreen says PNP

C1: 47uF
C4: 2.2uF
C11:47uf

[rankot]

Interesting thing is that diodes are connected reversed (cathode to + sign on silkscreen), but it is the same as in the article.

PNP is probably BC560.

Try to measure voltages on Q1 pins.

[Rob Strand]

Rob, I believe both FETs are 2N7000, and it can be seen on one of the photos, indeed.

Ah! of course.  Thanks.  I did that before I went to bed and I didn't even think of 2N7000
(When I posted my schem I noticed you had posted one also.    For the thread your schematic can be the master copy.)

Rob, if you could share that ltSpice file, it would be of a great help

It's not LT Spice it's a *really* old spice.  If you want to try it I can send it.

PNP is probably BC560.

IMHO, it is definitely PNP.  It is marked on the PCB as "pnp" and the circuit doesn't makes more sense.

It looks like I had to put in a jumper and R14 was not installed.

That makes a lot more sense.  There's no way it would work with the 1ohm shown in the magazine!
I also noticed the 1k's in the magazine are 1k2 on your board.

Mine was working but something is now causing Q1 (FET 2N700)  to fail, I've replaced it twice it works for a short while then no compression and the bypass tone gets the life sucked out of it.

Maybe ESD (static electricity)?   I normally don't blame ESD unless there's some evidence like known zaps in that area or the building or very low humidity.

[anotherjim]

Interesting thing is that diodes are connected reversed (cathode to + sign on silkscreen), but it is the same as in the article.

That's a correct thing in some circles. If it was a psu rectifier, the + supply would be coming out of the cathode end. It can cause a lot of confusion, especially as it could also be applied to silkscreen for LED's.

[rankot]

I will check diode orientation and whole that rectification part and compare with Rob's schematic in the morning, now I'm just too tired. Whole day of drafting, then driving, then drafting again...

[rankot]

It's not LT Spice it's a *really* old spice.  If you want to try it I can send it.

Lets try!

That makes a lot more sense.  There's no way it would work with the 1ohm shown in the magazine!
I also noticed the 1k's in the magazine are 1k2 on your board.

Indeed, I've noticed that too!

Also some other resistor values, that I have read wrong (4k7 resistors on this board, while I thought they're 470 Ohms and few others). I will post corrections in the morning.

[rankot]

I also did a rough trace:

I don't see this R24 of your schematic on PCB! Cathodes of those two diodes are connected to 100k+100n (their other leads are connected to anodes of other diode pair), 22k to 9V and 10k to ground. I definitely don't see 10k going to Q4 base from that point. Are you sure there is a resistor there?

[Rob Strand]

Lets try!

OK.  There's nothing to lose.  If it works it would be really cool.  You will have to edit the transistor models names to suit your libraries (to be honest I only put any old junk in there to get the schematic done).   I don't know how it will convert the symbols.

Also some other resistor values, that I have read wrong (4k7 resistors on this board, while I thought they're 470 Ohms and few others). I will post corrections in the morning.

There was a few other places as well.  Like the 100k to ground on the rectifier.  2k7 not 2k4 on the source of the MOSFET.  I'm fairly certain the MIX pot should not connect to the wiper of the Squeeze pot.

[Rob Strand]

I don't see this R24 of your schematic on PCB! Cathodes of those two diodes are connected to 100k+100n (their other leads are connected to anodes of other diode pair), 22k to 9V and 10k to ground. I definitely don't see 10k going to Q4 base from that point. Are you sure there is a resistor there?

R24 is R14 on my schematic.   That divider is OK.

I reverse the values on the other divider, which doesn't help trying to find them:
Rob                       Rankot
R11 10k-wrong     R20 22k-correct
R12 22k-wrong     R21 100k-wrong           should be 10k

My resistor designators do not match the PCB.  I couldn't read them all from the pics.

[rankot]

I had in mind R24 on your schematic. 

[Rob Strand]

I had in mind R24 on your schematic

That will cause you some trouble  .

[Rob Strand]

That's a correct thing in some circles. If it was a psu rectifier, the + supply would be coming out of the cathode end. It can cause a lot of confusion, especially as it could also be applied to silkscreen for LED's.

Best not to use the '+' at all.   The diode *symbol* already has the polarity.   The '+' causes confusion.

[FlyingWild]

Interesting thing is that diodes are connected reversed (cathode to + sign on silkscreen), but it is the same as in the article.

PNP is probably BC560.

Try to measure voltages on Q1 pins.

You've reminded me that I contacted the place I got the kit regarding the diodes, and got the following response:

Hi Chris

Fit the diodes as in the photo. Some confusion here as to what the anode and
the cathode look like on a diode, I expect. Anyway the  photo orientation is
correct.

I'll measure the voltages tomorrow, my multimeter isn't to hand at the moment.