Author Topic: Palissade clone : trouble with FET clipping  (Read 280 times)

eh la bas ma

Palissade clone : trouble with FET clipping
« on: June 07, 2021, 03:47:08 PM »
Hello,

A friend from this forum gave me, very kindly, a pcb.
It's working, but I am not sure if the FET clipping is properly working :

Q4: D= 4.35V
       G= 4.50V
       S= 4.50V

Q5  D= 4.50V
       G= 4.35V
       S= 4.35V

Moreover, the 6th position does a big volume drop, without clipping. I am not sure what it should do on schematics : Is it every diodes + transistors alltogether ?

According to instructions p. 6, the person who designed it was unable to get FET transistors working.
I wonder if it is even possible.

I am using this pinout (2n7000) from my supplier :


https://www.musikding.de/2N7000G_1

I tested for continuity between rotary switch's pad 5 and both FET :

On Q5 there is continuity between pad 5 and Source, and on Q4 there is continuity on Drain and Gate.

I did the same test between both transistors : There is continuity between Q5 Source and Q4 Drain and gate, and between Q5 Gate and Q4 Source.







The schematics mention BS170. Based on a quick research, they have opposite pinout than 2N7000. So I soldered my 2N7000  accordingly, opposite from the pcb layout.

Maybe my transistors have a different pinout. It is written :

2N7000
S9 (or 59)

Any help would be welcome !

Edit : this is p.6



"use the ring of the switch to limit to 4 steps" : Is it possible to limit the rotary switch steps without opening it ? What does he mean by "use the ring" ?
« Last Edit: June 07, 2021, 05:51:22 PM by eh la bas ma »
"unity-gain non-inverting stage", "ground leg","post-pick moments", I love this colorful forum.

PRR

Re: Palissade clone : trouble with JFet clipping
« Reply #1 on: June 07, 2021, 05:31:56 PM »
> if the Jfet clipping ................(2n7000)....

2N7000 is NOT a JFET, even though Musikding thinks it is. It is a MOSFET.
  • SUPPORTER

eh la bas ma

Re: Palissade clone : trouble with JFet clipping
« Reply #2 on: June 07, 2021, 05:47:41 PM »
> if the Jfet clipping ................(2n7000)....

2N7000 is NOT a JFET, even though Musikding thinks it is. It is a MOSFET.

Oups... thanks. I'll correct my post and I'll just write FET.
"unity-gain non-inverting stage", "ground leg","post-pick moments", I love this colorful forum.

idy

Re: Palissade clone : trouble with FET clipping
« Reply #3 on: June 07, 2021, 06:43:28 PM »
The schematic you showed is puzzling because position 6 is a short circuit. If you are using this as diode clipping in a feedback loop, this will give you something like a buffer, no gain. If you are using them to ground after an amplifier stage, you will lose everything.

I think if you are using this setup, you could set the rotary switch up to only give 5 positions, problem solved.

Now I read the bottom of your post. The rotary switches we usually use come with a "ring" which is like a washer but it has a little tab. This can fit into one of 10 little holes. You select the number of positions you need, put the little tab in that hole, and either tape it in place until you put the switch in the box and tighten its nut, or if you have room inside the box (usually not) you put an extra nut on the ring to hold in place. Because these switches are taller than the common PCB pots we get, this second method is usually impossible. And the build doc is incorrect I think. There are 5 combinations available, not 4 or 6.

The schematic on the build doc does not show the short circuit on position 6.

Also the position with MOSFETs will sound the same as a pair of silicon diodes. They need extra diodes to "block" the body diodes they have inside that work the opposite direction of the MOSFET and turn on at lower voltage. Very strange, but some really good designers did this a long time ago and it got copied, but is a waste to transistors and doesn't allow you to hear what MOSFETs can do. I put the G and D together and add a diode  to the S pointing away (band pointing away.) Note that using a single MOSFET as drawn and leaving its neighbor empty will give you a nice assymetry: MOSFET in one direction, body diode in the other.

PRR

Re: Palissade clone : trouble with FET clipping
« Reply #4 on: June 07, 2021, 09:36:35 PM »
I think you want JFETs, not MOSFETs which, as IDY says, work so differently especially here that there is no relation.
  • SUPPORTER

eh la bas ma

Re: Palissade clone : trouble with FET clipping
« Reply #5 on: June 07, 2021, 09:59:42 PM »
I think you want JFETs, not MOSFETs which, as IDY says, work so differently especially here that there is no relation.

Would j201 be ok ?

I found this quote in an old thread :

"(...) using a pair of J201 JFET's as clippers, but his directions explicitly state that the drain and source should be tied together, with no mention of a series diode at all. I used this configuration in the Triangle v.1 Beaver I recently built, and it works very well, giving the pedal a noticeably smoother, warmer tone in that clipping mode."

https://www.diystompboxes.com/smfforum/index.php?topic=76337.0
"unity-gain non-inverting stage", "ground leg","post-pick moments", I love this colorful forum.

eh la bas ma

Re: Palissade clone : trouble with FET clipping
« Reply #6 on: June 07, 2021, 10:18:00 PM »
The schematic you showed is puzzling because position 6 is a short circuit. If you are using this as diode clipping in a feedback loop, this will give you something like a buffer, no gain. If you are using them to ground after an amplifier stage, you will lose everything.



About position 6, here is an other piece of schematics showing where the clipping section is located :

« Last Edit: June 07, 2021, 10:19:51 PM by eh la bas ma »
"unity-gain non-inverting stage", "ground leg","post-pick moments", I love this colorful forum.

idy

Re: Palissade clone : trouble with FET clipping
« Reply #7 on: June 07, 2021, 11:38:58 PM »
J201 is a JFET and could probably be used as a clipper. I think the Mad Professor Stone Grey 2n3819 JFETS. Not many people use them but you are welcome to experiment.

The schematic shows, assuming "1" is on top
1) Schotky diodes, very low clipping voltage, nice compressed.
2) Pair of Silicon. Works for the tube screamer. What's not to like. A little louder.
3) Assymetrical Silicon, 1vs 2. Again a little louder, one direction is clipped higher than the other.
4) Asymmetric Mosfet vs MOSFET+ LED. This is done wrong, I think. The side with the LED will never conduct. But the solo MOSFET will, in both directions. Then you get MOSFET vs diode, asymmetric and again a step louder than 3. So not a bad choice
5)Pair of LEDs. Loudest, a much loved sound ever since the "Rockman." Or Craig Anderton, or.....
6)No diodes. Now you have a clean boost until the opamp itself starts to clip. Again, you may like this. A reasonable and very loud setting.

Rob Strand

Re: Palissade clone : trouble with FET clipping
« Reply #8 on: June 08, 2021, 12:12:19 AM »
It's pretty normal to use MOSFETs as clippers.

It should work.

Here's an example of a straight clipper, see top connection.  However they work fine with this connection in a feedback set-up like a TS-9.

http://www.muzique.com/lab/zenmos.htm

There's two ways to connect MOSFETs clippers but when they are connected in parallel they should act like two parallel diodes connect back-to-back.    It is customary to connect the gate to the drain in these connections.

[clarified one detail here for testing with gate connected to drain]
If you take a  single MOSFET you can identify D and S by using the diode function on your multimeter.   It should measure about 0.6V with the meter "anode" is on source and the meter "cathode" is on the drain.    If you set the meter to a high resistance setting then swap the D and S connections from the diode case you should see a high resistance or open circuit.   If you do this with the gate connected to the drain it could be 1M or 2M but it could be open circuit and if it gives a low resistance you might suspect  a low VTO MOSFET.

If you are not confident of the pinouts it's best to do this with the gate open however it will work with the gate open or connected to the drain - open gate is safer for pin identification.   Finding the "diode" terminals will identify D and S.  There should be no continuity to the gate from the other two terminals.  If there is it is a sign that the MOSFET is blown.    If get a low impedance in both direction between D and S that's also an indication the MOSFET is blown, especially if the resistance is the same in both directions.

There's one case which should not apply to the 2N7000 or BS170 and that's a very low turn on/off voltage, VTO.   The case might turn-up with dodgy ebay parts.    You might get to this point if the previous tests don't work out.   In this case you *have to* have the gate open for the diode test.  In fact you might even try shorting the gate to the source.    If you discover you have very low VTO device then the only way you can get the MOSFET clipper to work is to short the gate to the source not the drain.  (FYI normally the diodes in back-to back MOSFET connection keep the MOSFETs off however if the MOSFET turn-off voltage is low the MOSFET will conduct before the diodes.)


Here's another test set-up, two separate tests shown on same diagram,




In the left side test, opening the gate or connecting it to ground should make the voltage rise to 9V.

EDIT:
I clarified one detail in the above because you expect slightly different results with gate connected to drain.
« Last Edit: June 09, 2021, 07:16:12 PM by Rob Strand »
The internet:  answers without the need for understanding.