Rockman X100
- Yazoo55
- Breadboard Brother
I am still trying to get a Rockman X100 I built a year or so ago fully working. I am getting too much compression despite adding a potentiometer to try to control it better. I have been looking at rockman.fr which is a mine of information on all the Rockman products. I think my problem relates to the 2n4339 FET transistor which needs to be properly biased. On the schematic, there are two resistors with guide values of 6.8 M and 5.1 M with a note that the specific values are factory selected to match the characteristics of the FET.
Does anybody know how to select the specific values?
Does anybody know how to select the specific values?
- darthoverdrive
- Breadboard Brother
While I don’t have the answer to your question perhaps this will help. This guy Jark built a compressor based on the Rockman’s . Here are some links to the post and articles. In one he chooses the resistors for the FET used. I hope your French is better than mine. Google translate helped but much of the articles made no sense.
http://techniguitare.com/forum/fiches-t ... t6058.html
The articles are here,
http://www.techniguitare.com/Fiches/Comp_JFET_v1.0.pdf
http://www.techniguitare.com/Fiches/Comp_JFET_v3.0.pdf
http://techniguitare.com/forum/fiches-t ... t6058.html
The articles are here,
http://www.techniguitare.com/Fiches/Comp_JFET_v1.0.pdf
http://www.techniguitare.com/Fiches/Comp_JFET_v3.0.pdf
The JFET used in the Rockman is a selected one based on screening at the factory, but I have never seen any documentation of what the criteria for accepting a particular JFET for use in a Rockman compressor circuit is.
Be aware that the Vpinchoff and Idss for a batch of JFETs will vary a LOT, and some part numbers are just selected parts from the same run based on their measured characteristic after they are put in the package (molded plastic, metal can, SMT, etc.). But even within even a single part number there is still a wide range of device characteristics which all get stamped with the same part number. This is why the X100 schematic has the "*selected" next to the JFET number on the schematic and parts list.
I suggest buying a few, measuring them (Vp in particular), and trying different ones in the circuit and listen to the results. RG or Jack has a diagram of a measuring jig on their site if you need additional information on how to do this. You don't need to buy or make something to do the measuring, just hook up the few components on a breadboard and use your DVM.
You can also use clip leads to try various bias resistor values, which I'm sure is exactly what they did at the factory to select them. If you have a 'scope (or a sound card and software oscilliscope program, play your guitar and look at the envelope signal amplitude excursions. The closer to 0 volts the peak of the waveform goes, the greater the gain as the signal fades.
You WILL have to spend some time tweaking it. You also may not like the way SRD was setting the compression on that particular circuit (X100) - taste is very much involved here. I heard ones I didn't care for as well as ones I did. But you can tweak it to YOUR taste in a long afternoon. And don't let JFET part numbers restrict you too much. PRR says there are only a few masks used to make JFETS, and doping is a pretty imprecise process for this kind of analog process. USUALLY, JFETs with higher Vpinchoff tend to be easier to tame for compression circuits.
And if the problem is just "too much compression", you can always just put a resistor between the drain of the JFET and the resistor & resistor-capacitor ladder to limit the maximum gain, although you may have to tweak the feedback resistor to make it sound the way you like.
mr coffee
Be aware that the Vpinchoff and Idss for a batch of JFETs will vary a LOT, and some part numbers are just selected parts from the same run based on their measured characteristic after they are put in the package (molded plastic, metal can, SMT, etc.). But even within even a single part number there is still a wide range of device characteristics which all get stamped with the same part number. This is why the X100 schematic has the "*selected" next to the JFET number on the schematic and parts list.
I suggest buying a few, measuring them (Vp in particular), and trying different ones in the circuit and listen to the results. RG or Jack has a diagram of a measuring jig on their site if you need additional information on how to do this. You don't need to buy or make something to do the measuring, just hook up the few components on a breadboard and use your DVM.
You can also use clip leads to try various bias resistor values, which I'm sure is exactly what they did at the factory to select them. If you have a 'scope (or a sound card and software oscilliscope program, play your guitar and look at the envelope signal amplitude excursions. The closer to 0 volts the peak of the waveform goes, the greater the gain as the signal fades.
You WILL have to spend some time tweaking it. You also may not like the way SRD was setting the compression on that particular circuit (X100) - taste is very much involved here. I heard ones I didn't care for as well as ones I did. But you can tweak it to YOUR taste in a long afternoon. And don't let JFET part numbers restrict you too much. PRR says there are only a few masks used to make JFETS, and doping is a pretty imprecise process for this kind of analog process. USUALLY, JFETs with higher Vpinchoff tend to be easier to tame for compression circuits.
And if the problem is just "too much compression", you can always just put a resistor between the drain of the JFET and the resistor & resistor-capacitor ladder to limit the maximum gain, although you may have to tweak the feedback resistor to make it sound the way you like.
mr coffee
- Yazoo55
- Breadboard Brother
Many thanks for your guidance. I bought a 2n4339 for this because I thought it was the correct part, but the specific transistor I have may not be suitable. Would it be worth trying J201s? I have a few of these in stock.
Glad to help - the X100 is an amazing piece of digitally-controlled analog circuitry, particularly in view of the time it was designed and built. I have heard a couple, and they can produce some really GREAT electric guitar tones.
Yazoo55 said:
Measure the pinchoff voltage for the 2N4339 you've got, and we can get some idea of how likely it is to be workable.
Yazoo55 said:
And please post the SCHEMATIC of how you have it hooked up now. It doesn't have to b e pretty.
I have the schematic for the SRD Sustainor, which I believe is the same or very similar to the X100, and the schematic for the original SRD Rockman, but neither of them use a potentiometer or a trimpot to adjust the compressor. You mentioned that you added a pot, but where in the circuit you added it isn't clear to me. It would be very helpful to be able to see what your EXACT circuit is, and it would be much easier to discuss here if we all were referring to the same schematic and parts' designations.
mr coffee
Yazoo55 said:
Nope. The Vpinchoff for the J201 is very low. Not suitable for voltage-variable resistor service. It's also a short channel device.Would it be worth trying J201s? I have a few of these in stock.
Measure the pinchoff voltage for the 2N4339 you've got, and we can get some idea of how likely it is to be workable.
Yazoo55 said:
Can you describe, or better, post a SOUND CLIP of the way the compressor sounds NOW?I am getting too much compression despite adding a potentiometer to try to control it better.
And please post the SCHEMATIC of how you have it hooked up now. It doesn't have to b e pretty.
I have the schematic for the SRD Sustainor, which I believe is the same or very similar to the X100, and the schematic for the original SRD Rockman, but neither of them use a potentiometer or a trimpot to adjust the compressor. You mentioned that you added a pot, but where in the circuit you added it isn't clear to me. It would be very helpful to be able to see what your EXACT circuit is, and it would be much easier to discuss here if we all were referring to the same schematic and parts' designations.
mr coffee
- Yazoo55
- Breadboard Brother
Thanks for all your help. I think I've solved the problems I was having. I've replaced the resistors R106 and R107 connected to Q101, the FET transistor, with a 10M trimpot set roughly to match. I have also removed the potentiometer I had added. I thought this would tame the over-compression but this was just wrong.
The Edge setting seemed to be to low in volume and insipid before. There is still a bit of a volume loss compared to the Distortion setting but it now sounds good.
I based my build on the schematics in this forum in the post "Rockman - Headphone Amp". I also used the images posted in the Diystompboxes forum under the topic "Some Rockman X100 Mods". There is a good quality image of the main board and of the copper side of the board. I used these to check the component values against the schematic. There are a couple of problems with the schematic "Rockman Rev 10.pdf". Some of the component values don't match the values I found on the board photo and there is also a mistake - R175 and R176 are shown to connect to R111 and C107. I'm pretty sure this is wrong. It does not match the photo of the copper side of the board.
Now to play some ZZ Top!
The Edge setting seemed to be to low in volume and insipid before. There is still a bit of a volume loss compared to the Distortion setting but it now sounds good.
I based my build on the schematics in this forum in the post "Rockman - Headphone Amp". I also used the images posted in the Diystompboxes forum under the topic "Some Rockman X100 Mods". There is a good quality image of the main board and of the copper side of the board. I used these to check the component values against the schematic. There are a couple of problems with the schematic "Rockman Rev 10.pdf". Some of the component values don't match the values I found on the board photo and there is also a mistake - R175 and R176 are shown to connect to R111 and C107. I'm pretty sure this is wrong. It does not match the photo of the copper side of the board.
Now to play some ZZ Top!
I solved this puzzle about a year ago.
I had access to data from 16 different X100.
And I did a SPICE analysis with the goal to study, correlate the above data to be finally able to predict the correct values for a wide range of new FETs.
That lead me to breaking the key to this safe.
I've found that a J201, 2N5457, or 2N4339 with a Vgs(Off) value of ~-1.00v (±0.05v) will work a treat when R18 is 6M2 and R17 is 5M1
Most of the time, R17 can be left at a value of 5M1 and use a 10M trimpot in place of R18. Then you should be able to dial in most any J201, 2N5457 or 2N4339 as long as their Vgs(Off) is between -0.85. and -1.15v
Most of the time, R17 can be left at a value of 5M1 and use a 10M trimpot in place of R18. Then you should be able to dial in most any J201, 2N5457 or 2N4339 as long as their Vgs(Off) is between -0.85. and -1.15v