Friedman BE-100 JFET emu  [documentation]

Original effects with schematics, layouts and instructions, freely contributed by members or found in publications. Cannot be used for commercial purposes without the consent of the owners of the copyright.
User avatar
J0K3RX
Degoop Doctor
Information
Posts: 1057
Joined: 29 Jun 2011, 01:25
Location: US Florida
Has thanked: 315 times
Been thanked: 556 times

Post by J0K3RX »

caspercody wrote:The pedal still has a lot of treble to it, and I am not happy with the tone I am getting out of it. If I turn the mid control to 0, and turn the treble up to about 8 or 9 o'clock, then it sounds good. I really want to figure out a good tone stack to use with this, or any JFET clone. I do feel scaling it down I did loose bass response, meaning the bass control does not seem to have as much variance when you turn the pot.

I did change my guitar strings, and that seemed to get rid of the weird noise in the middle strings, and I changed the pick up from a SD 59 to a Dimarzio Super distortion. I think it was mostly the old strings.
What modifications did you make besides scaling the tonestack? Scaling the tonestack or not, either way you will need to change the value of R31 to 10k or 22k.. or whatever gives you better control/range of the tonestack and sounds good. You mentioned trying a Baxandall, you can try a Framus Cobra/Dragon tonestack.. On the lead and crunch channels it's like a James with a added mid control in front of it and on the Clean channel is like a mid control in front of a Fender Deluxe series tone stack which may be a good option to try also. The Cobra is one of my favorite metal amps of all time, has a unique tone and voicing.. Uber is also one of my favorite metal amps of all time and it has a unique deep growl to it.. Both amps don't sound alike but, they both have some magical shit goin on.

User avatar
roseblood11
Tube Twister
Information
Posts: 1886
Joined: 23 Aug 2008, 14:21
Has thanked: 366 times
Been thanked: 313 times

Post by roseblood11 »

Maybe try the tonestack of the Catalinbread Dirty little secret mk3? It has a very useful range.

User avatar
J0K3RX
Degoop Doctor
Information
Posts: 1057
Joined: 29 Jun 2011, 01:25
Location: US Florida
Has thanked: 315 times
Been thanked: 556 times

Post by J0K3RX »

roseblood11 wrote:Maybe try the tonestack of the Catalinbread Dirty little secret mk3? It has a very useful range.
Looks nearly the same as the BE100 :scratch: Straight up Marshall tonestack if ever I seen one.. :wink:

He said that he tried that and didn't like it.. but, if the preamp is jacked up sounding before it gets to the tonestack then it's probably gonna sound jacked up after the tonestack as well.. I wouldn't worry too much with trying to get the bias at each jfet set at half the supply voltage. If you are gonna do that then go back over them by ear and set each one to where they sound good because that is what you are going for, right? If they are all set at 4.75v and it sounds like shit then you need to go back over it by ear and see where the problem is. In any number of the jfet preamps I have built there will be 1 or 2 jfets that bias anywhere between 5.5v and 6.5v. If one jfet is off it can make the whole thing sound like it has too much treble or too much bass or whatever.

User avatar
caspercody
Resistor Ronker
Information
Posts: 388
Joined: 15 Jul 2009, 14:34
Has thanked: 34 times
Been thanked: 48 times

Post by caspercody »

I have not had much time to work on this, but the last thing I did was removed everything after R33 from the circuit. I then added the tonestack and presence control from the Dr Boogey project. Way better tone control.

I did try to bias by ear and setting at half of voltage. I keep thinking I can tweak it better. I just need to set it and forget it. But the sound sounds thin, if that makes sense. I am trying to make a 4x5 (holes) vero board with long leads so I can plug them into the JFET sockets on the BE100 board. And on these 4x5 vero boards I put more sockets so I can try different combinations of components. Like on the Dr Boogey they added capacitors from Gate to Source, or diode from Gate to Ground or different JFET to JFET like the AMT designs. But I need a P channel switch JFET.
Attachments
JFET add on.png
JFET add on.png (11.68 KiB) Viewed 2855 times
Last edited by caspercody on 21 Sep 2018, 16:11, edited 1 time in total.

User avatar
caspercody
Resistor Ronker
Information
Posts: 388
Joined: 15 Jul 2009, 14:34
Has thanked: 34 times
Been thanked: 48 times

Post by caspercody »

oops forgot, yes I did change R31 to 10k

User avatar
J0K3RX
Degoop Doctor
Information
Posts: 1057
Joined: 29 Jun 2011, 01:25
Location: US Florida
Has thanked: 315 times
Been thanked: 556 times

Post by J0K3RX »

caspercody wrote:I have not had much time to work on this, but the last thing I did was removed everything after R33 from the circuit. I then added the tonestack and presence control from the Dr Boogey project. Way better tone control.

hmm, that's odd.. The only differences between the Dr.Boogey stack and the scaled down Marshall version I posted is the treble HP filter cap value and the slope resistor value. The presence pot must be getting you where you want to be with the tone is all I can think? I used to add the presence pot between the treble and vol pots but I stopped doing that after I figured out how to add it after the output buffer which eliminates the possibility of any negative interaction with tonestack function.

User avatar
caspercody
Resistor Ronker
Information
Posts: 388
Joined: 15 Jul 2009, 14:34
Has thanked: 34 times
Been thanked: 48 times

Post by caspercody »

The vero board does work. I have changed the tone stack to the tone stack used in the Dr Boogie project. I removed all pars after R33 on the BE100 schematic.

I changed the resistor to 10K for R31.

I put a 2n5457 in the first stage and trimmed the voltages to ear. There is a little noise, but now realized it is from my amp.

User avatar
askwho69
Information
Posts: 15
Joined: 03 Aug 2015, 12:44
Has thanked: 2 times

Post by askwho69 »

Apply the runoffgroove no tonestock, from their Thor Schematics, or from BSIAB 2 tone control. Apply the SRRP as well.

Or build the JCM800 schem first then apply the Be-od modes one by one.


Lastly, add the phase inverter part of the preamp, see the jcm800 schematics like the Pal800, they include the phase inverter. Maybe it sounded thin because it's out of phase.
:D :D :D

User avatar
caspercody
Resistor Ronker
Information
Posts: 388
Joined: 15 Jul 2009, 14:34
Has thanked: 34 times
Been thanked: 48 times

Post by caspercody »

I do have the schematic for the PAL JCM800, is the phase inverter the components starting at the 22n (after pin 2 from treble) capacitor to the output?
Attachments
PAL_JCM800_Schematic_2.JPG

User avatar
askwho69
Information
Posts: 15
Joined: 03 Aug 2015, 12:44
Has thanked: 2 times

Post by askwho69 »

caspercody wrote:I do have the schematic for the PAL JCM800, is the phase inverter the components starting at the 22n (after pin 2 from treble) capacitor to the output?

Yes, after the "Bk2" after the treble control,it should have a master volume first then, 22n into phase inverter. In you simulation try to put the inverter after your volume control.

User avatar
teemuk
Breadboard Brother
Information
Posts: 147
Joined: 17 Aug 2009, 19:54
Has thanked: 3 times
Been thanked: 55 times

Post by teemuk »

These things are never going to nail the tone of the original tube design because a JFET simply isn't a "drop in" substitute for a 12AX7 triode tube.

Key elements that create "tone" of a circuit are overdrive characteristics and "voicing". These JFET "emulation" circuits can somewhat mimick the "voicing" aspect, as the interstage filtering circuits are more or less following the architecture of the original tube-based design. (FETs make this moderately easy because they have high impedance output and "voltage drive" input for gate, similarly to vacuum tube circuits). However, one needs to acknowledge why these filters aren't exactly identical to those in vacuum tube circuits: Source and drain impedances of JFETs are different from those introduced by a 12AX7 in "identical" circuit. This will alter turnover frequencies of all associated RC filters, whether they are in source bypass circuit, or in the interstage coupling circuit. Why moderately good results in emulating the tone are acquired is because these turnover frequencies sometimes still land "close enough" not to alter the tone dramatically into different direction. Sometimes.

Overdrive characteristics are a totally different ballgame, though. First, one needs to acknowledge that transfer functions of a common source JFET amplifier and common cathode triode amplifier are actually very different, triode is much more linear than a FET. FET will therefore inherently produce a very different pattern of harmonic distortion. In overdrive designs, especially those employing cascodes of overdriven stages, one also needs to "match" symmetry versus asymmetry of the clipping distortion (because it defines magnitude of evens versus odds), and "headroom" (because it defines at which sensitivity and magnitude each stage overdrives in respect to another). These aspects affect harmonic pattern of the distortion, as well as "touch sensitivity" of it.

FETs also do not have similar "grid conduction" characteristic as triode tubes. Grid conduction is an important overdrive mechanism, and these designs neglect it. This is why several FET designs feature those gate diodes: It's an attempt to capture this important aspect in how a real 12AX7 tube operates.

Overdrive circuits often do not produce a "fixed" pattern of harmonic distortion either. In fact, the pattern "dances" as sliding DC bias shifts dynamically control magnitude of even vs. odd harmonic distortion in real time. This is another element of "touch sensitivity". All this needs to match to the tube circuit to nail the tone as well as "feel" of playing guitar through the circuit.

These JFET "emulations" simply fail short in all of this. In fact, these "emulations" will never even provide any consistency in this aspect. Why?
- Because the design does not even try to match headrooms of the JFET circuit versus triode circuit
- Because the design does not even try to match symmetry vs. asymmetry of stage clipping of FET and triode circuits
- Because drain impedance and gain are entirely dependent on trimmer resistance, which in turn is dependent on widely varying JFET characteristics

Each time the drain trimmer is adjusted to bias the fet "properly" the adjustment changes gain of the stage, drain's idle DC voltage, and output impedance of the gain stage. All these changes have effect on clipping characteristics and "voicing".

Runoffgroove -style designs also often advise "center bias", meaning biasing for equal headroom of signal swing towards positive and negative directions. This is a nice "HiFi" design rule but overdriven tube gain stages actually often employ different bias point than "center bias" because bias point adjusts symmetry or asymmetry of clipping characteristics, and resulting tone and "feel". These designs neglect this important design aspect.

There's also something to be said about overdrive characteristics of different circuit architectures, a point demonstrated for example by the JCM800's FET "emulation". The "emulation" tries to emulate a phase inverter with two cascaded common source amps (with inconsistent trimmers) and negative feedback loop. However, JCM800's phase inverter is a differential amplifier circuit, and differential amplifier has totally different overdrive characteristics than a single-ended common source amp. Differential clips quite symmetrically and softly, common source amp clips asymmetrically and hard. You can fit a PPIMV to a JCM800 because differential's clipping sounds actually quite similar to overdrive of the final power amp (which is also moderately symmetric and soft). You can't expect even remotedly similar results from a totally different circuit architecture.

Why people encourage to use generic OpAmp & Diode clipping architectures for tube emulation is because such circuitry does not provide the same false sense of belief that all that is needed to mimic tube circuits is dropping B+ to 9 volts, subbing triode tube with a JFET, and adding a drain trimmer. Nope. With them you can't just follow the triode circuit's architecture and "paint by numbers" without a slightest clue of what you're actually even trying to achieve. As the circuit architecture changes (while still retaining its original function) the designer actually has to research how the tube circuit operates in regards of "voicing" and "overdrive characteristics" and then must devise a circuit, a very different kind of circuit than the initial tube-based, which emulates these characteristics as accurately as it can. It's certainly not as "easy" as runoffgroove -style "emulation" but can actually provide a far more accurate outcome when done properly. Also, generic OpAmps and diode clippers are far more predictable and consistent design elements than JFETs.

User avatar
caspercody
Resistor Ronker
Information
Posts: 388
Joined: 15 Jul 2009, 14:34
Has thanked: 34 times
Been thanked: 48 times

Post by caspercody »

Thanks Teemuk

I have noticed that the IC based pedals (BE-OD, VH4) do sound really good.

The reason I am trying the JFET is like you wrote it is easy to drop in a JFET in the tube position, and like you wrote I am finding the sound disappointing. It is missing something, but i can not pinpoint it.

I would love to try to do IC based emulations of amps, but I do not understand the criteria to do this. I have made the UBS version by BAJA, and it lacked gain, but I did figure out to put a pot in the feedback loop of the first and second op amps and got some values to try to get more gain. Next I will divide the caps by the same value i increased the resistors to to try to keep the same response.

User avatar
askwho69
Information
Posts: 15
Joined: 03 Aug 2015, 12:44
Has thanked: 2 times

Post by askwho69 »

Amazing explaination Teemuk, my brain exploded!! Otherwise, it is a good read.
Anyways, i heard the Pinnacle by Wampler demos it actually sounded like EVH, well it is copy of BSIAB ll with some modification. Maybe we can implement the BSIAB ll jfet style circuitry to emulate the BE-OD tubes schematics?

User avatar
askwho69
Information
Posts: 15
Joined: 03 Aug 2015, 12:44
Has thanked: 2 times

Post by askwho69 »

Casper, I just read a bunch of articles about jfet, you should remove the capacitors from the Source. Try reading Fetzer revisited at runoffgroove and tillman preamp too.

Post Reply