BLACK FOREST High Gain Preamp - Based on the Bogner Uberschall

[MyKeul]

@ Maik : Thanx for UK-Electronics; not so far from me and I didn't knew them.

@ JOK3RX : I was looking for "standard" FET and hadn't thought about SMD as I don't really like them. I prefer the standard one because it's easier to swap components when they're on DIP brackets. But thanx for your suggestion 

I have to order every components (maybe today after work)...

[jymaze]

Bias at lower than center maximizes the available gain (bigger drain resistance on the trimmer) if you care about that (who does not care about crazy high gain anyway...).

On a similar design, I used some 2N5457 and it worked fine. I had to sort the lowest VGSoff I could find (-0.6 to 0.8volts) and bias around 4volts. It works very fine.

[ggedamed]

Bias at lower than center maximizes the available gain (bigger drain resistance on the trimmer) if you care about that (who does not care about crazy high gain anyway...).

On a similar design, I used some 2N5457 and it worked fine. I had to sort the lowest VGSoff I could find (-0.6 to 0.8volts) and bias around 4volts. It works very fine.

Note that for the signal that passes through the source capacitor the gain is already maximized. The drain resistor / source resistor ratio has an effect only if the signal is outside the source capacitor bandwith or if there is no source capacitor.

[Bliss]

Thanks for all the information. We've played around with the voltages and it starts to sound quite allright. Though not like the Uberschall. Is it in any way possible to get it to sound like the Uberschall? Since that was our intention when we startet this project, but it seems to be a completely different sound....

Another point of trouble in our building is the boost, it still doesn't function at all. We've got no clue on the source of this problem.... Could it be an idea to add the variac board with two trimpots for volume and gain at the boost switch to add some more boost?

[J0K3RX]

@ Maik : Thanx for UK-Electronics; not so far from me and I didn't knew them.

@ JOK3RX : I was looking for "standard" FET and hadn't thought about SMD as I don't really like them. I prefer the standard one because it's easier to swap components when they're on DIP brackets. But thanx for your suggestion  

I have to order every components (maybe today after work)...

whatever works for you...



Better example here
https://dl.dropboxusercontent.com/u/11178619/J201%20SMT%20converters.pdf

May seem like a lot of work but I can make a whole sheet of them, solder them, cut them and use in no time... I use the really thin pcb from ABCFAB on ebay.  good practice

Another thing I noticed is that I have yet to find a dud in the SMD j201's and they are all relatively on spec... call it luck, but I'm sold!

[J0K3RX]

Thanks for all the information. We've played around with the voltages and it starts to sound quite allright. Though not like the Uberschall. Is it in any way possible to get it to sound like the Uberschall? Since that was our intention when we startet this project, but it seems to be a completely different sound....

Another point of trouble in our building is the boost, it still doesn't function at all. We've got no clue on the source of this problem.... Could it be an idea to add the variac board with two trimpots for volume and gain at the boost switch to add some more boost?

Have you heard my sound clips? That is what mine sounds like... Says "based on Bogner Uberschall" not sounds just like a Bogner Uberschall. For obvious reasons that ain't gonna happen but mine sounds pretty damn good, I think?

in regards to the boost switch, you absolutely need to have at least one of those resistors going to ground or it will not work right or sound good at all! Why don't you forget the switch and just solder the switch end of the 68k directly to ground? Try it and report back.

By the way, if you don't even have the base layout working then building the variac board will only make matters worse!! Highly recommend holding off on that until you get the pedal working...

Keep in mind, this pedal has to have a combination of all things coming together and working correctly to produce a satisfying result! If one thing is outa whack it will not work correctly... Not a good beginner pedal to say the least! 

[jymaze]

ggedamed,

Even when considering AC through the source bypass cap, there is the jfet intrinsic source resistor of the jfet and the ESR of the cap to take into account for the gain equation. It does not seem like much, but it count for a few dB of gain on an already relatively low gain device, then multiplied by 4 or five stages... You see the picture!

Actually, Rd always count in the gain equation of a jfet, even in pure common source configuration: the gain is proportional to Rd. It is also true for a bjt, I am not sure why you had the impression Rd (or Rc for that matter) did not influence the gain.

So in a nutshell, Rd is VERY important.

[MyKeul]

Thanx JOK3RX, using your kind of "SMD PCB Adaptor" is a very good idea (why haven't I thought of this before ).

Can I use the Black Forest which is on page 41 as it is in the post or are they any mods to do ? I must admit it's not really simple to me reading and understanding everything in English (maybe you could answer in French ? ).

[jymaze]

I speak French fluently, so send me a private message if you have any question you want to ask.

[ggedamed]

ggedamed,

Even when considering AC through the source bypass cap, there is the jfet intrinsic source resistor of the jfet and the ESR of the cap to take into account for the gain equation. It does not seem like much, but it count for a few dB of gain on an already relatively low gain device, then multiplied by 4 or five stages... You see the picture!

Actually, Rd always count in the gain equation of a jfet, even in pure common source configuration: the gain is proportional to Rd. It is also true for a bjt, I am not sure why you had the impression Rd (or Rc for that matter) did not influence the gain.

So in a nutshell, Rd is VERY important.

You're right. And thank you for making me to finally crunch some numbers.

I'll try here to make some thinking at my n00bish level.

I'm not sure about the JFET intrinsic source resistor and the ESR of the source cap - I'll say that first is in the tens of ohms regions and the second in the ohms region. Let's presume that the sum of the two resistances is 50 Ohm, for the sake of this example.

Let's take a J201 JFET with Vp = -0.8V and Idss = 0.75mA.

I'll settle for Id = 0.4mA and Rs = 1 kOhm, because it's almost half the usual 1k8 value for the source resistor - this  will maximize the effects of the drain resistor modifications in the absence of the source capacitor.

The drain voltage for maximum headroom will then be (9 - 0.8 )/2 + 0.8 = 4.9V, which gives Rd = (9V - 4.9V) / 0.4mA = 10.25 kOhm. The headroom on the both sides of the amplified signal is 4.1V (theoretically).

Maximum gain is Av = - (gm * Rd) / (1 + gm * Rs), where gm is the transconductance of the JFET, maximum 0.5mS for J201 (and a typical 15mS for 2SK117!).

So the gain of the stage will go along the bandwith from -3.42 times (~10.67dB) to -5 times (13.98dB).

Now, let's increase Rd until the headroom on the negative side is only 0.5V. This means that the drain voltage will be 0.8 + 0.5 = 1.3V, which leads to Rd = (9V - 1.3V) / 0.4mA = 19.25 kOhm.

The gain will now be between 6.42 times (16.15dB) to 9.39 times (19.45dB). Meaning an average increase of 5.5dB. Almost double.

Which is good, BUT the signal will be extremely clipped. The resulting signal will be distorted from almost the start. Which could be quite fine for many. What I like when the bias aims for maximum headroom is that the sound will pass a much larger range of transformation from almost clean to extremely distorted, depending on how hard one picks the strings.
So, this additional gain wil come at the expense of the "responsiveness" of the effect.

One more efficient way to increase gain seems to use JFETs with larger transconductance like 2SK117. When calculated for a 2SK117-Y with Vp = -0.55V with the same 1 kOhm source resistor, the gains for the maximum output headroom were from 9.9 times (19.91dB) to 90.54 times (39.14dB!!!). Not counting the much lower noise of 2SK117.
I'm not sure my calculations are entirely correct, yet I know that I only needed 3 stages with 2SK117 instead of 4 or 5 stages with J201 for this example of bad playing:

[pakrat]

You have a pretty versatile distortion there. Great vid ggedamed!

[jymaze]

ggedamed,

I agree, the trick to maximize the gain is to bias under V+/2, maximizing Rd without killing the sound quality. Probably about 3.5 to 4 volts is still fine.

Switching to higher gain jfet is even better, as long as the Vgs off / Vp is still low enough to allow Vd to stay high...

Subtle!

[palobonilla]

Hey guys i'm new here and i've readed the entire post... that was inspiring. I would like to sahre with you this pedal/preamp i've designed recently y hope you enjoy it


subir gif

[palobonilla]

Here a sound test https://soundcloud.com/palo-bonilla/aggression-mu-test

[pakrat]

Great sounding clip palobonilla, thanks for sharing. Welcome to the forum!

[kingswayguitar]

Here a sound test https://soundcloud.com/palo-bonilla/aggression-mu-test

holy crap
3 mu-amps in series
over the very top 4 sure
love it

[J0K3RX]

Hey guys i'm new here and i've readed the entire post... that was inspiring. I would like to sahre with you this pedal/preamp i've designed recently y hope you enjoy it


subir gif

oh baby! Please share layout!!! 

Love it!!!

[kingswayguitar]

@ ggedamed

scoopy...i love it
there's a guy at work, last name Schoolcraft
we call him scoopy
long story

[palobonilla]

i dont have a good layut by now :S but i will work in that for sure

[posmaster]

Hey guys i'm new here and i've readed the entire post... that was inspiring. I would like to sahre with you this pedal/preamp i've designed recently y hope you enjoy it


subir gif

Any particular reason for the 1458, or would any decent dual op-amp work (TL072, etc)