New FET distortion schematic...sensation?

[Jered]

OK, I built the D90x2 circuit with J201 fets and I'm getting no sound. Are the diodes silicon or germaniums? I used 1N914's.
Jered

[xbananoff]

The circuit(scheme) does not work because of errors in assembly most likely.
it is necessary to test conditions on a direct current. On a sink T1, T6 there should be half of supply voltage, if all is collected correctly.
In D90x2 the silicon diodes utilised.

[xbananoff]

Sorry, ".. on drain T1, T6" - this is correct.

[Medved]

Practically a year has gone... I'm very surprised that nobody not even tried to make it. Ok, lets aweken your interest.
http://webdrive.purga.ru/egorov/slo/sloII11.rar
Heh

[goosonique]

"xyeBo" (khuyovo)  thats one cool name for this project 
'from Russia with love' this one is ...

[Medved]

А по делу? Матерится мы все умеем, слава богу... 

[Medved]

Hey, is it realy uninteresting?

[DDD]

Hi all,
Medved is surprised that nobody reports on building the subject. To be frank, the fact (if true, of course) is more than surprising.
I can say for sure that more than a half of Russian-speaking stompbox DIYers are crazy about the devices from Xbananoff (the author). Maybe, because they've built and compared the devices with the other distortion stuff?
To remind: the "heart" of the device is a cascode circuit based on couple "so-so complementary" JFETs. That gives soft clipping together with tremendous gain in a very wide dynamic range of the input signal. Moreover, such method of amplification\clipping produces very "musical" and pleasantly distorted sound. Also viirtually endless sustain and negligible noise.
At the same time the JFET buffers,  JFET speaker simulating stages e.t.c give some pleasant colour to the sound.
The sound is actually great. One can easily try it. The sound clip is in the link above.

[Medved]

DDD, hi! It's very pleasant for mee to see here familiar nicknames;)
And it's really, really strange that this topic which was so poular here a year ago now is called

"xyeBo"

   
Ok,lets try again to awaken an interest. Here are two more samples:
http://webdrive.purga.ru/egorov/d90xgt/d90xgt1.rar
http://webdrive.purga.ru/egorov/d90xgt/d90xgt3.rar

Maybe, because they've built and compared the devices with the other distortion stuff?

DDD, please, don't instigate me to begin a flame about JFETS vs All 

[gaussmarkov]

i just noticed this because of your recent comments.  that d90x2 is one big circuit.     it might help interest if there were an english translation of the schematic.  the russian version is clear to an english reader except in a few spots.

is C1 a 1uF cap?
what do the asterisks/stars on C2, C3, C4 denote?

cheers, gm

[Medved]

The d90x2 schematic is rather old, but it is still rather good. Today, there are new tendences in russian  JFET schematics mostly thanks to xbananov and some more guys (me, Gunpowder, Dionis and... that's all  ) You can find us here http://forum.gtlab.net
Today we have these projects:
-Bsiab mod's and unique "Bsiab Extrem" by xBananov
-Slo100 on JFET's by me
-D90x/D900 by xbananov and me
-Some variations on Orange MKII by Gunpowder.
-SpeakerSim on JFET's by xbananov
-High accuracy SpeakerSim on JFETS by Gunpowder
-Simple guitar compressor on JFETS by xbananov and Lart.

I think, that it'll be better to discuss each project in separate topic, and most common things about xbananov's circuit technique here.
As for me I  see no reason to build the distortion + spikersim in one circuit (D90x2). Individually these devices are more universal.

c1 is a 1uf cup
Star means that this cup shoud be filled in individually according to your guitar and tastes.

[stm]

The d90x2 schematic is rather old, but it is still rather good. Today, there are new tendences in russian  JFET schematics mostly thanks to xbananov and some more guys (me, Gunpowder, Dionis and... that's all  ) You can find us here http://forum.gtlab.net
Today we have these projects:
-Bsiab mod's and unique "Bsiab Extrem" by xBananov
-Slo100 on JFET's by me
-D90x/D900 by xbananov and me
-Some variations on Orange MKII by Gunpowder.
-SpeakerSim on JFET's by xbananov
-High accuracy SpeakerSim on JFETS by Gunpowder
-Simple guitar compressor on JFETS by xbananov and Lart.

I think, that it'll be better to discuss each project in separate topic, and most common things about xbananov's circuit technique here.
As for me I  see no reason to build the distortion + spikersim in one circuit (D90x2). Individually these devices are more universal.

All the projects you mention sound really interesting, especially the BSIAB EXTREME, D90X/D and the Speaker Sim ones.  It would be great if you could post the links to the threads where the actual schematics are. I tried to navigate through the forum but it is really difficult, since it is not a matter of a different language, but also a different alphabet as well!

For instance, typing BSIAB in the search box gave me just one hit, which is a copy of your last post. Also, typing D900 gave me some hits, but couldn't get to the schematic either...

Best regards.

[Medved]

Heh, most of them are on our site www.gtlab.net
Ok, "Bsiab Extrem":
http://gtlab.net/gtlab2/readarticle.php?article_id=33
SpeakerSim on JFET's by xbananov
http://gtlab.net/gtlab2/readarticle.php?article_id=36
And rather old SpeakerSim by xbananov is in his famous article
High accuracy SpeakerSim on JFETS by Gunpowder:
http://gtlab.net/gtlab2/readarticle.php?article_id=31
About D900 you can (cannot ) read in russian topic which is named
"Frequency correction between cascades":
http://forum.gtlab.net/cgi-bin/yabb2/YaBB.pl?num=1118341835
it's a link to the first page of a discussion

http://forum.gtlab.net/cgi-bin/yabb2/YaBB.pl?num=1118341835;start=all
it's a link to ALL posts in topic (>18pages), so be patient. There you can see even some clipping oscillograms of xbananov cascades.
Link to scheme of my view of D900 (which i call D90xGT)  is
http://forum.gtlab.net/yabbfiles/Attachments/Dist90xGT4.PNG
Samples of D90xGT are in my post on 15-11-2005, 23:47:36, sample of Slo 100 on JFET's is in my first post (12-11-2005, 20:23:50)

[Medved]

Hmmm. I've forgotten to tell that we have a very big time difference- 11 hours. So it's a late night in my city now. I have only 4 hours till the morning, so i'd better go to bad  . See you $)

[stm]

Thanks a lot.  Very interesting reading.

Babel fish does an average job in the translation, so I'm trying to fill in the gaps and missing words!

The advanced speaker simulator is very interesting, especially the "rectified" response, which is much in accordance to what you actually *hear* in comparison to the "classic" response which is more in accordance to the theoretical response in an anechoic chamber (if I understood correctly).

Regarding your D90x Ampsim, there are four schems from D90xGT1 to GT4.  I assume the numbering is chronological and increasing in quality (?) or does it have something to do with a particular model of the D900?

Best regards,

STM

[Medved]

Babel fish does an average job in the translation, so I'm trying to fill in the gaps and missing words!

Sorry, I can't understand "Babel fish". Is it a name of a translator?

The advanced speaker simulator is very interesting, especially the "rectified" response, which is much in accordance to what you actually *hear* in comparison to the "classic" response which is more in accordance to the theoretical response in an anechoic chamber (if I understood correctly).

Hmm I really don't look at the classic/rectified mode from this point of view. Interesting %) As far as i know the main idea was to reproduce the amplitude-frequency characteristic of classic loudspeakers.

Regarding your D90x Ampsim, there are four schems from D90xGT1 to GT4.  I assume the numbering is chronological and increasing in quality (?) or does it have something to do with a particular model of the D900?

First of all, D90x isn't an Ampsim. It's a serious preamp like Soldano Slo-100, Bogner, Engl tube preamps. It isn't JFET schematics problems that mentioned above tube preamps cannot work from batteries and cannot be placed in a small box. 
Schems GT1 to GT3 were an evolution of the preamp with a clipping stage after the classical marshall tone stack. It was not a good idea. I've build it- not for my tastes. GT4 is a completed schema which sounds great.
There are two more samples, which show how D90xGT process music intervals and complex accords.
http://webdrive.purga.ru/egorov/d90xgt/d90xgt4.rar
http://webdrive.purga.ru/egorov/d90xgt/d90xgt5.rar
Only few very serious tube preamps can do something simiiar.

[stm]

Sorry, I can't understand "Babel fish". Is it a name of a translator?

Yes, it is an online translator capable of processing web pages as well.

Hmm I really don't look at the classic/rectified mode from this point of view. Interesting %) As far as i know the main idea was to reproduce the amplitude-frequency characteristic of classic loudspeakers.

You are right, after reading the translation several times I came up to understand that the term "office" (given by the translator) was refferring to the "speaker cabinet" and not to the "room" where the speaker was in.

First of all, D90x isn't an Ampsim. It's a serious preamp like Soldano Slo-100, Bogner, Engl tube preamps. It isn't JFET schematics problems that mentioned above tube preamps cannot work from batteries and cannot be placed in a small box. 

I see, so if you place a Magnum MkII speaker simulator after you are ready to go!

Schems GT1 to GT3 were an evolution of the preamp with a clipping stage after the classical marshall tone stack. It was not a good idea. I've build it- not for my tastes. GT4 is a completed schema which sounds great.

From what I found in the links you posted, the D900x (xbananov) and D90xGT (mod by you) preceded the GT1 to GT4 series.

There are two more samples, which show how D90xGT process music intervals and complex accords.
...
Only few very serious tube preamps can do something simiiar.

I agree, lots of clarity and sustain!

Please be aware that you have opened a can of worms, so now many questins have arised.

1) Is the LED shown in the source pins of the cascode JFETS a standard RED LED or other type?

2) I haven't been able to find information on the different diodes shown across the schematics. Consider there are at least five diode categories (apart from zeners) to consider:

a. Silicon rectifier (like 1N4001 and 1N4007)
b. Silicon signal (like 1N4148 and 1N914)
c. Shottky rectifier (like 1N5819)
d. Shottky signal (like 1N5711 and BAT45)
e. Germanium detector (like 1N34 and 1N60)

Could you please indicate the corresponding type for the following russian diodes:

KD509A
KD510A
KD521A
GD507A / 1D507A (are they the same?)

Best regards,

stm

[Medved]

I see, so if you place a Magnum MkII speaker simulator after you are ready to go!

Of course! Speakersim is very usefull thing even in a guitar head, not only in a stompbox.

From what I found in the links you posted, the D900x (xbananov) and D90xGT (mod by you) preceded the GT1 to GT4 series.

M... I'll try to explain... I number my schems chronologically. When i have a new idea i use a new number to publish the scheme.
The situation around this schemes is:
1) D900x (xbananov) & D90xGT - practically one thing. In GT i used bass boost in first clipping stage at it's minimum gain. That's all difference. This schems rather old.
2) GT1-GT3 is really poor choice. It was just a check-up of some ideas (mostly about tonestack and clipping stage after it). Of cource, you can build it, but result will be bad. At least I've got a bad result.
3) GT4 is the latest version of d900 at that moment. Bananov and me discussed ideas of this schema, he even made it, but didn't published the final schema. So I published my variant and named it GT4.

Please be aware that you have opened a can of worms, so now many questins have arised.

Heh, it was my goal $)

1) Is the LED shown in the source pins of the cascode JFETS a standard RED LED or other type?

First of all, lets come to an agreement about terminology.
1) About JFET
Vgs- gate-source voltage
V0 - gate-source cut-off voltage
I0 - drain current when Vgs=0

2) About cascades

It's called mu-cascade.
T1 and T2 should have equal V0
According to transistors vo diode is:
v0<1v silicon signal with voltage drop ~0.5v
1v<v0<2v IR led with voltage drop ~1v
2v<v0<3v red led with voltage drop ~1.4v
v0>3v green led with voltage drop ~1.8v
The main rule is that:
if bias on gate t2 is 4.5 then output bias should be approximatly the same (4-5v).
Instead of diode resistor with nominal of R2 can be used


It's called mu-cascade on cascode JFET's or just cascode.
T1 and T3 should have equal V0
T2 and T4 should have equal V0
V0 of T2,T4 should be 2-4 times bigger then V0 of T1,T3
Good values are 0.5-0.8v for t1,t3 and ~2v for t2,t4.
All transistors, especially t2,t4 should be with low noise!
Instead of diode resistor with nominal of R3 can be used

That's why you shouldn't install Leds in cascode! -it's the answer on your question.

KD509A,KD510A,KD521A are small signal silicon diodes.
GD507A / 1D507A are germanium diodes.

Now, you know main principles of basic schematics and we can discuss schems more deeply.

[Medved]

http://www.online-translator.com/srvurl.asp?lang=en
Try to use this translator- may be it'll give better results. Or you can just compare translations.

[stm]

Nice explanation ;D  Your comments are KEY to understand and design these FET circuits 
Certainly we needed to establish the basics in order to be able to go deep into this.

I think there are three main causes for these FET circuits not gain so much popularity in the past here:

1) The lack of understanding of the principles behind (which you have clarified now!)  This has to do with the language barrier as well.

2) The need of different types of JFETS with well-defined parameters. This may be a barrier for those who don't have more than one type of JFET available. FETS are also not so easy to get by.  In this respect an equivalence between the russian FETS and the more familiar J201, MPF102, 2N5458, etc here might help to break this barrier.  For instance, I can see that pairs like J201 and J202, as well as 2N5457 and 2N5458/2N5459 are good candidates for the 4-FET cascode.

3) Finally, the very large number of JFETS involved in the designs may scare away many people from experimenting, especially beginners, since chances of a mistake and a circuit not working are quite high  :icon_confused:  I'm not scared at all, however I must confess I'm lazy!  :icon_redface:  Nevertheless I fell in love with the Magnum MkII speaker sim. Especially the "rectified" mode.

I have a question regarding the samples. What was the setup you used? Guitar amp/speaker sim?

I believe you have achieved a great Marshall sound.  I'm more into cleaner sounds, like a Fender amp at breakup (with a bell-like or glass-like sound) and Vox amps as well.  I would like to experiment with this FET circuits along these lines.

Thanks again, and let's keep the channels open!  I'm confident more people will jump in as we start making progress.

STM

P.D. I'll give a try to the translator you suggest. It's good to have a second opinion, however not from doctors, since you end up with two different diseases!