New FET distortion schematic...sensation?

stm · December 16, 2005, 09:25:20 PM

Quote from: Medved on December 16, 2005, 04:21:16 PM
Setup is the same:
Guitar yamaha rgx-112-->Clean channel-->Power amp tda7294-->Cabinet 1*12 2/3 closed back--> Mic Shure 565sd -->Audigy

First of all, you should try supply voltage near 25V with adjusting R3 and maybe R11. I have only 15V, but it's sounds very good. Soon, i'm going to add a tremolo %)

Nice sound considering you are using a SS Power amp, which as I understand, it has basically flat frequency response and negligible THD.

It is important to notice the higher voltage requirement. For a typical pedal this circuit could be implemented with two 9V batteries, for a total of 18V. I did some experimentation last weekend with the Fetzer Valve, Marsha Valve and MuAmp (using 9V), but wasn't very satifsied with the sound when they started clipping. Gotta try with higher voltage and use the MuAmps instead of single FETs. In fact I got better results with the experiments I did with a CD4049 in respect to low OD levels for "clean" guitar playing.

MR COFFEE · January 13, 2006, 01:09:01 PM

Bad sign for this COOL international exchange - all the links on

http://webdrive.purga.ru/egorov/

went dead.

Let's hope Medved just changed servers...

The clean channel sounds a lot like the old single tube preamp Alembic used to make... Mainly the sound of the tone controls. Sounds better just before it starts distorting audibly. I suppose it is just clipping the transient peaks, but my oscilliscope is on the bench to fix this month

Anybody else playing with the quad cascode these days? It does have some interesting sonic properties if you play around with various JFETs and circuit values.

MR COFFEE · January 15, 2006, 07:12:21 PM

Hey folks,

Just switch the web addrress to

http://webdrive.reborn.ru/egorov/

and Medved is back on line.

Anyone working on the translation business further (trying to clean up the machine translation and symbols to match the USA versions) ?

Medved · February 04, 2006, 03:38:32 AM

QuoteJust switch the web addrress to http://webdrive.reborn.ru/egorov/ and Medved is back on line.

Yes, absolutely right

QuoteIt is important to notice the higher voltage requirement. For a typical pedal this circuit could be implemented with two 9V batteries, for a total of 18V.

Yes, for pedal 18v is the best solution. I didn't try to use 25v... There could be some problems... 9v is to low for my clean scheme. If you want to use 9v you should change first cascade to follower.

Guys, if you want a quick response, write me directly on my mail : nospamtomson@bk.ru. Without nospam of course...

Medved · February 04, 2006, 07:14:12 AM

http://webdrive.reborn.ru/egorov/d90xgt/d90xgt11.rar
In style of Napalm Death. Tunning in B...

MR COFFEE · February 04, 2006, 11:28:16 AM

Wecome back!!

Medved · February 04, 2006, 02:35:36 PM

Thanks.

ragtime8922 · February 04, 2006, 03:23:33 PM

Nice sound. Is the schematic link dead? I'd love to breadboard this circuit.

theman · February 05, 2006, 01:09:25 AM

hi medved,

i've seen lots of versions of the d90x series. the latest seems to use a lot of complementary push-pull stages. how good does the matching have to be? this seems to be a drawback of using these.

Medved · February 05, 2006, 02:26:10 AM

Quote from: theman on February 05, 2006, 01:09:25 AM
i've seen lots of versions of the d90x series. the latest seems to use a lot of complementary push-pull stages. how good does the matching have to be? this seems to be a drawback of using these.

Complementary push-pull stages are just buffers. I can redraw scheme without it.

theman · February 05, 2006, 08:37:44 AM

yes, if you could, please post your latest design schematic using buffers which are not push-pull. what do you recommend (jfet source follower with another jfet as a current source?).

Medved · February 05, 2006, 09:00:34 AM

Ok.
http://webdrive.reborn.ru/egorov/d90xgt/Dist90xGT_b.PNG
It's moderately high gain (clipping begins from 0.5-1m of input signal). All trannies are nJFET.
KP303A<-->low noise, V0~-0.6-0.8v, I0~1ma (j201, but it is more noisy then KP303A)
KP303G<-->low noise, V0~-2v, I0~6..10ma or so (2n3819 maybe... but i can't say anything about it's noise...bf245b seems to go... Again i don't know about it's noise...)
Sound:
http://webdrive.reborn.ru/egorov/d90xgt/d90xgt12.rar

theman · February 06, 2006, 05:17:32 PM

thanks!

btw, how do you change the design and get clips out so fast? is the design on a proto-board or breadboard?

Medved · February 07, 2006, 06:05:41 AM

Quote from: theman on February 06, 2006, 05:17:32 PM
btw, how do you change the design and get clips out so fast? is the design on a proto-board or breadboard?

Yes.

Medved · February 10, 2006, 03:52:49 PM

Some minor changes were maid in last scheme dist90x_b.png, so check it out...
About diods. D1 is any led with 1-1.5v drop (With kp303g (v0=-2v) i use 1v drop infrared led)
D2 is any germanium diod, D3- silicon. You can try any other type of diodes and choose one you like best.

Here is sound in linein and processed one with impulse of 2*12 cab.
http://webdrive.reborn.ru/egorov/d90xgt/Line1.rar
Some comments. Max.gain (more than max.gain of bognerTG by two or three times!) sounds rather pleasant in linein: clear attack, good lows, moderate hiss, accords are well readable...

theman · February 10, 2006, 09:07:47 PM

are the minor changes you mention just the diode changes you talked about in your last post?

Medved · February 11, 2006, 06:31:58 AM

No. Some capacitors, etc... Soon, i'll publish more or less finished one...

Medved · February 12, 2006, 06:45:16 AM

Ok. Some more changes were maid in dist90x_b.png. Now it's good.

Technical characteristics:
Supply voltage: 9..24v
Quiescent current: 5-6ma
For Russian JFET's input voltage noise at 1kHz: 3.5nv/vHz.
Gain up to 90dB at 1k. With 15 supply it starts clipping at 0.5-0.6mv double amplitude input voltage.

KP303A<-->low noise, V0~-0.6..-0.8v, I0~1ma.
KP303G<-->low noise, V0~-1.8..-2.5v, I0~6..10ma or so.
D1 is any led with 1-1.5v drop.
D2 is any germanium diod.
D3, D4- silicon.

J1,J2 should have equal V0.
J5,J6 should have equal V0.
J7,J7 should have equal V0.
J4 should have I0 5-15% bigger than I0 of J3.
J9 should have I0 5-15% bigger than I0 of J10.
J12 should have I0 5-15% bigger than I0 of J11.

theman · February 12, 2006, 07:57:42 AM

thanks for the new info. please post the new schematic when you are finished. regarding the I0 differences ... these are for the same type of transistor. will the design work if the transistors do not meet the "5-15% bigger" requirement? also -- would like to understand the advantages and disadvantages of the buffers you use. the push-pull complementary seem to be good, but might need matching as opposed to the JFET source follower with a jfet current source. which is preferred? do you hear sonic differences?

Medved · February 12, 2006, 08:45:42 AM

Quote from: theman on February 12, 2006, 07:57:42 AM
thanks for the new info. please post the new schematic when you are finished.

New scheme is in old place. So, download it again.

Quote from: theman on February 12, 2006, 07:57:42 AM
regarding the I0 differences ... these are for the same type of transistor. will the design work if the transistors do not meet the "5-15% bigger" requirement?

From 6 JFETs you'll find 3 pairs. Belive me

It will work only if upper JFET has bigger I0. If its I0 is much bigger, than buffer can clip signal. Anyway, just check the output bias- if it's less than 100-200mv then it's ok.

Quote from: theman on February 12, 2006, 07:57:42 AM
also -- would like to understand the advantages and disadvantages of the buffers you use. the push-pull complementary seem to be good, but might need matching as opposed to the JFET source follower with a jfet current source. which is preferred? do you hear sonic differences?

Complimentary buffers have smaller output impedance, a bit (1..2db) less noise and soft clipping.
But in this case it is no difference between buffers at all.

News:

New FET distortion schematic...sensation?