Triode to JFet. The Champ prototype

[seadi123]

Put this on perfboard but it won't work. The connections on the perf are fine.  I designed this using ROG's triode to Jfet method, the only thing i did different was using a voltage divider instead of biasing each transistor with individual trimpot for 4.5V.

https://s2.postimg.org/ke9m2236h/IMG_0036.jpg

[amptramp]

It looks like you have shorted all your drains together and taken them to the V_R point.  You need resistors from each drain to the 9VDC supply.  It will not work until you do that.

[R.G.]

What he said. With the drains all tied together, it can't amplify usefully.

Even once you get a setup with individual drain resistors and so on so that it will amplify, you then have to remember that
>>> JFETs are not triodes, and any similarity of sound between them is coincidental <<<

Simply subbing a JFET into a triode circuit is not particularly likely to sound like the original triode circuit. Granted, this method has its appeal to simplicity and straightforwardness, but really, if JFETs sounded like triodes, every guitar amp since about 1968 would have had only JFETs in its preamp section, right?

[Plexi]

I agree.
Use trimpots there (V to Gate), and bias each one...

[Plexi]

I've tried the Fetzer valve.
But never 3 in series..do you have any link to that "triode to jfet"?
I would like to know the logic about that...I tried to reach the Echorec preamp.

[KarenColumbo]

Peruse this: http://milas.spb.ru/~kmg/tubeemulation_en.html. Good read, if only a few steps away from Mr. Danyuk's original schematic (which seems to be hard to find nowadays)

[seadi123]

I've tried the Fetzer valve.
But never 3 in series..do you have any link to that "triode to jfet"?
I would like to know the logic about that...I tried to reach the Echorec preamp.

There is no real guide, you have to check their various articles that use the method and kind of work it out. Start with the professor tweed, it's the easiest one and its almost the same design as the champ, very easy to understand

[duck_arse]

your first fet gate has no bias or reference. [if you are using in and out caps, draw them on your circuit.]

[Plexi]

I've tried the Fetzer valve.
But never 3 in series..do you have any link to that "triode to jfet"?
I would like to know the logic about that...I tried to reach the Echorec preamp.

There is no real guide, you have to check their various articles that use the method and kind of work it out. Start with the professor tweed, it's the easiest one and its almost the same design as the champ, very easy to understand

Great...googling now
Thanks!

[sajy_ho]

In the past I searched for tube sound emulation with FET's alot! But finally I got to this conclusion that simulating the tube sound is not just simulating any particular tube stage; it's about everything: frequncy shaping(the most important factor IMO), cabinet and speaker, ...
If you look at Runoffgroove's schematics during years, you can see they got to this conclusion as they improved their circuits; take Professor tweed for example, It was the first attemp to simulate Fender sound with FET's, then take a look at Azabache's schematic and see how it got more complex. The same with Thunderchief, Thor and Thunderbird...
IMO Azabache is very close to Fender sound, and it's so versatile, you can get all Fender models sound(black, blonde, tweed,...) just by tweaking some knobs...

[Plexi]

In the past I searched for tube sound emulation with FET's alot! But finally I got to this conclusion that simulating the tube sound is not just simulating any particular tube stage; it's about everything: frequncy shaping(the most important factor IMO), cabinet and speaker, ...
If you look at Runoffgroove's schematics during years, you can see they got to this conclusion as they improved their circuits; take Professor tweed for example, It was the first attemp to simulate Fender sound with FET's, then take a look at Azabache's schematic and see how it got more complex. The same with Thunderchief, Thor and Thunderbird...
IMO Azabache is very close to Fender sound, and it's so versatile, you can get all Fender models sound(black, blonde, tweed,...) just by tweaking some knobs...

I'm reading now, and reach almost the same conclusion at the moment.
Looking the ROG:
- Umble (Dumble): http://runoffgroove.com/umble.html
- Eighteen (Marshall): http://runoffgroove.com/eighteen.html
- Britannia (AC30): http://runoffgroove.com/britannia.html
- Ginger (Ampeg SB-12 Portaflex): http://www.runoffgroove.com/ginger.html
- Supreaux Deux (Supro 16T): http://runoffgroove.com/sd.html

Maybe I forgot one..

D*amn...how I love ROG 

[seadi123]

Update:
Put 10k resistors on each drain, the voltage was 1.6v on Q1, 1.7v on Q2 and 0.7v Q3. The circuit had very low volume and a farty sound. Replaced the 10k with 3.9k resistors. Q1 is 5.6v, Q2 is 5.9v and Q3 around 1.9v. Higher volume, too little headroom, the circuit sounds very good at full gain, but i'm not impressed with the clean sound when gain is at 30-40%. Will try other resistor values (don't have trimpots right now). Overall still not impressed with the sound, maybe i will ommit the 22k feedback resistor and completely redesign the circuit in a marshall-like sound

[Plexi]

That's why, I think, you need to balance with Source to ground resistor.
Which Jfets are you using?
Each one are a specia case (that's why you'll find trimpots in almost all the circuits), and every every species (J201, 2N5457, etc) is a particular animal.
I'll try to copy that values from the ROG's designs that I mentioned: sometimes you'll find some standard values (like 1/1k2 from S to Ground on J201).

[seadi123]

At the moment i only have 2n5458

[Plexi]

I've never had to bias a 5458, only used 'as is in the recovery stage of the Rat.
But I'm a huge fan of the 2N5457...and they're very close.

As I can see, you used bias values that usually are find on the J' family.

Try to play with 22k or close values (place there a 50k trimpot) at drain to get 4,5/5v (máx headroom at 9v), and 3/3k3 fixed from source to ground. 
If you can use here 18v, it'll be better.

[seadi123]

Update: Just wanted to say that i used trimpots and biased each transistor to 4.5V, and it works wonderfully & sounds great

[Plexi]

Great!
Pics/samples!