j-FET / Triode Phono front-end

What people are working on at the moment
JamesD
Old Hand
Posts: 997
Joined: Tue Oct 09, 2007 4:26 pm
Location: North Yorkshire

#241

Post by JamesD »

Damn! I screwed up the folded cascode design! It won't work as intended... Funnily enough it does work as a normal cascode but not properly and that is due to JFETs being, at the physical level, symmetrical devices so that drain and source can be swapped and the beastie will still work but not with the right parameters....

Sorry I'll have to change this design a bit to get it right. Amazing how trying to calculate the component values can show things up :shock:

J
Andrew
Eternally single
Posts: 4206
Joined: Thu May 24, 2007 2:18 pm

#242

Post by Andrew »

Fets are fun aren't they?

How does the folded thinymajig compare the the front end of some of those hyper fancy j-fet input op amps. I seem to remember hagtech did a fine job of a head amp using jfet input op amps, Clive had one, he loaned me it for a liesten I was very impressed, very good detail just the bass was a tad dry. I really liked it dunno if Jim is still doing it as a kit tho?

Andrew
Analogue, the lost world that lies between 0 and 1.
JamesD
Old Hand
Posts: 997
Joined: Tue Oct 09, 2007 4:26 pm
Location: North Yorkshire

#243

Post by JamesD »

John Curl uses folded cascode in his jfet mc pres such as the Mark Levinson JC-1 and JC-2 for instance. These get fabulous reviews There is a thread or two on this on DIYAUDIO - the Blowtorch pre thread for instance Beware this is a very, very noisy thread - over 44,000 postings and probably only 10% on topic! Huge debates/rants/comments between the subjectivists and the double blind testing only brigade! I only discovered this earlier this week and its quite fascinating but very time consuming... I've only covered about 10% of it so far...

JFET cascodes have been used for even longer. The first JFET mc headamp I built (1978) was differential JFET cascodes with a JFET CCS in the tail. It was a bit good - it was flat to 5MHz!!! The noise was audible but didn't detract from its excellent performance - it blew away my AT SUT for everything except noise. Of course in those days my amps were ss as I had given up valves by then - funny how the circle of life turns! Of course this wasn't
the first use of JFET cascodes! I pinched the topology idea from Tektronix
:D

John Curls view is that the discrete folded cascode design is 6-8dB quieter and achieves comparable distortion whilst being open loop.

Its common that in the application notes for even the ultra quiet OP amps that to improve the performance they often add a JFET diff pair on the input..

J
JamesD
Old Hand
Posts: 997
Joined: Tue Oct 09, 2007 4:26 pm
Location: North Yorkshire

#244

Post by JamesD »

Had a re-think about the design and come up with this revision... This actually has folded cascodes :D
Attachments
JFET complimentary folded cascode MC head gain block v2.png
JFET complimentary folded cascode MC head gain block v2.png (8.18 KiB) Viewed 10297 times
User avatar
IslandPink
Amstrad Tower of Power
Posts: 10041
Joined: Tue May 29, 2007 7:01 pm
Location: Denbigh, N.Wales

#245

Post by IslandPink »

Some of you ( eg. Andrew ) are sitting on a gold mine out there ! :
http://www.audiocircle.com/index.php?topic=122057.0
"Once you find out ... the Circumstances ; then you can go out"
User avatar
izzy wizzy
Old Hand
Posts: 1496
Joined: Fri Nov 02, 2007 7:02 pm
Location: Auckland NZ
Contact:

#246

Post by izzy wizzy »

Before that transformer had turned itself into unobtanium, I let a spare pair go for what I paid for them - £100 the pair. Always wish I'd kept them. Now I really really wish I'd kept them.

cheers,

Stephen
User avatar
IslandPink
Amstrad Tower of Power
Posts: 10041
Joined: Tue May 29, 2007 7:01 pm
Location: Denbigh, N.Wales

#247

Post by IslandPink »

Good time to mention that Pieter is working on some Mk.2 step-ups similar to the 1:10 cobalt amorphous ones I have, but with more primary inductance, for carts like 103 & 103R .
"Once you find out ... the Circumstances ; then you can go out"
User avatar
slowmotion
Old Hand
Posts: 282
Joined: Fri Aug 21, 2009 4:07 pm
Location: Norway

#248

Post by slowmotion »

I've just finished reading the whole thread.
Haven't got anything to contribute, but I think it is fascinating stuff.
Have to go outside to clear the driveway for snow now,
but my mind will be full of silver wire transformers and other stuff I can't afford .... 8) :D
- Jan -
User avatar
Dave the bass
Amstrad Tower of Power
Posts: 12273
Joined: Tue May 22, 2007 4:36 pm
Location: NW Kent, Darn Sarf innit.

#249

Post by Dave the bass »

slowmotion wrote: Have to go outside to clear the driveway for snow now,
Wow! Thats amazing! You have your snow delivered to the driveway? :shock:

Once its been delivered I'm guessing you then have to spread it over the roads and house and garden I'd imagine yeah? :-)

DTB.
"The fat bourgeois and his doppelganger"
User avatar
slowmotion
Old Hand
Posts: 282
Joined: Fri Aug 21, 2009 4:07 pm
Location: Norway

#250

Post by slowmotion »

heheh :D

What I would like is snow that fall everywhere but roads and driveways.
And only for Christmas! :)
- Jan -
User avatar
IslandPink
Amstrad Tower of Power
Posts: 10041
Joined: Tue May 29, 2007 7:01 pm
Location: Denbigh, N.Wales

#251

Post by IslandPink »

Better than my friend who went back to Ontario after New Year and found his house buried under 4ft+ of snow, and the basement flooded with water from a burst pipe !

This thread might resurrect if Nick wants to post something about his latest phono experiments. It might go on another thread, though . This is using JC Morrison's 'Gm amp' concept, and he reckons you can get it quiet enough for direct-MC input, no step-ups .
"Once you find out ... the Circumstances ; then you can go out"
User avatar
Nick
Site Admin
Posts: 15706
Joined: Sun May 06, 2007 10:20 am
Location: West Yorkshire

#252

Post by Nick »

Well, its borderline quiet enough, in practice its ok, but in reality a step up is still needed with a 103r.

I will draw up some diagrams when I have it all down. But basically I am using a d3a as a pentode, then loading it with a CCS (yes I know), but running the CCS with a resistor in parallel to to give the effect of a high load resistor without needing the large B+ that would go with it. Its done that way as I misunderstood the description of the JCM stage where the resistor is between the anode and ground. At AC effectively the same, I think his way will have better PSRR, but I like mine as it maes me worry about using a simple CCS less. Then I take the anode of that into a ecc88 cathode follower and that drives the RIAA. Then it gets a bit odd. The g2 voltage is derived from the other triode in the 88 being used as a cathode follower to provide a low impedance voltage source for g2. But just for fun, I take some of the signal from the anode and add it to the DC so there is feedback into g2. So the d3a is in effect running in the same was as a UL output stage, so the g2 feedback reduces gain and lowers the output impedance. Between the d3a and the ecc88, I get a gain of 200 and a output impedance down in the 100 ohm range. Less if the g2 feedback is taken from the load output.
Whenever an honest man discovers that he's mistaken, he will either cease to be mistaken or he will cease to be honest.
IDM
Old Hand
Posts: 249
Joined: Wed May 07, 2008 5:50 pm

#253 Re: j-FET / Triode Phono front-end

Post by IDM »

Hi I thought I would resurrect this thread as I am interested in trying a phono without the need for an SUT.

I have managed to source some 2SK369's and have been measuring their IDSS in order to match them. The process of measuring the IDSS is straightforward but as I thought about it I became confused as to what matching means, is it:

1. All 4 FET's to have the same IDSS?
2. Is it each channel to have the same total IDSS (when you add the 2 FET values together?
3. Is it to have the 2 FET's in each channel to have the same IDSS, but not necessarily the same as the other channel?with

Also how matched is matched? I have several with IDSS values within 5% of each other, is that good enough?

Hope someone can throw some light on this before I ask the next beginners question!
Cheers
Ian
User avatar
IslandPink
Amstrad Tower of Power
Posts: 10041
Joined: Tue May 29, 2007 7:01 pm
Location: Denbigh, N.Wales

#254 Re: j-FET / Triode Phono front-end

Post by IslandPink »

If I look back at the beginning of the thread I think I received a few jFets from Nick to get this going. I don't remember measuring Idss myself. I think getting the same value on each side from each pair will be enough, and I tend to think that 5% tolerances are fine. The signal is very small so distortion profiles from slightly different jFets are not going to come into it much, so I suppose you just want to match gain and I suppose the Idss for the pair sets the gain of the cascode (?)
Nick, please correct me if I am wrong.
I tend to thing 'what's the worst that can happen', so probably best just to build it.
"Once you find out ... the Circumstances ; then you can go out"
IDM
Old Hand
Posts: 249
Joined: Wed May 07, 2008 5:50 pm

#255 Re: j-FET / Triode Phono front-end

Post by IDM »

Hi Mark,

Yes, I will build it, it would be useful to know which way to think regarding the matching though. Hopefully Nick might enlighten me at some point. I had a quick Google to try and find articles on the importance of matching, and whilst there is loads on how to do it, there seems little (easily accessed) on why and how to implement.

It will take me awhile to build the phono as the 6800pf silver micas are in transit from the Ukraine and all my projects progress on a geological time scale.

Cheers
Ian
Post Reply