MPP

the Jfet / 417a Cascode origins

I thought that Arthur Loesch was the first one that used a Fet-Tube cascode as MC input.
In the latest Tempo phonostage there is only a WE417 at the input so i may be wrong. Yes, the other Lösch ( Thorsten ) has a Fet-Tube cascode in his Toccata.
The Rauschfrei cabinet is in part made by Modushop and in part in Germany and the Netherlands. I found the Modushop front and back plate not of sufficient quality for a piece like this.


I can't and do not take credit for much at all. For this I will take a modest amount of insignificant credit. After all a cascode is hardly a new idea.


But Arthur tried the JFet/417a cascode arrangement a number of years after I decided to try and use it myself. I did not offer any sort of product with it, and built it into my own phono stage, which quite frankly was mostly the same as Arthur's.

The reason for that was that Arthur lives only a few miles from me and in the 90s I ran an audio club in the area, of which Arthur was a member. So, having met him even before the audio club started I got to watch him try almost every possible configuration of phono stage over a period of time. His original phono stage (published version) was really quite beguiling in terms of the overall sound, wonderful to listen to. But with a single 417a in the front end it was really necessary to match and select tubes with low noise. Even then it still had a bit too much noise... so I made the jfet cascode for myself. Look ma, almost no noise! :D

To the best of my recollection, both Arthur's and Thorsten's versions of the Jfet/417a came chronologically later than the version I made. iirc, Allen Wright also tried the same sort of cascode...

Now to be accurate, I did not tell or show Arthur my circuit. But I showed it off to a number of people, audio friends, from our same local group who doubtless "spilled the beans". ;)

I would not be surprised to find a jfet lurking in Joe's Tempo phonostage. dunno. could be xfmrs too.. dunno.

I know JC subsequently used a similar front end on at least one of his custom phonostages...

ok, that's my pitch for high-end fame and immortality... ha!

sorry to present this so late, but the post about this just popped up while i was looking for something else...

_-_-bear
 
The Fet at the bottom can be a 2SK170 for example. That combined with an ECC88 can make a low noise MC input. Well, not that low noise that we do here but something close to 1nV/qHz. That is the noise of ca.50 Ohm. The Fet has higher Gm to begin with and tubes have a rather high 1/F corner so even if they are steep they have a lot of noise in the bass. The 2SK170 has a somewhat lower 1/F frequency, not as low as bipolars though but sufficient. If you use a tube that can take higher idle current you can use several Fets in parallel or Fets with higher Gm. Then the noise can be rather low. An even better solution can be a folded Fet-Tube cascode like done by JC Morison.
I have published that circuit before here and i think i also explained it.
 
Something like this .....
 

Attachments

  • hybrid-folded-cascode for RC.gif
    hybrid-folded-cascode for RC.gif
    164.1 KB · Views: 344
You can put coils in the gates and then 8 is the limit. I even managed a prototype with 13 Fets at the input. Sometimes it likes to work, sometimes not.
Ricardo, the "Loesch" circuit has the Fet at the bottom in the cathode chain of the ( top ) tube. That way the noise of the tube is not correlated with the noise in the Fet, just as in the Simplistic Salas. The top BC550 has 140 Ohm Rbe`, still it does not contribute much to the total noise of the circuit. The BC550 is common base so it only amplifies the voltage and not the current.
 
Joachim
Tanks for that SCH
I use a similar MC front end on my Salas Simplistic.

Just a couple more stages with J fets not Valves.

Suply of inversed polarity as I use 4 2SK170.

There are a couple of things In your drawing that I will try up
I have 33 Homs as source resistors and ommited the 25000U capacitor.