diyAudio

diyAudio (http://www.diyaudio.com/forums/)
-   Pass Labs (http://www.diyaudio.com/forums/pass-labs/)
-   -   new SemiSouth SJDP120R085 "normally on" JFET (http://www.diyaudio.com/forums/pass-labs/166144-new-semisouth-sjdp120r085-normally-jfet.html)

zhoufang 1st May 2010 09:28 PM

new SemiSouth SJDP120R085 "normally on" JFET
 
1 Attachment(s)
Now they have depletion mode SiC JFET!

SemiSouth Laboratories, Inc. | Products
http://www.semisouth.com/products/up...085_rev1.0.pdf

what is amazing is that the Id vs Vgs looks rather different.
If the graph is right, this thing has maximum transconductance when Id is zero.
And at zero or near zero Id, the transconductance is >20S!
And look at the linearity of the curve at zero Id!
I hope it is not an mistake in the datasheet.

Melon Head 2nd May 2010 01:31 PM

Quote:

Originally Posted by zhoufang (Post 2172577)
what is amazing is that the Id vs Vgs looks rather different.
If the graph is right, this thing has maximum transconductance when Id is zero.
And at zero or near zero Id, the transconductance is >20S!
And look at the linearity of the curve at zero Id!
I hope it is not an mistake in the datasheet.

The scale of the y axis is too coarse (ie 10A per division) it would be nice to see the graph reproduced with 1A per division or even 0.5A per division, over a range of 0A to 10A (instead of 0A to 70A)
They do look interesting though.
How much are they?
When I can buy these for less than $10 each then I will be definitely interested.

zhoufang 2nd May 2010 02:44 PM

Thanh1973,

I think you made an excellent point.
Look at fig.1 - Id vs Vds at various Vgs:
Id at Vds=5V and Vgs=-2V = appox. 29A, which is close enough to Fig. 4 which tells 30A. (With this I assume Fig. 4 is at 25 degC)
And at Vds=5V and Vgs=-1V, ID= appox. 45A.
So the average transconductance for -2V < Vgs < -1V is 16V/1A = 16S.
While the average transconductance for -4V < Vgs < -2V is 29V/2A = 14.5S.
This shows that maximum transconductance should be between 29A < Id < 45A, not at zero Id as suggested by Fig. 4

My guess is that for figure 4 they take too few points and smoothed the curve too much, thus it is useless for our interest.
This is not the first time that this happened, the Id vs Vgs curve of the R100 is also pretty much useless.

But the self biasing is still interesting though.

$10 a piece might be unrealistic for now, but I'm sure it will be possible in just a few years time.

Nelson Pass 2nd May 2010 04:44 PM

This is a new part that replaces one that I have been working
with in house. The earlier part is very nice, and I used it to
make a version of Hiraga's Nemesis. It looks like they were
waiting for the newer version before putting it on the price list.

You may recall that Jan Didden got involved in having a
transformer made for a Nemesis in the U.S., but decided that
shipping costs to Europe were prohibitive. I took that project
over, and after some additional work developed a single-ended
amplifier using a revised transformer and the predecessor to
this part. The article is scheduled to go to Jan's new magazine
at the end of May.

:cool:

Bigun 3rd May 2010 03:34 AM

It looks very interesting and I'll be looking forward to learning more about this.

A side question, and a small matter, but I don't understand why the spec. sheet shows the operating temperature limited to 150C. SiC can operate at much higher temperatures, so is it a limitation due to other materials in the packaging?

I hope Jan will get to hear it one day.

ra7 3rd May 2010 04:03 AM

Quote:

Originally Posted by Nelson Pass (Post 2173327)
This is a new part that replaces one that I have been working
with in house. The earlier part is very nice, and I used it to
make a version of Hiraga's Nemesis. It looks like they were
waiting for the newer version before putting it on the price list.

You may recall that Jan Didden got involved in having a
transformer made for a Nemesis in the U.S., but decided that
shipping costs to Europe were prohibitive. I took that project
over, and after some additional work developed a single-ended
amplifier using a revised transformer and the predecessor to
this part. The article is scheduled to go to Jan's new magazine
at the end of May.

:cool:

Great news!

So, will this new part be replacing the one you were working on?

Nelson Pass 6th May 2010 09:22 PM

Pops right in.

:cool:

Melon Head 7th June 2010 03:48 AM

Anymore news on this part.
Has anyone tried it?

Nelson Pass 7th June 2010 07:49 PM

I've tried them, they work great. They do require a negative
bias on the Gate, but for single-ended Class A operation they
have quite low distortion, and it is dependent on the specifics
of the voltage, bias, load and gain.

Melon Head 7th June 2010 09:38 PM

So would you say they perform better than lu1014d jfets?
Does this mean there could be an F3 mkII using these devices?


All times are GMT. The time now is 02:25 PM.


vBulletin Optimisation provided by vB Optimise (Pro) - vBulletin Mods & Addons Copyright © 2014 DragonByte Technologies Ltd.
Copyright 1999-2014 diyAudio


Content Relevant URLs by vBSEO 3.3.2