Yeah, holidays reading material! I'm very much looking forward to it!
On a side note, I've curve traced 40 fqp3n30 and 30 fqp3p20 FETs and got four n-p matches that are reasonably close, the yield is indeed as low as others have said before... Maybe we should somehow pool unmatched FETs that can't be used in other builds.
On a side note, I've curve traced 40 fqp3n30 and 30 fqp3p20 FETs and got four n-p matches that are reasonably close, the yield is indeed as low as others have said before... Maybe we should somehow pool unmatched FETs that can't be used in other builds.
As promised a comprehenive description.
http://xen-audio.com/documents/f5ha/F5-HA Description V1.4.pdf
Most of the circuit related information have already been posted here before, including Spice models.
So for those who want to scratch build, there should be nothing stopping you.
The additional information relates to our particular build shown here and in the Build Thread.
And even then you also now have all the information.
Happy building, and Merry Christmas,
Patrick
http://xen-audio.com/documents/f5ha/F5-HA Description V1.4.pdf
Most of the circuit related information have already been posted here before, including Spice models.
So for those who want to scratch build, there should be nothing stopping you.
The additional information relates to our particular build shown here and in the Build Thread.
And even then you also now have all the information.
Happy building, and Merry Christmas,
Patrick
Not this one http://www.diyaudio.com/forums/anal...uator-new-passive-preamp-100.html#post1390908 ?
(post1000)
(post1000)
Fisrt of all, thank you Patrick for sharing your design.
Secondly, I noticed you changed semis in the cap multiplier schm. Are the Toshibas better than the previous D44/45H11? Can we use the BC550/560C instead of the new BC807/817 just to avoid smd components?
Last but not least, you mentioned a Jfet follower using e.g. a mp of 2SK209. Can you give us an example?
Marry Xmas!
Secondly, I noticed you changed semis in the cap multiplier schm. Are the Toshibas better than the previous D44/45H11? Can we use the BC550/560C instead of the new BC807/817 just to avoid smd components?
Last but not least, you mentioned a Jfet follower using e.g. a mp of 2SK209. Can you give us an example?
Marry Xmas!
Very well spotted. Good to know someone actually read my article in detail.
The Toshiba's have higher hfe, but lower power dissipation.
For the application here, the former outweighs the latter.
You can use BC550/560, but my PCBs are all SMD and 78xx/79xx compatible.
Please refer to the build thread for details.
For source follower, see :
http://m.eet.com/media/1196037/F3x600.jpg
J1 = J2 = 2K209BL MP or LSK389
R2 = R3 = 0R to 10R
Add 220R gate stoppers.
Patrick
The Toshiba's have higher hfe, but lower power dissipation.
For the application here, the former outweighs the latter.
You can use BC550/560, but my PCBs are all SMD and 78xx/79xx compatible.
Please refer to the build thread for details.
For source follower, see :
http://m.eet.com/media/1196037/F3x600.jpg
J1 = J2 = 2K209BL MP or LSK389
R2 = R3 = 0R to 10R
Add 220R gate stoppers.
Patrick
Thank you Patrick. I actually read your article twice
Yes, for a CM a higher hfe is always beneficial
Source follower: R2, R3 = 0-10R. Lower with higher Idss, right?
Unfortunately Mouser run out the 2SK209 BL grade and, apparently, no order pending. Any other reliable source?
Digikey has them in stock, but it's generally pricey and the minimum order for free shipping is also high
Yes, for a CM a higher hfe is always beneficial
Source follower: R2, R3 = 0-10R. Lower with higher Idss, right?
Unfortunately Mouser run out the 2SK209 BL grade and, apparently, no order pending. Any other reliable source?
Digikey has them in stock, but it's generally pricey and the minimum order for free shipping is also high
The source resistor value is a matter of taste.
The higher the value, the lower the bias and the higher the output impedance.
So I use 0R most of the time.
But using 0R means that you positive bias the JFET with a positive output current.
So it depends on the load you are driving.
It is real bad news that BL is not available anymore.
It also seems that DK has a real left, and probably will follow suit.
They are probably not selling enough loose ones to be bothered in the future.
If I were you I would stock up now.
Alternatives are LSK170, LSK389, BF862, 2SK3557, 2SK2394, ......
Patrick
The higher the value, the lower the bias and the higher the output impedance.
So I use 0R most of the time.
But using 0R means that you positive bias the JFET with a positive output current.
So it depends on the load you are driving.
It is real bad news that BL is not available anymore.
It also seems that DK has a real left, and probably will follow suit.
They are probably not selling enough loose ones to be bothered in the future.
If I were you I would stock up now.
Alternatives are LSK170, LSK389, BF862, 2SK3557, 2SK2394, ......
Patrick