I've got JFETs on the brain for some reason. In the context of an earlier question regarding reversal of the Drain and Source leads on JFETs, I just ran across a part that explicitly mentions the capability of doing so:
http://www.onsemi.com/pub/Collateral/J110-D.PDF
Whether that property is useful to us in audio or not is another question, but if there's one part like that out there it's likely that there are others. At least you know it won't pop if you hook it up backwards.
For some reason I keep getting the idea that it would be fun to build a headphone amp out of parallel J310s. This concept is getting really annoying, as I have other things I need to be doing. I have a perfectly wonderful headphone amp already (simple follower: Motorola MTP7N20E [not unlike the IRF610] with +-25V rails and 50 ohm Caddock for bias...yes, it's overkill, but it sounds bloody marvelous). And yet...and yet...
I hate this hobby.
Grey
http://www.onsemi.com/pub/Collateral/J110-D.PDF
Whether that property is useful to us in audio or not is another question, but if there's one part like that out there it's likely that there are others. At least you know it won't pop if you hook it up backwards.
For some reason I keep getting the idea that it would be fun to build a headphone amp out of parallel J310s. This concept is getting really annoying, as I have other things I need to be doing. I have a perfectly wonderful headphone amp already (simple follower: Motorola MTP7N20E [not unlike the IRF610] with +-25V rails and 50 ohm Caddock for bias...yes, it's overkill, but it sounds bloody marvelous). And yet...and yet...
I hate this hobby.
Grey
Call me old fashioned but I like to connect the components the right way, not a big fan of inverted GC's either, I digress.
Grey,
Instead of changing the way you perceive the same musical piece sometimes is useful to change musical piece altogether: every time I feel like building something I go out and buy a few CDs. It saves money in the long run.
Grey,
Instead of changing the way you perceive the same musical piece sometimes is useful to change musical piece altogether: every time I feel like building something I go out and buy a few CDs. It saves money in the long run.
Casoded MOSFET
Hi!
Would it be possible to build the original JFET amp circuit with only MOSFET instead of the lovoltech JFET???
I can not wait until these JFETS will be available for "mortal" people here in Europe.... 🙂
Greets:
Tyimo
Hi!
Would it be possible to build the original JFET amp circuit with only MOSFET instead of the lovoltech JFET???
I can not wait until these JFETS will be available for "mortal" people here in Europe.... 🙂
Greets:
Tyimo
Re: Casoded MOSFET
Absolutely. Just Cascode a Zen. My original design for the F3
was precisely that, but did not use feedback. After I started
playing with the JFET I decided that the world wanted it with
somw damping factor, and so I put about 10 dB around it for a
DF of 3 or so.
😎
Tyimo said:
Absolutely. Just Cascode a Zen. My original design for the F3
was precisely that, but did not use feedback. After I started
playing with the JFET I decided that the world wanted it with
somw damping factor, and so I put about 10 dB around it for a
DF of 3 or so.
😎
I have been playing with variants on the circuit. For me--and those who have watched me grind slooowly through the R&D process will know that I usually take at least three or four months, minimum, to get to a workable circuit--I have made rapid progress. This partly due to the fact that Nelson's paper dovetailed neatly with something else I had already been working on...it's much easier to dance when there are two. Even if it's coast-to-coast shadow-dancing.
All of this boils down to my usual long-winded way of saying, yes, by all means pursue the MOSFET variation if that intrigues you.
I hope to have more to say on this at a later date. If all goes well (and my usual timing applies), perhaps by Christmas.
Grey
P.S.: 50V is good, but 8W is kinda in a never-never land...neither enough for a true power transistor, nor entirely useful for a small signal transistor. Perhaps a driver (aka VAS) stage?
All of this boils down to my usual long-winded way of saying, yes, by all means pursue the MOSFET variation if that intrigues you.
I hope to have more to say on this at a later date. If all goes well (and my usual timing applies), perhaps by Christmas.
Grey
P.S.: 50V is good, but 8W is kinda in a never-never land...neither enough for a true power transistor, nor entirely useful for a small signal transistor. Perhaps a driver (aka VAS) stage?
The Crystalonics devices look interesting. An array of them could possibly make some nice Class-A type noises with a cascode FET to help them out, but I betcha they are hilariously, even heart-stoppingly expensive (I just looked at the price list, and the CP666 is $14.70 in 1-99 Qty)... J105s (commonly available) have a high IDSS too, and the voltage rating of 25V is no problem using a cascode, but the guys are in a TO-92 package, so you can't get the heat out om them.
powerbecker said:
Yeah, I've seen those. But I believe I've heard that RF JFETs such as these aren't terribly useful in audio circuits (they're intended to be used common gate). Perhaps I've heard wrong. Has anyone used RF JFETs with any success in audio applications?
se
Thanks Mr. Pass for the answer. How do you mean to cascode a Zen? The whole Zen circuit?
Basicly I thought only to replace the lovoltech JFET to an IRFP240 Mosfet in the JFET amp circuit, because I would like to build this amp with no feadback and with lightbulbs....
Tyimo
Use a low voltage high transconductans device instead. At least something like IRFP-054 or IRFP-3703. It will improve performance a lot. (i belive)
Se Mr Pass article about device selection. http://www.passdiy.com/pdf/retofzen.pdf
The gfs is a "more is better" parameter but considerations for the Iq of your design versus the actuall gfs at that Id should be considered. Cascoding affords the ability to run at higher Id's for more gain and linearity without melting the device... I beleive😀 😀 😀
O.K., that part looks interesting, until I see the high C #s. The Cascoding helps deal with the non-linearity there but is'nt the absolute C values still a little high to deal with???
If you run the amp without any feedback from the drain but say a 0,5 ohm resistor between source and ground, I dont think its much of a problem. (When cascoded)
If you use feedback you will probably need to use an input buffer like in the Penultimate Zen.
- Status
- Not open for further replies.