Is there a proof, that the company ever claimed, that these devices are MOSFET or J-FET ?
I'm pretty sure Nelson Pass is just preparing his next design with it ..
Maybe it's only a mistake by me and farnell:
Attachments
The SiC jfets sound very fine if you pretend that they're pentodes without filaments (and no screen/suppressor grids, either) and run them in a circuit that looks very much like a pentode - pentode partial feedback tube amp with output transformer - check the "SiC Puppy" thread in the Pass forum for one such implementation. I probably will do one more pass on this particular circuit to incorporate a source follower drive to get around the pathetic gate leakage specs for the Semisouth normally-on jfets. Even without that particular refinement, the "SiC Puppy" sounded very fine at the 2013 Burning Amp, driving Variac's speakers ( it sounded very nice, driving my speakers at home, too...).
Last edited:
I understand that they're used in switching circuits but it must be high voltage circuits since for a couple hundred volts silicon MOSFETs are better and much cheaper.
So my question is: what are the switching circuits where more than 1000V are used ?
I have worked with "Off-Line" smps, which means you plug it into the mains, and we always did a universal "worldwide" design. It was designed to operate from 65-265VAC in. All you have to do is change the plug. This would be rectified to almost 400Volts DC and fed to some kind of buck/flyback converter. The switching on and off of 400VDC, as Gyuri said, generates upwards of 600-700Volt spikes that must be dealt with somehow. I could be talking about your phone charger, your laptop charger or your 50" Plasma TV. Charging big batteries or powering your whole house would be another level altogether and raising voltage as the power steps up is appropriate to avoid resistive losses in the system. Industrial equipment also generally operates at higher power and voltage levels.
What were SemiSouth SiC JFETs intended for?
You are all right to one degree or another, especially the remark that it was an "accident" that they made it to the DIY Audio community. Of course, the JFET had the helping hand of Nelson.
The killer ap for SiC at this time is at 1200 V. FLG has put his finger on where SiC unipolar devices have become a "must have" part of BOMs in the server power supply market. The SiC Schottky Barrier Diode goes where no Si Schottky Barrier Diode can go (600 V and much higher) and its incumbent, the Si ultrafast recovery pn diode, is less expensive per part but more expensive (and the solution is too large) per universal AC/DC adapter. Ergo, that is a recurring business that Infineon and Cree have largely divided outside of Japan. SemiSouth was building orders for these parts while still alive.
The same idea was pushing the order books for the SemiSouth SiC JFET, but the displaced part was the Si IGBT. The Si MOSFET can reach higher voltage than the Si Schottky Barrier Diode, but even these reach ridiculous uselessness at 1200 V except for a few captive sockets in thinks like automatic defibrillators. Ever try to use a high voltage Si MOSFET in a hard-switching application? Me has, and it is NEVER as good an experience as using a SemiSouth SiC JFET. You get spoiled using the best real fast.
Where were the big volume design wins a few years ago? (There all over the place now with other companies of course.) Solar inverters. Here's a link to a press release dating from 2009 that suggests why and when SemiSouth's ramp in orders for the JFET began.
New Power Inverter Efficiency Record Set : Renewable Energy News :
I just gave a presentation at a workshop at the University of Texas Austin last Thursday that went through the whole experience.
You are all right to one degree or another, especially the remark that it was an "accident" that they made it to the DIY Audio community. Of course, the JFET had the helping hand of Nelson.
The killer ap for SiC at this time is at 1200 V. FLG has put his finger on where SiC unipolar devices have become a "must have" part of BOMs in the server power supply market. The SiC Schottky Barrier Diode goes where no Si Schottky Barrier Diode can go (600 V and much higher) and its incumbent, the Si ultrafast recovery pn diode, is less expensive per part but more expensive (and the solution is too large) per universal AC/DC adapter. Ergo, that is a recurring business that Infineon and Cree have largely divided outside of Japan. SemiSouth was building orders for these parts while still alive.
The same idea was pushing the order books for the SemiSouth SiC JFET, but the displaced part was the Si IGBT. The Si MOSFET can reach higher voltage than the Si Schottky Barrier Diode, but even these reach ridiculous uselessness at 1200 V except for a few captive sockets in thinks like automatic defibrillators. Ever try to use a high voltage Si MOSFET in a hard-switching application? Me has, and it is NEVER as good an experience as using a SemiSouth SiC JFET. You get spoiled using the best real fast.
Where were the big volume design wins a few years ago? (There all over the place now with other companies of course.) Solar inverters. Here's a link to a press release dating from 2009 that suggests why and when SemiSouth's ramp in orders for the JFET began.
New Power Inverter Efficiency Record Set : Renewable Energy News :
I just gave a presentation at a workshop at the University of Texas Austin last Thursday that went through the whole experience.
Of course, the JFET had the helping hand of Nelson.
Unadulterated altruism, I assure you.
Was googling jfets out of boredom and stumbled across this. Noticed the seemingly familiar part number, remembered this thread, searched ( lazily ) and didn't find it mentioned. Thought I would throw it out there for those qualified to evaluate its meaning ( or lack thereof ) .
http://www.micross.com/pdf/ASJD1200R045.pdf
Micross Components | Silicon Carbide SiC | JFETs
.http://www.micross.com/pdf/ASJD1200R045.pdf
Micross Components | Silicon Carbide SiC | JFETs
He shoots.....and misses the basket completely
Sounds like they'd look neat, looks like we'll never use them for sound. I think what we really need here is a good old fashioned conspiracy aimed at seeing that products are always being manufactured that just so happen to meet our needs and those of larger markets simultaneously.
OR
The whole thing could be "above board" ( boring ) and involve the creation of an organization dedicated to bringing together those who design semiconductors and those who work in the the multitude of audio related industries.
Seems so many good devices have been lost to history. Perhaps there's some way to stem the tide.
( my vote is for both )
Sounds like they'd look neat, looks like we'll never use them for sound.
I think what we really need here is a good old fashioned conspiracy aimed at seeing that products are always being manufactured that just so happen to meet our needs and those of larger markets simultaneously. OR
The whole thing could be "above board" ( boring ) and involve the creation of an organization dedicated to bringing together those who design semiconductors and those who work in the the multitude of audio related industries.
Seems so many good devices have been lost to history. Perhaps there's some way to stem the tide.
( my vote is for both )
I find it a bit sad that there is no further discussion on the "RF LDMOS" topic introduced by "Jon Lord" (some time ago admittedly but I'm reading it only now). Even though the user seems to cultivate a slightly hostile tone of voice I find the topic very interesting. I'm aware of at least one commercial concept using a single powerful MOSFET, the Thorens TEM 3200.
Does anybody know of available DIY concepts that employ such a high power device in a simplistic Class A circuit?
Regards
Sven
Does anybody know of available DIY concepts that employ such a high power device in a simplistic Class A circuit?
Regards
Sven
Completely agree with Jon.
GeneSiC's SJTs offer the best I-V linearity than ANY other transistor made in Silicon Carbide. The latest generation of SiC Junction Transistors (eg. GA10JT12-247) will offer better audio quality than anything else on the market.
Semisouthfan is simply a failed loser. And Gyuri, you should make your opinions based on real data - not hearsay - that's not a mark of a good scientist or a good experimentalist.
GeneSiC's SJTs offer the best I-V linearity than ANY other transistor made in Silicon Carbide. The latest generation of SiC Junction Transistors (eg. GA10JT12-247) will offer better audio quality than anything else on the market.
Semisouthfan is simply a failed loser. And Gyuri, you should make your opinions based on real data - not hearsay - that's not a mark of a good scientist or a good experimentalist.
you know for proverb - put up or shut up ?
arrange these wonderful devices to be available to Greedy Diyaudio Boyz , learn some manners and then we can talk (in decent way) who is a looser , who not**
**we aren't introduced in semiconductor fab. circles dirty laundry ...... and we don't need that .
arrange these wonderful devices to be available to Greedy Diyaudio Boyz , learn some manners and then we can talk (in decent way) who is a looser , who not**
**we aren't introduced in semiconductor fab. circles dirty laundry ...... and we don't need that .