Anybody tried LDMOS Linear Power MOSFETS for FM/UHF transmitting stages ?
These are explicitely made for common source operation and have vanishing capacitances. If you get such a beast tamed in a Audio circuit and use a voltage devider around the gate to simulate a V-FET voltage gain (mu), I'm sure you get as close to SET nirvana as possible.
BLF578 (X) from NXP and others from Freescale (Formerly Motorola) and DSI (IXYS) come into my mind ! Up to 1200W and junction temperatures up to 250°C and parasitic capacitances of <30pF are available in this Class.
You don't get this from any commercially available SIT be it Si or SiC !
The only deviation from the absolute minimalistic approach is the voltage devider, which is anyhow much more linear/predictable than any voltage gain mechanism in a Triode/SIT.
One should really think about such a circuit as a Triode/SIT simulator.
If you see it like this the Zen would contain also a SIT without additional feedback on the pull side.
I used btw. such a "hybrid circuit" to replace 2SK77 V-FETs in my Yamaha B-I a few years ago and it worked great, it even sounded better !
The hybrid circuit approximated gm, mu and gate capacitances of the 2SK77, introduced a zero tempco (as the 2SK77 has), that the amplifier itself didn't need any change.
If you want to know more, search "Cold Triode" on Audio Karma .
These are explicitely made for common source operation and have vanishing capacitances. If you get such a beast tamed in a Audio circuit and use a voltage devider around the gate to simulate a V-FET voltage gain (mu), I'm sure you get as close to SET nirvana as possible.
BLF578 (X) from NXP and others from Freescale (Formerly Motorola) and DSI (IXYS) come into my mind ! Up to 1200W and junction temperatures up to 250°C and parasitic capacitances of <30pF are available in this Class.
You don't get this from any commercially available SIT be it Si or SiC !
The only deviation from the absolute minimalistic approach is the voltage devider, which is anyhow much more linear/predictable than any voltage gain mechanism in a Triode/SIT.
One should really think about such a circuit as a Triode/SIT simulator.
If you see it like this the Zen would contain also a SIT without additional feedback on the pull side.
I used btw. such a "hybrid circuit" to replace 2SK77 V-FETs in my Yamaha B-I a few years ago and it worked great, it even sounded better !
The hybrid circuit approximated gm, mu and gate capacitances of the 2SK77, introduced a zero tempco (as the 2SK77 has), that the amplifier itself didn't need any change.
If you want to know more, search "Cold Triode" on Audio Karma .
I'm not absolutely sure, but I think I can remember, that a diyaudio member (possibly from Uzbekistan) has realized such a project and has also reported about this here on diyaudio (either filed by "Solid State" or by "PassLabs".
The flag of this member was in any case that one from this URL:
http://www.welt-blick.de/flaggen/usbekistan2.gif
perhaps this is the topic:
http://www.diyaudio.com/forums/solid-state/183335-modified-follower-99-hf-transistors.html
and this the datasheets of the used high frequency parts:
http://www.rfparts.com/pdf_docs/RD/rd100hhf1.pdf
http://www.radio-kits.co.uk/radio-related/Linear_PA/ANVHF027B(RD100HHF1 PP).pdf
and here more such devices:
FINALI HI POWER RF. - radioassistances jimdo page!
Attachments
Last edited:
There are several commentaries on using the voltage divider to get the
"triode" character. It appears to be most commonly referred to as a
Schade circuit. The voltage divider has also been used to linearize the
character of a fet operated in the linear region.
I seem to recall lifting this off of this site:
Attachments
Static Induction Transisors: a US (hi)story
I’m glad to see people showing interest in Static Induction Transistor (SIT) audio amps.
IMHO this is a device with unusually good ‘potential’ but which was unjustly neglected!
I’ve been designing SITs since 1978, mostly for HF and RF use and I have over 30 patents issued for this work.
In 1974 we were approached by Kirkwood Rough (some of us who are ‘older’, may remember the now defunct Kirkwood Amplifiers?) who asked us to design SITs as vacuum triode replacements. He wanted to deploy a ‘low cost’ HiFi power audio amp line based on these new SITs.
We secured our work with US Pat. 5648664, BIFET vacuum tube replacement structure.
We displayed our SIT audio amps at various venues and to many ‘august’ audio experts. In the end, our effort failed, in my opinion, due to the price collapse of vacuum tubes in the mid 1990’s. We could not compete as, at the time, the marked was flooded by ECC83s and the like, sold for less than $1, by companies like Iskra, Tesla, Tungsram.
I hope this info adds a little perspective to this thread!
Good vibes to all,
AIC.
I’m glad to see people showing interest in Static Induction Transistor (SIT) audio amps.
IMHO this is a device with unusually good ‘potential’ but which was unjustly neglected!
I’ve been designing SITs since 1978, mostly for HF and RF use and I have over 30 patents issued for this work.
In 1974 we were approached by Kirkwood Rough (some of us who are ‘older’, may remember the now defunct Kirkwood Amplifiers?) who asked us to design SITs as vacuum triode replacements. He wanted to deploy a ‘low cost’ HiFi power audio amp line based on these new SITs.
We secured our work with US Pat. 5648664, BIFET vacuum tube replacement structure.
We displayed our SIT audio amps at various venues and to many ‘august’ audio experts. In the end, our effort failed, in my opinion, due to the price collapse of vacuum tubes in the mid 1990’s. We could not compete as, at the time, the marked was flooded by ECC83s and the like, sold for less than $1, by companies like Iskra, Tesla, Tungsram.
I hope this info adds a little perspective to this thread!
Good vibes to all,
AIC.
more on SIT (SST)
Great background info on SIT from JL.
Here is a little bit more. I hope it will add value and not clutter to this site’s audience!
SIT’s were made on Silicon since early 1970’s. The name, indeed, originated from J. Nishizawa and can be construed as a (somewhat) incorrect translation. The SIT is a Junction Field Effect Transistor (JFET), as compared to most common FETS that are MOS devices.
Besides us at MWT Inc. in US, practically only Sony and Yamaha deployed ‘working’ Si SITs. The first commercial SITs, however, came from Tohoku, Nishizawa’s company.
We’ve been making various HF SITs on SI since 1978. Our devices, although also VFETs (Vertical Channel FETs) were somewhat different, due to an original change of the field distribution in the channel. Some 30 US patents were issued for this work. Most of the applications work was for RF and HF amplifiers. We applied to have ours trademarked as Solid Sate Triodes (SST)! Those interested can find numerous technical SIT and SST article by ‘Goggling’ for them.
One feature that was not publicized enough is the SITs extremely good linearity. It’s IP3 performance exceeds that of any other power semiconductor device. The downside is that the SIT’s RF power gain is relatively low (about 10dB), lower than that of current RF LDMOST. The SIT/SST is also the most ‘rugged’ power semiconductor structure, being able to withstand huge energy transients.
In the last years, ‘SITs’ have experience a resurgence of sorts, as they are easier to fabricate on SIC or GAN that any other FET or Bipolar Junction Transistor (BJT) structures.
I’m encouraged to see some interest in SIT/SST (re)surfacing on this web site and, if anyone is interested, I can (try to) provide additional background.
Reliable audio power to all,
AIC.
Great background info on SIT from JL.
Here is a little bit more. I hope it will add value and not clutter to this site’s audience!
SIT’s were made on Silicon since early 1970’s. The name, indeed, originated from J. Nishizawa and can be construed as a (somewhat) incorrect translation. The SIT is a Junction Field Effect Transistor (JFET), as compared to most common FETS that are MOS devices.
Besides us at MWT Inc. in US, practically only Sony and Yamaha deployed ‘working’ Si SITs. The first commercial SITs, however, came from Tohoku, Nishizawa’s company.
We’ve been making various HF SITs on SI since 1978. Our devices, although also VFETs (Vertical Channel FETs) were somewhat different, due to an original change of the field distribution in the channel. Some 30 US patents were issued for this work. Most of the applications work was for RF and HF amplifiers. We applied to have ours trademarked as Solid Sate Triodes (SST)! Those interested can find numerous technical SIT and SST article by ‘Goggling’ for them.
One feature that was not publicized enough is the SITs extremely good linearity. It’s IP3 performance exceeds that of any other power semiconductor device. The downside is that the SIT’s RF power gain is relatively low (about 10dB), lower than that of current RF LDMOST. The SIT/SST is also the most ‘rugged’ power semiconductor structure, being able to withstand huge energy transients.
In the last years, ‘SITs’ have experience a resurgence of sorts, as they are easier to fabricate on SIC or GAN that any other FET or Bipolar Junction Transistor (BJT) structures.
I’m encouraged to see some interest in SIT/SST (re)surfacing on this web site and, if anyone is interested, I can (try to) provide additional background.
Reliable audio power to all,
AIC.
Indeed very nice !
Yamaha and Sony also marketed the amplifiers as VFET or V-FET amplifiers, this is strange but also understandable.
The name "Static Induction Transistor" was certainly created by a genius in Physics but not by a good English speaker.
There is certainly nothing static inside, nor is electromagnetic induction taking place in the traditional sense (But Nishizawa probably already envisioned GHz-Applications ..).
He probably wanted to say "Electrostatic Field Induced Voltage Transistor" "EFIVT" but ended up with "SIT". I'm sure the marketing people at Sony and Yamaha understood the principle quite well, but spoke a better English,and thus came to the point.
Nishizawa also wasn't really the inventor of the SIT, this can rather be attributed to Julius Edgar Lilienfeld, who filed a patent on the idea in the late 1920ies, just about 20 years before some future Nobel Price winners at Bell labs discovered the bipolar transistor, by trying and failing to implement Lilienfeld's idea ...
Strange also that patents expire exactly after 20 years, isn't it ?
What's wrong with the following sentence:
"SITs are a paticular breed or Power J-FETs like Triodes are a particular breed of Power Pentodes"
Jacco, we should get in contact, V-FET-lovers should unite !
Sorry for robbing this thread ! I get the V-FET blues from time to time ...
Yes, I was playing with these RF RD100HHF1 transistors quite successfully, finally I preferred a sound of this integrated amp
http://www.diyaudio.com/forums/solid-state/189044-integrated-no-nfb-son-follower.html
This amp has overtaken good and expensive tube SE 300B amp, and I expected that the DIY story is close to being completed for me. I even shunted the input relay selector in the tube amp, for making it even more transparent, but no, this tube amp (nearly 10K bucks) is remaining without use.
However, DIY story is not finished, to my "regret" the amp with RF jFETs at the output is even better
http://www.diyaudio.com/forums/solid-state/197493-jfet-amp-current-nfb.html
In the last schematics, there is near 10dB of current NFB, but all measures have been taken to make the NFB resistive divider short and non-inductive.
Do these SIT amps sound like JFET Common-Source amps?In any case, what is of interest to me is that SITs are just about the ideal
sort of part for the Common-Source mode minimalist no feedback
amplifiers that I find interesting. They beat the crap out of every other
single-transistor amplifier I have made, and you can't even try this with
triodes.
Everyone who has heard them so far has been pretty amazed, and some
of these people are not fans of little simple amplifiers.
I guess they would sound a little crisp or it that totally off the mark?
It would be so cool if we can get our hands on a couple!
Regular power Jfets are more like lower distortion Mosfets. SITs are a
different animal, with a quite different characteristic curve.
very interesting informations (also in the follow post) - thank you.
I don't understand this term:This thread is almost causing me physical pain. I need to build an odd-looking amplifier with mystical devices of awesomeness and glowing lights and a "Warning, HOT" sign on it. Where is that group buy already?
with mystical devices of awesomeness
Very good work (both threads) and good informations - thank you. Anywhere you have also investigeted various electrolytic caps and you have find out a cheap model with very good quality. Please let me know this thread - thanks in advance therefore.
For me the most important question is follow:
What is the perception difference in sound character by a listening test between an ordinary MOSFET (like IRFP 150 or 250) with a great amount of input-, output and reverse capacitances to all this here mentioned kinds of FET's/SIT's both in an one-stage power follower like Andrea Ciuffoli's topology and in an common source topology like the ZEN var. I ??
The kind of the influence through the unwanted capacitances must be actually very different in this two basic topologies. And also the different transfer curves. Who have heard the differences by a listening comparison test - at best by the use of a high resolution full range speaker ?
Last edited:
- Status
- This old topic is closed. If you want to reopen this topic, contact a moderator using the "Report Post" button.