Don't know where to put this, I thought it could be interesting.
This marketing document from Sanyo talks a little about the FBET/LSBT process, the source of some amazing devices manufactured there.
http://home.arcor.de/thomasfetzer/Sanjo/SY103A_e.pdf
Following the drawings and the short explanation, it's clear that this process uses shallow (probably ion implanted) local thin base regions, and polysilicon emitters. This is a seven to eight mask process, compared to three to four in a regular bipolar discrete shop. No wonder the performances and also no wonder why these devices are expensive.
The FBET/LSBT process was completed in 1981. For a process to survive in production for almost 30 years, that's really a long time!
The bad news is that if Sanyo is going to discontinue it's line of discretes, I don't see any other discrete manufacturer picking up and filling the gap. It's very expensive and unlikely to have a good quick ROI. It's much more complicated to second source the Sanyo bipolar devices than (e.g.) Toshiba's 2SK170/2SJ74
This marketing document from Sanyo talks a little about the FBET/LSBT process, the source of some amazing devices manufactured there.
http://home.arcor.de/thomasfetzer/Sanjo/SY103A_e.pdf
Following the drawings and the short explanation, it's clear that this process uses shallow (probably ion implanted) local thin base regions, and polysilicon emitters. This is a seven to eight mask process, compared to three to four in a regular bipolar discrete shop. No wonder the performances and also no wonder why these devices are expensive.
The FBET/LSBT process was completed in 1981. For a process to survive in production for almost 30 years, that's really a long time!
The bad news is that if Sanyo is going to discontinue it's line of discretes, I don't see any other discrete manufacturer picking up and filling the gap. It's very expensive and unlikely to have a good quick ROI. It's much more complicated to second source the Sanyo bipolar devices than (e.g.) Toshiba's 2SK170/2SJ74
Well, I am a little aware of the development concerning power mosfets, but that's for switching and I was more pointing in the small signal or in general linear direction...
That simple low-noise jfets (K170) are a thing of the seventies I can understand, but that the FBET-process dates also back to 81 I wouldn't have expected.
Maybe you could shortly elaborate what I'm missing?
Have fun, Hannes
That simple low-noise jfets (K170) are a thing of the seventies I can understand, but that the FBET-process dates also back to 81 I wouldn't have expected.
Maybe you could shortly elaborate what I'm missing?
Have fun, Hannes
syn08 said:Don't know where to put this, I thought it could be interesting.
This marketing document from Sanyo talks a little about the FBET/LSBT process, the source of some amazing devices manufactured there.
http://home.arcor.de/thomasfetzer/Sanjo/SY103A_e.pdf
Following the drawings and the short explanation, it's clear that this process uses shallow (probably ion implanted) local thin base regions, and polysilicon emitters. This is a seven to eight mask process, compared to three to four in a regular bipolar discrete shop. No wonder the performances and also no wonder why these devices are expensive.
The FBET/LSBT process was completed in 1981. For a process to survive in production for almost 30 years, that's really a long time!
The bad news is that if Sanyo is going to discontinue it's line of discretes, I don't see any other discrete manufacturer picking up and filling the gap. It's very expensive and unlikely to have a good quick ROI. It's much more complicated to second source the Sanyo bipolar devices than (e.g.) Toshiba's 2SK170/2SJ74
I doubt any these transistors have been used in audio anyway - the sanyo types. So a non issue to me. Remember ring emitter Q's was in production in 1984 and could challange some mosfets in certain areas easily but to no avail.
Regardless the 2SJ74 is a disgrace to the audio community with respect to very high non linear capacitance and high gm.
Cheers
Kevin
syn08 said:I don't see any other discrete manufacturer picking up and filling the gap. It's very expensive and unlikely to have a good quick ROI. It's much more complicated to second source the Sanyo bipolar devices than (e.g.) Toshiba's 2SK170/2SJ74
The spin was almost totally CRT displays a retrogression that has pleasantly been absent from the videophile crowd. I guess in the end BIG wins.
> Where does the Sanyo 2SK222 JFET fit? Its figure of merit sticks out as far better than the 2SK170 with respect to gm/Cgs.
I remember Kaneda saying in his book that 2SK117 is the best as far as gm/Cgs is concerned among JFETs.
I don't think he uses Sanyo, and I think 2SK223 is even better than 2SK222, if you ignore differences in Idss.
But of course BF862 is the clear winner. I love them, and cannot thank you enough for the tip.
> AFAIK they never made a P version.
That is precisely the problem, and why Toshiba is still DIYers' favorite.
Patrick
I remember Kaneda saying in his book that 2SK117 is the best as far as gm/Cgs is concerned among JFETs.
I don't think he uses Sanyo, and I think 2SK223 is even better than 2SK222, if you ignore differences in Idss.
But of course BF862 is the clear winner. I love them, and cannot thank you enough for the tip.
> AFAIK they never made a P version.
That is precisely the problem, and why Toshiba is still DIYers' favorite.
Patrick
Actually Sanyo did make complementary P and N channel Jfets, they had duals too, phased out in the early 80s. I had the datasheets somewhere on my backups, will try to locate them.
I came upon them through a friendship with a retired engineer that build audio amps and preamps when I was still just a teenager around 1987. He had boxes full of these parts. He has passed away but his wife is still alive and unfortuneatly doesnt want to part with any of her late husbands stuff yet, she is 92 now, fortuneatly I am friends with the 2 daughters as it would be a total waste for the parts to end up on the rubish heap. 5 Years later and I also started using exclusivelySanyo parts for building amps and preamps, they were easy to obtain where I grew up except for the Jfets. Sanyo kept low profile in those days, but already had excellent small and medium signal devices, parts that most are in awe about today and many that dont even know they exist. The west was trying to curb Japanese industrial might so alot of these parts were unknown and delibertly kept from their markets. We on the other hand because of sanctions imposed could only get Japanese electronic products. The only audio manufacturer using exclusively their parts from those years and earlier were Rotel and they still do so.
I came upon them through a friendship with a retired engineer that build audio amps and preamps when I was still just a teenager around 1987. He had boxes full of these parts. He has passed away but his wife is still alive and unfortuneatly doesnt want to part with any of her late husbands stuff yet, she is 92 now, fortuneatly I am friends with the 2 daughters as it would be a total waste for the parts to end up on the rubish heap. 5 Years later and I also started using exclusivelySanyo parts for building amps and preamps, they were easy to obtain where I grew up except for the Jfets. Sanyo kept low profile in those days, but already had excellent small and medium signal devices, parts that most are in awe about today and many that dont even know they exist. The west was trying to curb Japanese industrial might so alot of these parts were unknown and delibertly kept from their markets. We on the other hand because of sanctions imposed could only get Japanese electronic products. The only audio manufacturer using exclusively their parts from those years and earlier were Rotel and they still do so.
Regarding the Sanyo FBET process :
In the near future I will "experiment" by attempting to use 2SA1403/2SC3597 in the cascode sections on a Hafler P3000 amp.
One of the original MPSA06/MPSA56 pairs have apparently become temperature and mechanically sensitive. It's not a cold solder joint.
This will be interesting. The intended replacements are both faster and more linear than the originals. (faster inviting instability?)
Of course I will substitute a 100W lamp for the fuse on initial power-up.
If HF/VHF oscillations ensue I will counter with miniature ferrite beads on the emitters.
I have new MPSA06/MPSA56's just in case this doesn't come out as planned.
Good Day All
Rickey.
In the near future I will "experiment" by attempting to use 2SA1403/2SC3597 in the cascode sections on a Hafler P3000 amp.
One of the original MPSA06/MPSA56 pairs have apparently become temperature and mechanically sensitive. It's not a cold solder joint.
This will be interesting. The intended replacements are both faster and more linear than the originals. (faster inviting instability?)
Of course I will substitute a 100W lamp for the fuse on initial power-up.
If HF/VHF oscillations ensue I will counter with miniature ferrite beads on the emitters.
I have new MPSA06/MPSA56's just in case this doesn't come out as planned.
Good Day All
Rickey.
- Status
- This old topic is closed. If you want to reopen this topic, contact a moderator using the "Report Post" button.