Pass Labs, Ayre and Parasound can consolidate their efforts to keep jfets available for themselves...the extras can be made available to DIYers😉...Yeah, why not?!
EUVL,
point I was making is that because of limited payback, I doubt there will be new JFETS for audio coming along. Also, as you know, the pressure to maximise revenue per square foot in the fab is huge, so anything that can be done to improve this gets done (e.g. move from 6 inch to 8 inch, make more expensise/profitable parts). Even process re-qualification trials because of equipment changes are used as inflection points to cut declining or non performing products from the portfolio and make space for other stuff.
I agree with your other comments.
I think Linear probably run their parts in someone else's fab (vishay? maybe someone knows this). But keeping these alive will also be tough for them and might be part of th e reason we have not yet seen the devices promised quite some time ago - they are jus t having to stick with existing processes.
point I was making is that because of limited payback, I doubt there will be new JFETS for audio coming along. Also, as you know, the pressure to maximise revenue per square foot in the fab is huge, so anything that can be done to improve this gets done (e.g. move from 6 inch to 8 inch, make more expensise/profitable parts). Even process re-qualification trials because of equipment changes are used as inflection points to cut declining or non performing products from the portfolio and make space for other stuff.
I agree with your other comments.
I think Linear probably run their parts in someone else's fab (vishay? maybe someone knows this). But keeping these alive will also be tough for them and might be part of th e reason we have not yet seen the devices promised quite some time ago - they are jus t having to stick with existing processes.
BTW, just in case my post is misunderstood, mosfets, especially power mosfets, will be around for decades to come - main application is power supplies. These devices are in huge, growing demand - not to be confused with JFETS.
Linear Systems is as close to a home brew transistor company as you will find. It started as a dealer in used Semi equipment. The guy running it discovered he needed to show that the stuff worked to sell it so he started a processing line using the older equipment. (The story on their web site is a little different but its hard to sell semi's if people think you are a small player).
Then, Silicon Valley being as small as it is he managed to get the masks and process from several local companies who were discontinuing obsolete parts. I think he uses a 4" or possibly a 2" line so he doesn't need to make many parts at a time. Interfet also uses a 2" line.
For a major vendor like National or Fairchild running a small line is not very practical but for these smaller players it can work quite well. And there is a significant and very profitable commercial/industrial market for their products. Places like down-hole in the petroleum industry where money is literally no object and the need is real. Or space craft or IR vision equipment.
I have bought full wafers from Interfet in the past optimized for my needs and I see no reason why that would not be true today. Just very expensive.
Then, Silicon Valley being as small as it is he managed to get the masks and process from several local companies who were discontinuing obsolete parts. I think he uses a 4" or possibly a 2" line so he doesn't need to make many parts at a time. Interfet also uses a 2" line.
For a major vendor like National or Fairchild running a small line is not very practical but for these smaller players it can work quite well. And there is a significant and very profitable commercial/industrial market for their products. Places like down-hole in the petroleum industry where money is literally no object and the need is real. Or space craft or IR vision equipment.
I have bought full wafers from Interfet in the past optimized for my needs and I see no reason why that would not be true today. Just very expensive.
1audio, thanks for the feedback.
Do you have any view on why the promised NP matched pairs from Linear have not materialised? Do linear do any development or just work off existing mask sets and processes (sounds like it from your input).
Do you have any view on why the promised NP matched pairs from Linear have not materialised? Do linear do any development or just work off existing mask sets and processes (sounds like it from your input).
Bonsai said:
I think Vishay just buys up companies to pick their low-hanging fruit and discontinues any lower demand components. Adcom used to use lots of J555 CC JFETs to bias the op amps in Class-A (output to V- rail), but Vishay has discontinued the Siliconix CC parts. I hope Linear is as successful as the sound like they want to be.
I would think tubes are cheaper to set up for small batches. You don't need Sealex machines for exhaust-seal automation. M. U. does quite well in their niche products, making tubes by hand. It's all mechanical.
Chuck Hansen
I would like to bring out an important topic that relates to high end preamps, including Parasound, Ayre, and CTC.
This is the regarding the differences between relays and Shallco switches.
I use Shallco in my best designs, quality relays in my midpriced designs. Why?
This is the regarding the differences between relays and Shallco switches.
I use Shallco in my best designs, quality relays in my midpriced designs. Why?
If the diodes only see DC (excepting when the relays switch), why have a cap? In a power supply, yes. To dump the backlash from a relay coil...I'm not sure I follow. Yes, there will be a transient, but it will only be during the changing of program source (assuming relays are used to change inputs) or when the volume is being changed (nobody listens critically whilst changing the volume).
If it is a problem, then consider a fast/soft diode.
Grey
If it is a problem, then consider a fast/soft diode.
Grey
This is not really the point. What is important is why silver on silver switches sound different than gold on gold relays.
It is one of the points, as it changes coupling of unwanted signals from coil circuit to signal circuit. It is oversimplifyied to consider it just DC, indicating to no experience with the issue.
I design relay controlled preamps, as a rule. It is the contacts that I am considering important here, once the relay voltage is stable and RF free.
Every kind of contact has its contact phenomenae, i.e. voltage drop with shape dependent on current passing thru. Not only contact material is important, but also contact environment. Excellent results are obtained in vacuum and SF6.
Have you read Holm's book on contacts? What about JH Whitley's AMP articles on contact theory? Better prepared, than unprepared.
I used to work for 5 years as a research worker in the field of current interruption and contact/arc phenomenae.
john curl said:
Well, pure silver is the best conductor but can sulfidize if air can get to the contacts. Silver contacts are almost always pure metal throughout. Pure gold won't oxidize (or any other ...idize), but the gold flash or plating used in most relays is very thin and can be blown off by a high enough current through the contacts. Gold is only useful for "low-level" signals. It is also subject to mechanical wear over time if the relay is used frequently (more than 50,000 operations is the number I used in aerospace). I would think that line-level audio signals are too high to be accurately characterized as "low-level" and this could erode the gold over time. I'll have to check on the official definition of "low-level".
Good relays have a contact wiping action induced by over-travel of the actuator, so silver contacts could be kept clear of any films that form. Rotary switches are much better in this regard as the horizontal wiping of the movable contact across the fixed contacts is a superior film clearing feature.
BTW, cadmium oxide is added to the silver for power contactors to prevent the contact erosion that would otherwise be caused by interrupting high currents.
Best Regards,
Chuck Hansen
- Status
- Not open for further replies.