Thats awesome. at least they didnt erase the marksHere's a story for you. We were buying BJT output devices from a major semi
manufacturer (you can guess who, but they no longer make transistors and
it was forty years ago). We paid a premium to have all devices tested for
secondary breakdown as we had seen some problems with that, and we also
tested them in-house. We found a significant percentage breaking down
at voltages way below spec, so clearly they were either not testing them or
shipping them anyway. We called in the factory rep who was a real d**k
and after much discussion they agreed to replace them. We shipped the
bad parts back.
Later we received a batch of replacements, and found that they had our
felt pen test marks on them.
Been there and done that too. I have made it a policy whenever the need arises to return product, I make a small mark or note some small cosmetic defect that will provide positive identification. Surprising how often I have that deja vue sensation.
A co-worker once had a inexplicable habit of placing components that he had replaced back in the supply cabinets, I suppose they were magic healing boxes. Maybe that bizarre notion is more widespread than we know.
I just found that ROHM is into SIC as well, if there are some alternatives to the unobtainium please share and the race will be on to hoard them as well!
A co-worker once had a inexplicable habit of placing components that he had replaced back in the supply cabinets, I suppose they were magic healing boxes. Maybe that bizarre notion is more widespread than we know.
I just found that ROHM is into SIC as well, if there are some alternatives to the unobtainium please share and the race will be on to hoard them as well!
Prudent business practices by you at THRESHOLD. It pre-empted a quagmire of failed and returned amps plus possible ill reputation. Your scientific and business reputation is pristine clean due to all that.Here's a story for you. We were buying BJT output devices from a major semi
manufacturer (you can guess who, but they no longer make transistors and
it was forty years ago). We paid a premium to have all devices tested for
secondary breakdown as we had seen some problems with that, and we also
tested them in-house. We found a significant percentage breaking down
at voltages way below spec, so clearly they were either not testing them or
shipping them anyway. We called in the factory rep who was a real d**k
and after much discussion they agreed to replace them. We shipped the
bad parts back.
Later we received a batch of replacements, and found that they had our
felt pen test marks on them.
Thanks, amp-guy. Too kind for sure, but
I have wondered about the Cree MOSFET from an audio perspective myself. Attached is a LTSpice model of the CMF20120 MOSFET distributed by Cree and evaluated by one of my students in the second attachment for validity against the data sheet. The capacitance and transfer characteristics match up well which are of first order importance to linearity and performance. The output curves computed by the model are less valid which, while not so important to the intended customers of the MOSFET, definitely plays a role in the compensation of the square law dependence of the transfer curve. Hence the model may not tell you all you want to know about using it for audiophile work in lieu of a SemiSouth JFET. Getting a Cree MOSFET is easy so I assume folks here have already checked them out. Love to hear about experiences.
On another note, I'm putting the finishing touches on an article containing my thoughts on spotting SemiSouth fakes. I'm going to post the PDF in a new thread, which I'm planning to call the "SemiSouth boiler room." I expect to make the first post any day now. I look forward to feedback from the community and ideas for new topics.
I have wondered about the Cree MOSFET from an audio perspective myself. Attached is a LTSpice model of the CMF20120 MOSFET distributed by Cree and evaluated by one of my students in the second attachment for validity against the data sheet. The capacitance and transfer characteristics match up well which are of first order importance to linearity and performance. The output curves computed by the model are less valid which, while not so important to the intended customers of the MOSFET, definitely plays a role in the compensation of the square law dependence of the transfer curve. Hence the model may not tell you all you want to know about using it for audiophile work in lieu of a SemiSouth JFET. Getting a Cree MOSFET is easy so I assume folks here have already checked them out. Love to hear about experiences.
On another note, I'm putting the finishing touches on an article containing my thoughts on spotting SemiSouth fakes. I'm going to post the PDF in a new thread, which I'm planning to call the "SemiSouth boiler room." I expect to make the first post any day now. I look forward to feedback from the community and ideas for new topics.
The CMF10120 is $16 and they are in stock at Digikey.
The only potential issue I note is that the Crss increases quite a bit at Vds
below 30V. This means that you either want to drive it with a low capacitance
or keep the Vds high.
Nelson brings up a good point that power switching folks and audiophiles can agree on. The SiC MOSFET is more capacitance heavy than the SiC trench JFET. This is a result of the fact that the SemiSouth JFET has better figures of merit on specific on-resistance than is possible with current lateral channel SiC MOSFETs. This applies to the Rohm MOSFET too. A trench MOSFET by Rohm which is not commercially released yet should (eventually) benchmark better with the trench JFET.
Ideally, SemiSouth or assignee may/could do elemental analyses on these two devices as follows. Scrape some of the housing [epoxy?] , some of the ink, and some metal from the leads; and subject the scrapings to Scanning Electron Microscopy [SEM] for a fingerprint of which elements are present and their extent. May also do an X-Ray scan for additional signatures. The harder part is to determine the probability that the analyses match or not; given or granted that SemiSouth had already collected a solid reference for comparative analyses.
I made a mistake above. The test method is XRF or X-Ray Fluorescence instead of SEM.
The industry of electronic devices has moved away from using the element Lead [Pb] in their construction. A reputable supplier of devices will not bend/break this rule; unlike one [with maybe a good probability] who is illegaly making its functional copies. Analytical Chemistry to the rescue. There are kits which test for surface Lead as on the leads of the device. The web is full of them at less than $20.
Best regards.
labjr. I hope this helps answer some of your questions [from the web]. The site www.semisouth.com is not responsive, and draws a blank page.
24 October 2012
SemiSouth to be closed; 90 staff to be laid off
SemiSouth Laboratories Inc of Starkville, MS, USA, which designs and manufactures silicon carbide (SiC) devices for high-power, high-efficiency, harsh-environment power management and conversion applications, is to close down, according to a report by the Starkville Daily News.
SemiSouth was founded in 2001 by former Mississippi State University (MSU) faculty member Jeffrey Casady and current MSU electrical and computer engineering professor Michael Mazzola, and has since been housed at the Thad Cochran Research, Technology and Economic Development Park near MSU.
MSU director of university relations Sid Salter said that, while MSU does not have a formal business relationship with SemiSouth, the university assigned the technology’s intellectual property rights to non-profit organization MSU Research and Technology Corp, giving the university a 2% stake. As owner of the Cochran facility, MSU has managed SemiSouth’s lease of its portion of the building (yielding $300,000 in revenue annually).
In October 2010, in order to drive the expansion of its SiC fabrication facility, Power Integrations Inc of San Jose, CA, USA (which supplies high-voltage integrated circuits for compact, energy-efficient power conversion in electronic products for AC-DC, DC-DC and LED lighting applications) made a $30m strategic investment in SemiSouth (including an equity investment, a technology license and other financial commitments). The firms also collaborated on driving the adoption of SemiSouth’s SiC technology, aiming to speed the development of efficient power conversion devices for applications including solar and wind inverters as well as hybrid/electric vehicles.
In second-quarter 2011, to expand production capacity, SemiSouth installed an Aixtron AIX 2800G4 WW chemical vapor deposition (CVD) reactor in 10x100mm- and 6x150mm-wafer configuration for the production of power SiC junction field-effect transistor (JFET) and Schottky barrier diode (SBD) devices. More recently, just this March, SemiSouth announced its second major capacity expansion (worth $18m) within 18 months, intended to add 50% more capacity to meet demand for SiC power JFETs and power diode products from the solar inverter and industrial power supply markets.
However, now, Power Integrations says that its third-quarter 2012 results will include a pre-tax charge of about $60m from the closure of SemiSouth (resulting in a net loss). “The expected closure of SemiSouth is disappointing, but reflects the challenges and risks inherent in the quest for disruptive technologies,” says Power Integrations’ president & CEO Balu Balakrishnan. “Our strategic direction remains unchanged, and we continue to invest in promising technologies to expand our market opportunity within the realm of high-voltage power conversion,” he adds.
MSU vice president for research David Shaw hopes a solution can be developed to retain employees among the 90 affected. “We hope we can be a resource to some of these families looking for other job opportunities,” adds Jennifer Gregory, VP for tourism and development at the Greater Starkville Development Partnershipj
- Status
- This old topic is closed. If you want to reopen this topic, contact a moderator using the "Report Post" button.