Doesn't the manufacturer sort of put themselves at risk that the chip vendor decides to change the process or ship marginal product ("Nobody is buying our 100 MHz parts anymore, so we will save a few bucks and make and ship 40 MHz parts (that pass only to 45 MHz)")?
-- Joe Leikhim K4SAT "The RFI-EMI-GUY"© "Use only Genuine Interocitor Parts" Tom Servo ;-P
Yep. That's what I would be extremely concerned about. If I ever decided to run a part over its rated limits (speed wise), I would put in hooks, for production testing, to lower its supplies to their minimum and then run those tests at the maximum rated d.u.t. temperature.
Of course, in Larkin's case, the only person that can fire him for non performance and insubordination is his wife.
Bob
-- == All google group posts are automatically deleted due to spam ==
Yes, but that's the 3dB bandwidth, isn't it? i.e the frequency at which the voltage is 1/sqrt2 that of its value at DC. Hence you should still be able to see signals higher than 50Mhz. Plonk a signal generator on it and tell us what you you see ;)
Podcast:
Hey, nowadays I report to The Brat.
John
Sure. Life is risky. Given a product with 200 different parts, two of which are overstressed, there's probably more risk that one of the 198 other parts will be EOL'd.
John
Have you ever been caught in the situation of one of your over-stressed parts being changed by the manufacturer in such a way so as to render it unsuitable for your needs?
I was just going to say, didn't she show an interest in the biz lately and even do a layout? Remember, she may one day be the one deciding which nursing home you guys go to ;-)
-- Regards, Joerg http://www.analogconsultants.com/ "gmail" domain blocked because of excessive spam. Use another domain or send PM.
Happened to another engineer: Bought tons of 5% capacitors for production, went into Hilbert shifters. They always hand-picked the 1% parts out of this lot and used the rest for jelly-bean stuffing. One balmy summer day they found the first shipment that had a hole smack in the middle of the Gaussian tolerance curve. Obviously someone had detected a business opprotunity ... whoops.
-- Regards, Joerg http://www.analogconsultants.com/ "gmail" domain blocked because of excessive spam. Use another domain or send PM.
I've never been caught with any over-spec parts I've used in the past. It's always something else in the design that goes the way of the do-do. And of course it's always the seemingly most innocuous part.
Dave.
-- --------------------------------------------- Check out my Electronics Engineering Video Blog & Podcast: http://www.alternatezone.com/eevblog/
Sure production ends on the part and then they look in the stockroom and find a dusty box of parts, where did they come from? Probably someone else hand picked parts and sent the rest back for restock!
-- Joe Leikhim K4SAT "The RFI-EMI-GUY"© "Use only Genuine Interocitor Parts" Tom Servo ;-P
They couldn't be that brazen unless they'd have gone to great lengths such as re-belting them all. Because the good ones weren't all in front section.
-- Regards, Joerg http://www.analogconsultants.com/ "gmail" domain blocked because of excessive spam. Use another domain or send PM.
blackhead wrote in news: snipped-for-privacy@a7g2000yqk.googlegroups.com:
snipped-for-privacy@s31g2000yqs.googlegroups.c
Anti aliasing filters do not have to be purely low pass filters set below the Nyquist frequency. You can also use band pass filters to digitise a frequency above Nyquist or even HIGHER than the sampling frequency.
You will not get the original signal, but you will get an undersampled signal that still can be used in certain applications such as digital receivers sampling off IF chains.
The important part of the filter design is to have very good rejection of signals outside the designed centre frequency so no signals present at the usual "baseband" are present.
Obviously the ADC must be capable of functioning at the desired input frequency.
It works, and is quite commmon.
Ray
Siemens used to make the CLY5 power mesfet. It was rated for about 9 volts or something but was good to 22 or so. They discontinued it and TriQuint started making them, with a different process. The new ones die hard at 15 volts. Luckily, the product that we used it on never got beyond shipping a couple of prototypes.
Optek keeps changing their laser diodes, so you can't count on their typical performance.
But in general we haven't had much trouble. EOL'd regular parts are a bigger hassle because there are so many more "regular"-use parts in a product.
We do test parts to destruction to see what sorts of margins we have. Some parts don't even survive their datasheet abs max ratings.
John
She is doing all our PCB layouts and managing engineering [1]. And doing all our press releases. Not bad for a 22-yo with a minior in psychology and majors in softball and beer pong.
I asked her a couple of months ago, "Say, do you want this company?" She thought about it for a few hundred milliseconds and said yes.
John
[1] And she does my schematic entry for me. I still draw pencil on blue-grid D-size vellum.
Great. I remember a few years ago when you hinted that there wasn't much hope she'd jump in. For a friend of mine that turned out to be the case. One of their daughters would have the technical knack to run his business but it seems she's not very interested. Similar age.
I do that less and less. Most of the time when a client needs a re-design I fire up the WP and CAD and have at it. Except when at the pool, like 30 minutes ago. Two Clearprint 1000HP-4 vellum pads always at the ready.
-- Regards, Joerg http://www.analogconsultants.com/ "gmail" domain blocked because of excessive spam. Use another domain or send PM.
of
198 1%=20and=20
Not to be too picky, but why would production end on a profitable 5% parts line?
Likely selected out before packaging.
e
I think you can place and wire parts of it by hand to get around some of it, but it would still change with temperature, voltage, etc.
Some Xilinx fpgas have programmable delays in the IOs, 64taps each at
1/64th of a ~200MHz (afair) reference clock. a closed loop keeps the delay constant with temp etc.-Lasse
He meant by a customer who then sent it back. If the manufacturer did that there would be no re-stocking process, they do it right off the conveyor belt.
-- Regards, Joerg http://www.analogconsultants.com/ "gmail" domain blocked because of excessive spam. Use another domain or send PM.
You might do it once. You'll never maintain the mess.
Yes, there is an I/O delay, but it's certainly not that large. There is also delay in the DLLs on that order, but there are a lot of restrictions on those (clocks only).
We've had stuff put onto reels. There are companies that do this. I doubt it would be worth it for capacitors, but maybe. My bet is that it was the manufacturer doing the binning.
ElectronDepot website is not affiliated with any of the manufacturers or service providers discussed here. All logos and trade names are the property of their respective owners.