Say Joerg,
I've heard the "must have a second source" line from you before, and I've often been curious... just what do you do for any reasonably complex IC such as a DSP, FPGA, CPU, Ethernet controller, etc.? Since -- other than really old parts -- no one makes second sources.
I'm using the AD620 WAY outside of its temperature range and there are non-linearities that are cropping up.
I've looked and there are several pin compatable drop-ins from Linear and TI(BB) that have very similar specifications.
My question is if anyone knows if these are three independant designs or did the same guy bounce from company to company and carbon copy the same silicon for each?
Hello Mook,
Maybe just a 2nd source agreement. There are companies and engineers (like me...) who are very hesitant to design in single sourced parts. In mission critical apps 2nd source is usually mandatory or you aren't allowed to design it in.
Regards, Joerg
Would you guess that tha die are duplicates of each other of two seperate approaches to achieve the very similar specifications.
Say Analog uses its opamp silicon design to make the in-amp on a single die, but Linear and TI uses their own with different opamp silicon but achieving the same perforance specification.
I know the AD620 for 1X gain has a 1.5% gain reduction at high temperatures (far above datasheet rating). Its flat out to about 10C south of my max temperature. Then the gain drops sharply down -1.5% at max temp (>125C).
Those silicon resistors leaking maybe?
I'm considering testing other "2nd source" parts to see if they "hopefully" do this at higher temperature or (lucky lucky me) don't drift at all due to a different die process. If they are dupes, likely the answer will repeat.
Hello Mook,
No idea on this one. It is all a matter of contracts between them and that could include mask sharing. Those contracts are sometimes necessary when a potential large volume customer made the design-in of a part contingent upon 2nd sourcing. It is amazing to sit in the room when two Goliaths who would normally tear each other apart gather around the same table.
I think Jim (Arizona) would be the expert here. I always did my high temp stuff with discretes.
Regards, Joerg
Page 13 of the Analog Devices data sheet for the AD620 reads as if they thought they'd invented something.
Page 11 of the Linear Technology data sheet for the LT1168 looks very similar .... Page 11 of the Burr-Brown (now Texas Instruments) data sheet INA337 is less specific, but could well be describing the same structure. Note that the INA337 is specified over the automotive temperature range -40C to +125C.
It is unlikely that one designer bounced from company to to company - Linear Technology is in Silicon Valley, California (Milpitas, Ca
95035-7417), Analog Devices is in Massachusetts (Norwood, MA 02062-9106) and Burr-Brown was in Tucson, Arizona before they got taken over by Texas Instruments - Jim Thompson probably knows all about them ...----------------------- Bill Sloman, Nijmegen
Only two possibilities:
(1) Linear and TI are licensed by Analog Devices to replicate the parts.
(2) Linear and TI have their own designs meeting the specification (most likely scenario).
...Jim Thompson
--
| James E.Thompson, P.E. | mens |
| Analog Innovations, Inc. | et |
| Analog/Mixed-Signal ASIC\'s and Discrete Systems | manus |
| Phoenix, Arizona Voice:(480)460-2350 | |
| E-mail Address at Website Fax:(480)460-2142 | Brass Rat |
| http://www.analog-innovations.com | 1962 |
I love to cook with wine. Sometimes I even put it in the food.
I doubt the same IC designer hopped that many companies. He would have had to sign non-disclosure agreements anyway, to protect company B from getting company A's trade secrets -- at least for a year or so.
My advice -- if you have to use the AD620 at a higher temp, go with the AD620S. It cost's a "bundle" but I use it all the time in my line of work (spacecraft/launch vehicle avionics).
Hello Joel,
Sometimes we have to bite the bullet and use them. For example, one of the future designs here may contain an MSP430 which won't have a 2nd source for years or probably forever. There are companies I trust more, some I trust less and some I don't trust at all (and won't design in any single source part from them).
Also, we take a very careful look at where a certain specialty part is used elsewhere. If it's consumer the scrutiny will be even higher because those products can fall from grace in no time and then the components in there will as well. Video or graphics cards are a perfect example here. Once I was very tempted to use a controller for that market. Thank God I resisted. Not only the part but the whole company pretty much vanished.
Other than that, if there is a single source part and there is a way to design the circuitry discrete at same or lower cost it will be designed discrete, with jelly bean parts.
Regards, Joerg
What _is_ pretty likely is that they are all using similar processes, i.e. the "parts" within all the ICs will have similar characteristics such as TCs.
To get different overall specs, you have to go to parts made with different technologies or at least different dopants, etc.
-frank
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.