MEMS

Dunno. I've had good success so far with MEMS devices. At one point in the early 1990's the company I was working for was all hot about using quartz MEMS gyros instead of rotating gyros for stabilizing imaging platforms -- it took some work to convince management that the salesmen weren't actually telling the whole truth.

--

Tim Wescott 
Wescott Design Services 
http://www.wescottdesign.com

Some things, like oscillators and DLPs, seem fine. MEMS relays sounded like a great idea, but seem to have reliability problems.

I'm "cautiously optimimistic" about things like this:

Oops, it appears to be obsolete.

--

John Larkin         Highland Technology, Inc 
picosecond timing   laser drivers and controllers 

jlarkin att highlandtechnology dott com 
http://www.highlandtechnology.com

Yeah, I'm noticing that! So, are the parts being withdrawn because

*better* parts are now available (though the fact that the earlier part is no longer ALSO offered suggests it may have been ruled "undesireable" by The Market)? Manufacturing problems? Poor acceptance? [I am not fond of bleeding edge in any technology!]

Any particular *types* of devices? Or, "in general"?

I was looking at some accelerometers, microphones, etc. Primarily for the size reduction. But, if they "aren't worth a damn" then I'd resort to more conventional offerings...

Tiny MEMS relays would be fabulous high frequency switches, and have been just around the corner for 15 years or so. I think the problem is contact sticking. Maybe it's some scaling issue, like stickiness vs spring force as things scale down in size.

MEMS oscillators can in theory be cheaper than quartz crystal oscillators and (almost) as good, but that's not reality, so far. Even silicon oscillators (all CMOS, no moving parts) have a hard time competing with crystals.

--

John Larkin         Highland Technology, Inc 
picosecond timing   precision measurement  

jlarkin att highlandtechnology dott com 
http://www.highlandtechnology.com

I'm not interested in "clever". Or, necessarily "cheap(er)". Size is driving my interest. But, reliability (in reasonably harsh -- from a mechanical standpoint -- environments) is a big concern. With long expected product lifetimes, I simply can't afford to replace units that fail in the field. Nor, field complaints from users who *expected* longer useful life. I.e., products (technologies) with longer manufacturing histories might be a safer bet (?)

MEMS accelerometers are well esablished and not much alternatives anyway. Non-MEMS accelerometers are bulky and pricey, and not much more reliable anyway.

They can build MEMS structures more or less reliable, but just a matter of cost.

MEMS accelerometers have been around for a long time. Cell phones, tablets take a beating and keep on ticking. The new GA glass cockpits use them in the AHRS. I would not consider them bleeding edge any more. I think MEMS microphones were in the original iPad.

--
Chisolm 
Republic of Texas

But, do they *really*? How often do you drop your phone? And, over the course of how *few* years? I suspect the phone manufacturers wouldn't consider that sort of abuse in their design specifications...

If you dropped it several times a day over the course of a decade or more, would those components still be viable (I'm sure the phone itself will have given up the ghost long before)?

The same *commercial* units? Or, the $500 version thereof?

They have problems with hot switching. They can even die because of it, not merely "stick." They have had reliability problems even with cold switching.

I would not believe any MFR claims about life expectancy. I would test it first. Or maybe I would not even bother testing it.

They was a "top ten career killers" list on microwaves101. MEMS on your resume was on the list. lol.

That is funny.

--

John Larkin         Highland Technology, Inc 
picosecond timing   precision measurement  

jlarkin att highlandtechnology dott com 
http://www.highlandtechnology.com

What is your use case? Are you planning to strap this thing to the side of a rock crusher or put it in the washing machine for several days? Actually, MEMS are used to detect load imbalance in washing machines. See:

Use the ADXL343 as a simple example, +/-16G measurement limit, 10,000G powered/unpowered absolute max limit. What are your expected G loads?

There is no such thing as a $500 GA anything :) The MEMS are used in certified AHRS.

--
Chisolm 
Republic of Texas

Imagine a two year old using your phone on a daily basis for many years. E.g., expect it to be dropped, thrown, stepped on, dropped in the toilet, left out in the hot sun, on a car dashboard, etc.

(literally)

Probably hundreds of g's. E.g., disk drives don't like being dropped on the floor -- yet are rated as being able to tolerate a few hundred G's (non-operating). Imagine the drive being thrown at a wall (or, kicked across the floor and crashing into a wall).

Of course, I don't need to be able to *measure* those forces. But, can't tolerate failures in the presence of them, repeatedly.

I meant the cost of that actual *component*! :>

Yes, but the same "commercial" units that you would nominally purchase? Or, "select at test"?