mini acceleration switch prefer SMT 1

[firenock]

I urgently need to locate a mini acceleration switch for my application. It will be high volume, 1-5 million in quantity or better when in final production. It have to be really TINY and light weight. Packaging is not as critical since we do epoxy coated the entire unit. The current unit we use used a reed switch. Due to difficult situation, we cannot use a magnet, and we truly believe the only other way to improve upon and maintain the functionality of the application is to use acceleration switch that can detect reliably the acceleration (150 fps). The reed switch we current use is SRC UM23XX. Any one who have info or idea please advice ASAP.

 

[Skogsgurra]

MEMS probably out of question? Or isn't it?

What supply voltage do you have available? Is this acceleration in one direction? Or more axis? What is "tiny"? Like a transistor - or smaller? Or larger?

Do you need a logic signal or a switch that can handle some power? How long is the acceleration period? Some detectors react in microseconds while others need milliseconds. Shal the switch stay activated after acceleration limit is exceeded? Or shall it reset itself?

Gunnar Englund

http://www.gke.org

--------------------------------------
100 % recycled posting: Electrons, ideas, finger-tips have been used over and over again...

 

[MacGyverS2000]

That's about 5Gs, well within the range of many accelerometers. At those quantitites, I easily imagine a dollar or two for single-axis units.

Dan - Owner

http://www.Hi-TecDesigns.com

 

[firenock]

I understand it is well within range of a lot of acceleration switch. What is critical is size, weight and power consumption. The closet I can find is by Star XPF 100 which is right on size and weight. But it is an IC and need power. The application I want need to be only a switch. Since it need to in a application for no less than 3 years by the power source alone. So any power drain will kill the entire app. the power source for this is a 25mA battery. the ideal size can be as big as 5mm X 1mm X 2 mm with the direction force along the 5mm.
FYI, Star-M pdf link below

 

[VEBill]

Minor nitpik: fps is a speed. An acceleration would be 'feet per second per second', or 'feet per second squared'.

A switch with a mass would be very simple, but it certainly won't be tiny.

Is this a one-time trigger? Or need to be used many times?

 

[tinyaccel]

I agree, Sorry for the typing. I just not sure what to type on the screen for that square is hard to do. so 5G switch is what I am looking for. It need to be really small. MEMS is what I have in mind, but that is not that simple. So it is not simple or anyone and everyone would have done it. It has to be a multiple used unit.

 

[VEBill]

Related?

http://www.freepatentsonline.com/7274118.html

 

[Madcow]

Some thing like this?

http://www.inertiaswitch.com/pcbminiature.html

 

[tinyaccel]

VE1BLL, no, nothing to do with that.

Madcow. That is similar, but I need it to be 25% size of it. It is way too big for what I need. It cannot exceed 1.4 mm in height.

 

[Skogsgurra]

You are one tough customer!

Do you need 1/4 of a TO-18 size? I am afraid you will have a very tough time finding that. Especially if you do not want any electronics involved.

You may be able to do something in a PCB - directly from glass-epoxy and Cu. But I doubt if you can get any reliability from such a construct.

Gunnar Englund

http://www.gke.org

--------------------------------------
100 % recycled posting: Electrons, ideas, finger-tips have been used over and over again...

 

[tinyaccel]

I only need it to work MOST of the time. like 80+% would be good enough and low cost. My entire application including battery is only 1.75" X .24" and it weight less than 1 gram, battery included.

 

[MacGyverS2000]

Tough to have MEMS without any sort of electronic control.

Dan - Owner

http://www.Hi-TecDesigns.com

 

[tinyaccel]

I also find out that MEMS is not the greatest in being a non electronic control switch. That is why I am looking at other mini mechanical switch. may be I have to repackage the unit altogether. But this is worth a try, IMHO.

 

[itsmoked]

I suspect you could probably do this with electronics that use less power than your battery does self-discharging.

Does firenock = tinyaccel? I'm confused.

Keith Cress
Flamin Systems, Inc.-

http://www.flaminsystems.com

 

[tinyaccel]

It is the same. just that I forget about my password and I send out 4 times with no respond. So, I just decided to use a new name. Sorry for the confusion.

 

[MacGyverS2000]

For one product I designed, the processor goes to sleep and wakes up every 100ms to check on inputs... I think it used about 60uA, on average.


Dan - Owner

http://www.Hi-TecDesigns.com

 

[VEBill]

ST Microelectronics: P/N LIS202DL

"suits motion-activated user interfaces", "wake-up feature"

5 x 3 x 0.9-mm LGA-14

$3 (10,000)

http://www.st.com

Source: Mentioned in EDN Oct 25, 2007 on page 84.

 

[MacGyverS2000]

Still requires a micro to run, as well as some form of communications channel (I2C/SPI).

Dan - Owner

http://www.Hi-TecDesigns.com

 

[VEBill]

Right. After a quick read, I'd thought that the programmable wake-up function applied to the device itself. But it looks like it needs to be ON (consuming some power) to wake-up the micro. Oh well.

 

[tinyaccel]

After I talked with a MEMS company, they want something like 100K design fee to start with, then prototype fee, etc. I think base on the amount and my budget, that is definitely out. I do think the current package in COMUS have potential. Do anyone know of another company that may have something smaller?

The volume is not small as you can see. COMUS ASLS-2 is reasonable small, but I prefer it to be smaller. Any suggestions?