-20 degree C

Thread Starter

bug13

Joined Feb 13, 2012
2,002
Hi guys

I may need to design a timer to control something under -20 degree C in high amplitude, what are the problem I will need to solve and what advice do you have for me.

Is it going to be very hard, or easy?

Any info is appreciated, thanks guys!!
 

joeyd999

Joined Jun 6, 2011
5,237
-20C isn't *that* cold -- but some things don't work well at such temps, like LCD displays and Alkaline batteries.

If you need a display, use LEDs. If you need a battery(ies), use Lithium.

Also, keep in mind that voltage regulators, references, crystals, crystal oscillators are designed to have their nominal values around 25C. You will need to compensate/adjust these things if accuracy is desired.
 

Ron H

Joined Apr 14, 2005
7,063
Hi guys

I may need to design a timer to control something under -20 degree C in high amplitude, what are the problem I will need to solve and what advice do you have for me.

Is it going to be very hard, or easy?

Any info is appreciated, thanks guys!!
High amplitude? Did you mean high altitude?
 

WBahn

Joined Mar 31, 2012
29,979
You will want to pay the extra money for Industrial Grade components since most Commercial Grade components are only rated to 0°C or perhaps -10°C.
 

Thread Starter

bug13

Joined Feb 13, 2012
2,002
-20C isn't *that* cold -- but some things don't work well at such temps, like LCD displays and Alkaline batteries.

If you need a display, use LEDs. If you need a battery(ies), use Lithium.

Also, keep in mind that voltage regulators, references, crystals, crystal oscillators are designed to have their nominal values around 25C. You will need to compensate/adjust these things if accuracy is desired.
Thanks for your info, I am glad that -20 isn't that cold. :)
 

joeyd999

Joined Jun 6, 2011
5,237
Being from Florida, have you ever experienced -20°C? Of course, being originally from Wisconsin, I know the natives often say it's getting a little "nippy" out when it gets that cold. :rolleyes:
-20C would be unbearable to me. But electrons -- they don't mind so much, *usually*.
 

crutschow

Joined Mar 14, 2008
34,285
It's interesting the BJTs tend to slow down as it gets cold while MOSFETs speed up. I've actually worked on CMOS multiplexers that work down to around 10°K for cryogenically cooled IR detectors.
 

Thread Starter

bug13

Joined Feb 13, 2012
2,002
It's interesting the BJTs tend to slow down as it gets cold while MOSFETs speed up. I've actually worked on CMOS multiplexers that work down to around 10°K for cryogenically cooled IR detectors.
That's very interesting to know, thanks.

Have you come across an app note that talk about all the think I will need to know when designing a circuit at low temperature?
 

WBahn

Joined Mar 31, 2012
29,979
It's interesting the BJTs tend to slow down as it gets cold while MOSFETs speed up. I've actually worked on CMOS multiplexers that work down to around 10°K for cryogenically cooled IR detectors.
MOSFETs also speed up as the temperature goes up. This makes them a good choice for bore-hole instrumentation in deep wells. The speed up for each are due to different mechanisms. IIRC, the slowest temp is not too far away from room temp.
 

bwack

Joined Nov 15, 2011
113
As for wiring you will need wires with insulation material that doesn't crack in the freeze. I don't remember all the kinds.

I've had trouble with power cables from a batterybox to a light on the bicycle handlebars in winter time. The cable shorted the battery durring a cold ride. Then I switched the cable, and the same happened again after the next ride in the cold. It was a spiral cable with multiple wires inside and it had the standard pvc insulation material. The insulation had cracked and shorted the + and - wires. Probably due to stress and stiff/frozen material.

Then I read somewhere that a nordic distributor of some chinese led bicycle lights (magicshine?) switch the original cables with silicone coated wires for "nordic conditions" (cold). :) There are other matierals too. Anyway, mind the insulation material at -20C !
 

Thread Starter

bug13

Joined Feb 13, 2012
2,002
As for wiring you will need wires with insulation material that doesn't crack in the freeze.
That's useful for me, thanks!

That lead me to think, Do you think the PCB will crack too?? As the timer I mention in my OP, it will be mounted somehow on a little rocket. So I think it will have quite a bit of vibration too when it launch.
 

bwack

Joined Nov 15, 2011
113
I haven't heard of problems with pcb's in the cold.
Rockets, that sounds like a stressing environment, but for a small pcb maybe not that critical as for a larger pcb mounted in a housing that can bend, vibrate or expand to heat/cold.

I must pass on this one I don't know.
 

crutschow

Joined Mar 14, 2008
34,285
For high stress environments you can encapsulate the PCB in a solid potting compound. We did that for electronics in a smart canon shell that was subject to about 20,000 G's when the shell was fired. For that, the IC bond wires between the IC chip and the pins needed to be aluminum since gold wires (often used in high rel ICs) were so dense they would sag under that much force.
 

Thread Starter

bug13

Joined Feb 13, 2012
2,002
For high stress environments you can encapsulate the PCB in a solid potting compound. We did that for electronics in a smart canon shell that was subject to about 20,000 G's when the shell was fired. For that, the IC bond wires between the IC chip and the pins needed to be aluminum since gold wires (often used in high rel ICs) were so dense they would sag under that much force.
Thanks crutschow, encapsulate the PCB in a solid potting compound sounds like a good idea, I have never thought of that.

Do you need to use a crystal oscillator in your design? Does the acceleration at launch have a big impact on the reliablity of the clock? I'm hoping to have at least +/- 10ms accuracy of my timer.
 
Top