Low Temp Effects on 555, 386, and TL082

[Starbuckin]

Greetings fellow circuit builders...

I built an outside/inside intercom system for my house using TL082's and LM386N-4's...

There is a TL082/LM386 pair inside the house for transmitting to an outside speaker. There is a TL082/Lm386 pair outside in a box on my fence that transmits back to my living room. I then have a "push to talk outside" switch that controls a dpdt relay that simply switches (alternates) between the two speakers being on or off.
When not pressing the switch, the outside circuit drives the indoor speaker and I have an excellent monitor of all sounds going on outside. When I press the switch, the inside speaker goes off and the outside speaker is turned on, so that I can talk to visitors outside.

This system has been in operation for about 7 years and I love it.

My point is that one pair of the circuit is OUTSIDE in an aluminum enclosure.
I have recently added more circuitry to the outside enclosure including 555 timers that drive L.E.D. arrays. Furthermore, all the chips have in their data sheets in their "Absolute Maximum Ratings" Operating temperature 0 degrees Celsius to about 70 degrees Celsius... So.... That means that the chips are not supposed to be operated below 0 Celsius (Which is 32 degrees F, or freezing).

In my location here in Kentucky, Winter temperatures often are far below the 0 degree Celsius rating.

My questions... What effect does low temperature have on these IC's??

Why are the ratings so generous when it comes to high heat (70 degrees C is 158 degrees F) but they can't be operated below freezing?? REALLY??

All these circuits are functioning just fine right now and it's about 15 degrees F outside right now.

It's obvious what the dangers are of operating a chip OVER the Max temp, rating... the chip is going to burn up...

But what about operating them under their low rating??? Is my 555 going to just 'lock up' and stop sending clock pulses to my 4017b and "freeze" the moving dot L.E.D. bar circuit that I built??

 

[(*steve*)]

The temperature rating is mostly due to the packaging material. It may be slightly porous and freeze/thaw cycles can damage it due to trapped moisture being frozen (the same thing that breaks rocks).

There are higher spec devices that have extended temperature ranges.

You may find that the device does not operate within its specification at extremes of temperature. At high temperatures the device may need to be de-rated to zero power, at low temperatures perhaps the colour of the LED is different, or it is dimmer. Batteries are an example where phase change may mean operating them in certain conditions at certain temperatures is forbidden (lead acid batteries will freeze when discharged at a far higher temperature than when charged).

If the device is constantly powered, the heating from just running may be enough to keep them above 0C.

The chip won't "burn up" until the device gets *very* hot. Before that you'll get failures in small parts that may or may not spread to the rest of the device. Before then, you can get effects that will age the device quickly, and before that, you may get into thermal runaway where the device gets hotter, and hotter, ad hotter in a never-ending spiral. Or... (before that) It may just run as normal.

For a simple circuit that you envisage, at low speeds, low power, and low need for precision, it may just run normally.

 

[Starbuckin]

You may find that the device does not operate within its specification at extremes of temperature. At high temperatures the device may need to be de-rated to zero power, at low temperatures perhaps the colour of the LED is different, or it is dimmer. Batteries are an example where phase change may mean operating them in certain conditions at certain temperatures is forbidden (lead acid batteries will freeze when discharged at a far higher temperature than when charged).

If the device is constantly powered, the heating from just running may be enough to keep them above 0C.


For a simple circuit that you envisage, at low speeds, low power, and low need for precision, it may just run normally.

Thank you for your answer... I'm aware that there are milspec versions of these chips that will perform at an extended range of temps and other parameters... As I understand (in the past) they submitted components to a test and if they passed, they offered them as a milspec component but it was more of "cream of the crop" from the production line rather than a special or changed design from the original....

I've also considered that the devices operation and power disapation may be able to keep the chip sufficiently warm though I always use transistors to "current boost" the output in order to take the output current load OFF of the chips and only use them as voltage amplifiers.

For instance, the 555 is supposed to do 200mA output... But I use a PN222A between it and the L.E.D.s I drive so therefore the 555 only has to supply a few mA and the Transistor does the real work...

FYI... Most L.E.D.s produce more light at low temperatures and not less...

Your last statement comes closer to answering my main question than the rest of what you said. I suspect that each chips characteristics are slightly different than normal at low temps so that things such as their max speeds, impedances, etc, are slightly off from the norm... I agree that this shouldn't affect "normal" or general use circuits...

But I still seek the answer... What effect does operating these specific chips LOWER than their operating have on them...

I'm not worried about high temps because I know that there is no way that my enclosure is going to reach an ambient temp of 150 degrees F.

It is shaded by the fence...

 

[davenn]

.......................................
But I still seek the answer... What effect does operating these specific chips LOWER than their operating have on them...

I suspect you are looking in the wrong place for highly accurate specifics on this topic.
Im not aware of any of the forum members that are in the chip fabrication business
so all any of us could do is give you generalisations and point you to already visited datasheets.
For the hi-spec specifics you seem to be after, the only way you will get answers that are likely to satisfy you would be to talk to the manufacturers of the chips you are interested in and see what they have to say

Dave

 

[duke37]

The voltage drop of a semiconductor junction changes at about 2mV per degree Cent. An op amp contains a plethora of junctions so extreems of temperature could affect the operation even if there is no damage.

There will also be leakage problems at high temperature but this should not be a problem in your case.

 

[Starbuckin]

Thank you very much for the replies.

Duke, I find your reply very helpful.

I'm assuming that the 2 milliVolt change is a drop in voltage across the junction as temperature goes down. Is this correct?

 

[(*steve*)]

It's actually -2mV/degC, so it goes up as the temperature goes down.

 

[duke37]

Mullard Reference Manual of Transistor Circuits. (1961)

"Vbe decreases by roughly 2mV per degree Centigrade. This rate of change is almost constant for all transistors (germanium and silicon)."

The book also deals with leakage currents but this is not likely to be significant with silicon transistors.

 

[poor mystic]

Hi
The Ebers-Moll transistor model describes the influence of temperature on current and voltage in a bipolar transistor.
Wikipedia has an article, of course... http://en.wikipedia.org/wiki/Ebers%E2%80%93Moll_model#Ebers.E2.80.93Moll_model
In FET's, the situation is quite different; FET's do not exhibit 'thermal runaway'.

 

[Starbuckin]

Thanks guys for all the replies. This helps me better understand what happens to the chips in extreme cold. Extreme cold seems not to be as "hard" on the chips as is extreme heat (minus humidity). I've always had trouble with my LM386 circuits oscillating... Even using large and small capacitors directly across the supply pins, using shielded cable, and operating without the gain-boost capacitor (gain is fixed at 20, without it), they still will oscillate very easily...

I am doing some detailed research to help me prevent the oscillation...

As I said my intercom system has worked well for many years. It has always had a slight low (motorboating) oscillation to the inside speaker. I noticed the other night that around 14 degrees Fahrenheit, the motorboating stopped and the audio was crystal clear...

Perhaps this is due to the increased voltage drops across all the PN junctions that you have mentioned... This I'm sure, reduced the gain of the 386 as well as the TL082...

Now knowing that the low temperatures will not harm the chips, (I have a large silica gel desiccant pack in the enclosure so humidity isn't a factor), I will start to think of a way to modify the circuit so that the oscillation is gone all the time.

Thanks for the help!