AMAZING temperature range Mega168

Go To Last Post
19 posts / 0 new
Author
Message
#1
  • 1
  • 2
  • 3
  • 4
  • 5
Total votes: 0

Another thing that may interest you:
I held a little technical seminar today,
where the topic was to see if we could find max and min working/not working temperature range of several normal electronic devices.
For the test I used a normal two layer pcb and mounted a few parts.
The AVR microcontroller Mega168-20AU
I use is rated -40C to +85C a range of 125C
I found out this:
it still worked just fine at -183C fully covered in LOX,
and failed at +180C when it was floating in liquid solder.
a working temperature range of 363 Celcius, that is quite amazing :-)

want to see pictures/video clip of this test ?

Yours: Thomas Scherrer - Denmark
OZ2CPU www.webx.dk

  • 1
  • 2
  • 3
  • 4
  • 5
Total votes: 0

Quote:

want to see pictures/video clip of this test ?

The first think I thought of was the current thread on Mega48P (PA?) where 32kHz auxiliary crystal appears to be failing at freezer temperatures.

But that is a bit different than your test. ;)

You can put lipstick on a pig, but it is still a pig.

I've never met a pig I didn't like, as long as you have some salt and pepper.

  • 1
  • 2
  • 3
  • 4
  • 5
Total votes: 0

AVR is go for space exploration.
Quite impressive.
Id love to see a video.
You didn't have a PIC on the pcb to compare with?

  • 1
  • 2
  • 3
  • 4
  • 5
Total votes: 0

I find nothing abnormal with this. The liquid nitrogen temperature can only damage the package pastic after several repeated warmup/cooldown cycles, whereas the melted solder temperature would only make some chip timings going wrong due to increased logic gates'/transistor structures' delay.

Warning: Grumpy Old Chuff. Reading this post may severely damage your mental health.

  • 1
  • 2
  • 3
  • 4
  • 5
Total votes: 0

Does the test include writing and reading to the EEPROM?

I think that there have been statements to the effect that this is one of the more temperature sensitive internal operations.

Jim

Jim Wagner Oregon Research Electronics, Consulting Div. Tangent, OR, USA http://www.orelectronics.net

  • 1
  • 2
  • 3
  • 4
  • 5
Total votes: 0

Quote:
whereas the melted solder temperature

Or perhaps crack / explode the plastic case if there was significant humidity within the device and it was heated too quickly...

Very COOL Test...

JC

  • 1
  • 2
  • 3
  • 4
  • 5
Total votes: 0

It would be interesting to see how the internal oscillator performs at those temps. I have seen internal clock generators which have logic gates stop functioning due to the timing problems that Mbedder mentions.

Also would be interesting to see how the internal AREF (with the bandgap) performs at these temps.

  • 1
  • 2
  • 3
  • 4
  • 5
Total votes: 0

I guess it was horizontal. I also guess it would fail oriented vertically and with a minimal vibration frequency/amplitude.

Ross McKenzie ValuSoft Melbourne Australia

  • 1
  • 2
  • 3
  • 4
  • 5
Total votes: 0

Lennart wrote:
AVR is go for space exploration.
Quite impressive.
Id love to see a video.
You didn't have a PIC on the pcb to compare with?

Microchip has a range of parts that are qualified for 150C operation:

http://www.microchip.com/stellent/idcplg?IdcService=SS_GET_PAGE&nodeId=2844&param=en544687

They would probably work at an even higher temperature.

Leon Heller G1HSM

  • 1
  • 2
  • 3
  • 4
  • 5
Total votes: 0

Neat. Try the high temp. again with welded or brazed connections. :)

  • 1
  • 2
  • 3
  • 4
  • 5
Total votes: 0

leon_heller wrote:

Microchip has a range of parts that are qualified for 150C operation:

http://www.microchip.com/stellent/idcplg?IdcService=SS_GET_PAGE&nodeId=2844&param=en544687


So does Atmel - most (if not all) "automotive" AVRs are specified at 150 oC.

Warning: Grumpy Old Chuff. Reading this post may severely damage your mental health.

  • 1
  • 2
  • 3
  • 4
  • 5
Total votes: 0

we did find other parts not working at this temperature range, a very normally used 3.3V regulator LP2981 fail alot erlier !
offcourse still way out of its spec.
yes we are designing parts for space, but that project have nothing to do with this, test out of spec day. I have hard of a company that use AVR mega for +140C, that is out of spec for the parts, and their project is in hi volume production.
if my fail at +180C was due to parts moved in the floating solder or change in electrical parameters inside the AVR, I dont know, but it did resume program execution again when I made it a little bit colder, and it was the very same devices I have been running -183C tests on several times today.

Yours: Thomas Scherrer - Denmark
OZ2CPU www.webx.dk

  • 1
  • 2
  • 3
  • 4
  • 5
Total votes: 0

Torturing is prohibited by law :D

Warning: Grumpy Old Chuff. Reading this post may severely damage your mental health.

  • 1
  • 2
  • 3
  • 4
  • 5
Total votes: 0

MBedder wrote:
Torturing is prohibited by law :D
Unless you learn some valuable secrets ... (wasn't that "their" excuse?)

Ross McKenzie ValuSoft Melbourne Australia

  • 1
  • 2
  • 3
  • 4
  • 5
Total votes: 0

Just keep in mind there is a HUGE difference between "working" and "fully function". The limits the manufacture sets for devices (be it operating temperatures or clock frequency or voltage specs) are set to allow for the device to remain FULLY functional when everything is skewed to the "wrong" limits. The actual limits are set based on statistical probabilities more than "it was still running at 125C, so we can rate it at that."

But it IS still impressive that these little things can survive that kind of torture!

 

Clint

  • 1
  • 2
  • 3
  • 4
  • 5
Total votes: 0

Bah, you're all so smart with your fine theories on why it doesn't work when submitted in liquid solder.
It's obvious - it shorts! ;-)

/Jesper
http://www.yampp.com
The quick black AVR jumped over the lazy PIC.
What boots up, must come down.

  • 1
  • 2
  • 3
  • 4
  • 5
Total votes: 0

>It's obvious - it shorts!

the test was performed on a perfectly stable
and flat heat plate, and not moved a bit when
solder was liquid, and as you see in my txt explain,
the function was restored when temperature got a little bit colder, if there was a short,
it would continue to be a short after cooling down.
we have been using this methode for testing other parts too, and many parts works just fine when floading in liquid solder

Yours: Thomas Scherrer - Denmark
OZ2CPU www.webx.dk

  • 1
  • 2
  • 3
  • 4
  • 5
Total votes: 0

ThomasScherrer wrote:
we did find other parts not working at this temperature range, a very normally used 3.3V regulator LP2981 fail alot erlier !
offcourse still way out of its spec.

and they claim "over temperature protection" is a feature :).

  • 1
  • 2
  • 3
  • 4
  • 5
Total votes: 0

Quote:
and they claim "over temperature protection" is a feature

and they are quite right! It depends how you define "fail". Destructive failure as in "the smoke came out" or non-destructive failure to operate normally because of current foldback limiting!

Charles Darwin, Lord Kelvin & Murphy are always lurking about!
Lee -.-
Riddle me this...How did the serpent move around before the fall?