Go Down

Topic: How to measure chip (or other component) temperature? (Read 1 time) previous topic - next topic

oric_dan


Seen a 7805 open up once, guess the shutdown had an issue.  I also learned it's also a good idea to pay attention to voltage and polarity with tantalum caps.  Anybody got any really good incendiary stories?


What, you don't like my flaming 293? Another time, I had a TO-220 MOSFET literally melt off its
heat-tab. Little silvery beads came out from between the chip die and the heat tab.

L298 will go to 1.5-2 Amps, plus it's a Multiwatt15 package, so you can screw a big heatsink to it,
but they're still the same bipolar technology as L293. The 293/754410s are DIP16, but you can epoxy
a smallish heatsink on top if you like.

afremont

#16
Mar 08, 2013, 08:22 pm Last Edit: Mar 08, 2013, 08:24 pm by afremont Reason: 1
LOL, yeah caps can be dangerous in more ways than one.  Of course I liked your flaming h-bridge story Dan, that's what set me off.  I apologize for not specifically acknowledging it though.  

I learned to use fuses with my first car stereo self install.  I had a power lead coming out of an unfused connector on the fuse box and I didn't have any fuses inline.  It was a sloppy job, but it worked for quite a while until one day I got tangled in some wire and it shorted.  Lit up several feet of wire just like a toaster element.  The wire didn't burn in to, but it was the weirdest thing to see all the insulation just drop off the wire as it lit up.   :smiley-eek:  Fortunately I was able to yank it apart before things started catching fire.  I do things different now.  I was really lucky in that the wire feeding the fuse box was heavy enough to not melt down.
Experience, it's what you get when you were expecting something else.

oric_dan

Car electronics is a good place to pay attention to safety measures. Don't want to catch
fire out on the freeway. 50-cent fuse, $25,000 car.

afremont

Ham radio mobile HF with 30A from the battery to the interior for minimum noise.  Fuses on the hot and ground at both ends. 

I've seen a couple of cars burn up in my day and it's almost unnatural how they can be consumed like that.  Everything including the metal is flammable.  In the case of magnesium, especially the metal.
Experience, it's what you get when you were expecting something else.

SirNickity

I learned if you put a 7915 where a 7815 is supposed to go, a TO-220 will split apart between the tab and case.  I always test power supplies on a Variac, so I learned it only takes a few volts to do this, too.  :-)

Docedison

No, I think a 7905 is different enough (data sheets attached) the input lead (1) is ground the tab/center pin/ground (2) is - supply and the third pin is the - output lead...
The device won't work properly but sorry nothing really reversed... I know I've used
Now IF you REALLY want to EXPLODE a TO-220 transistor, connect the emitter of a TIP42 to a real stiff (> 10A) power supply and then just ground the base... even works with the little TO-92 types... In my career as an engineer I've probably popped... a hundred or more..
I used that "trick" to get rid of my boss... he had a bad habit of looking over my shoulder.. I could never tell if it was something he was looking for to criticize or something to learn... But a "Slip" with a screwdriver or a meter probe always worked... even one time when After I had done for the transistor.. he handed me the negative meter probe and walked away.

Bob
--> WA7EMS <--
"The solution of every problem is another problem." -Johann Wolfgang von Goethe
I do answer technical questions PM'd to me with whatever is in my clipboard

SirNickity

Come to think of it, it wasn't that I used the wrong part, it's just that I swapped the ground and input pins on the PCB.  First PSU design and I didn't realize they were in a different order between the 78 and 79 parts.  Positive rail worked fine.  Negative rail exploded.  Oops!

That's a great story, BTW.  Ever worry that the boss thought of you as "the clumsy guy that's always blowing up parts?"  ;-)

JimboZA

Just for laughs I tried the following experiment....

I put a 7805 on a 12V supply and loaded it with a 10  Ohm resistor. Voltages measured at 12 and 5, and current measured as expected (497 Fluke mA to be exact). (Yes the resistor is hefty enough: 10W) So this set up is losing 7V @ 500mA for a power dissipation of 3.5W.

Without a heatsink, that 7805's temp shot to 110C in mere seconds. I turned it off then, since the datasheet said 120C tops.

I rummaged around and the smallest heatsink I could find is about the size of 2-3 Arduino Unos. Mounted the 7805 on there (no conductive goo though) and it settled just under 40C.

I now realise the importance of heatsinks, so that was a useful 1/2 hour of Saturday morning buggering around which beats shopping with Mrs Jimbo.
My hovercraft is full of eels.

No PMs for help please.
DO NOT power servos from Arduino 5V: give them their own power and connect the grounds.

oric_dan

Quote
So this set up is losing 7V @ 500mA for a power dissipation of 3.5W.

Without a heatsink, that 7805's temp shot to 110C in mere seconds. I turned it off then, since the datasheet said 120C tops.

That's pretty close to what the d/s indicates to expect. If you find the lines that say
----------
package thermal data (see Note 1)
PACKAGE BOARD ?JC ?JA
POWER-FLEX (KTE) High K, JESD 51-5 3°C/W 23°C/W
TO-220 (KC/KCS) High K, JESD 51-5 3°C/W 19°C/W
NOTE 1: Maximum power dissipation is a function of TJ(max), ?JA, and TA. The maximum allowable power dissipation at any allowable ambient temperature is PD = (TJ(max) - TA)/?JA. Operating at the absolute maximum TJ of 150°C can affect reliability.
----------

and you plug in, you get 3.5W * (3°C/W + 19°C/W) = 77 °C rise --> 99 °C (including ambient).

There are various heatsinks with different capabilities, but I like a nice compact one (25.9°C/W),

http://www.jameco.com/webapp/wcs/stores/servlet/ProductDisplay?freeText=158051&langId=-1&storeId=10001&productId=158051

I'm not quite sure how to calculate overall heatsinking of chip+heatsink, but I guess the two will
work in parallel, so (22 || 25.9) = 12 °C/W --> 42 °C rise at 3.5W. Something like that.

There are also heatsinks made specifically for your DIP16 L293,

http://www.jameco.com/webapp/wcs/stores/servlet/ProductDisplay?freeText=2077335&langId=-1&storeId=10001&productId=2077335

Go Up