Actually the finger test is often a good indication. We've just got a
thermal camera here & it's "brilliant" for finding hotspots & getting
temperatures - or at least good estimates. A bit harder to use on
cased up boards however - we are looking at suitable window material.

RP

On 27/04/06, Picdude <picdude@narwani.net> wrote:
> So you're saying that heat testing with my finger on the board after building the prototype is not scientific enough? :-)  I'll run these numbers for my app to see where it stands, but I know in actual practice I can detect any heat rise using MMBF170's (finger method).  Only other thing is the there is a 7805 in the case generating some heat, so the air in there is already 20-30 deg C above ambient.... or more?  I'll have to get a temp sensor in there some time to find out for sure.
>
> Cheers,
> -Neil.
>
>
>
> > I think you have to work out the max die temperature from the
> > transient thermal impedance plot for the max temperature,  timing &
> > load you're using.
> > Then, as long as it's less than 150C, you should be OK. (preferably
> > significantly less).
> >
> > For your frequency & pulse rate the transient thermal impedance looks
> > to be about 40K / Watt.
> > At 250mA & 10 ohms the max power is 625mW which correspnds to a
> > temperature rise of 0.625 * 40 = 25K (or C)
> > Therefore at an operating ambient temperature of 50C the die
> > temperature will be 50+25 = 75C which is much less than the 150C
> > maximum. You should be OK unless I've stuffed up the calculation
> > somewhere.
> >
> > Note also the plot is for a device mounted on a metal clad substate -
> > For a pcb the continuous typical value is 350k/W cp 150 for the metal
> > clad.
> > There should be enough safety marging to cover this however.
> >
> > RP
> >
> > --
> > http://www.piclist.com PIC/SX FAQ & list archive
> > View/change your membership options at
> > http://mailman.mit.edu/mailman/listinfo/piclist
> --
> http://www.piclist.com PIC/SX FAQ & list archive
> View/change your membership options at
> http://mailman.mit.edu/mailman/listinfo/piclist
>

-- 
http://www.piclist.com PIC/SX FAQ & list archive
View/change your membership options at
http://mailman.mit.edu/mailman/listinfo/piclist