[sdiy] IC Temperature

Mattias Rickardsson mr at analogue.org
Mon Aug 3 14:11:50 CEST 2020


Hi Mike,

Den sön 2 aug. 2020 19:53Mike Beauchamp <list at mikebeauchamp.com> skrev:

> On 8/2/20 5:26 AM, Mattias Rickardsson wrote:
> > A heat camera can prove useful for design as well.
>
> Hi Mattais, I borrowed the FLIR for this exact reason - to check a
> prototype PCB design.
>
> The first thing I noticed was that all of my IC's get very hot. TL074's
> are all going up to around 50C in 20 minutes, NE5532 a bit hotter and
> the LM13700's stay pretty chilly. After running the board for about 20
> minutes, the whole thing is warm to the touch - something I'm not really
> used to in previous designs. But this is my first surface-mount design
> and it is probably twice as dense as previous stuff. I'm using a +/-15V
> power supply.
>

Around 50 degrees is what I'm experiencing as well on TL074s on densely
mounted double-sided SMT designs at +/-12 V supplies. At first quite
unexpected since we're used to old hole-mount synth circuits not going very
warm at all, but when measuring the current draw and summing up the
expected supply currents of all circuits it seems to be correct. I guess
the density is the main reason here, both having the heat sources much
closer to each other AND having much smaller area & less material for
dissipating the heat.

Btw, TL064s are hardly visible in the heat camera - they pull so little
current. Applying TL064 op-amps where noise & slewrate aren't critical
factors is an important trick in designs where power & heat & cost are.


> I checked the datasheet for the TL074 I'm using and it's "Operating
> Temperature" is 70C Maximum. I can't help being worried that my measured
> 50C is pretty close to this, although I'm 95% certain I'm not doing
> anything electronically "wrong" with all of these chips to put them in
> danger.
>
> Maybe "Operating Temperature" refers to the ambient temperature that the
> chips can operate in, not the temperature they reach?
>

It would be very hard to actually measure the temperature on the hottest
spot on (or even in) an IC, so from a design perspective I just assume that
"Operating Temperature" refers to the surrounding air and PCB - and then
the internal IC temperature lands on some reasonably well-known temperature
above this, since the delta temperature is given by the expected power draw
of the IC. Please do chime in, if someone can correct or clarify this! :-)

/mr
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://synth-diy.org/pipermail/synth-diy/attachments/20200803/3a8dc317/attachment.htm>


More information about the Synth-diy mailing list