At 12:19 PM 2/14/02 +0000, you wrote:
>Hi list,
>
>I'm working on a home project (a drum machine based around a PIC!) that uses
>a couple of 15V regulators in TO220 packages - a 7815 and 7915. They're both
>providing about 200mA, and I'm expecting the total voltage drop across them
>to be about 5V (15-0-15vrms transformer...), hence they'll both be
>dissipating roughly 1W. According to the National Semi. datasheet and having
>done some rough calculations myself I know that the device temperature will
>be well within spec even if the device is used in free-air, i.e. without a
>heatsink. However, I'm wondering if I should go back and redesign my (rather
>dense) PCB before getting it printed, to allow space for a heatsink anyway.

My rule of thumb is not to go over 600mW with an un-heatsinked TO-220. But I
am rather conservative and like to make stuff that never quits. Without doing
any calculations, and not knowing the details of your enclosure, my gut feel
is that you'd probably get away with this for one piece, but it isn't great
design.

>My question is, can anyone suggest a reason why I *should* consider a
>heatsink, especially if the device is switched on for very long periods of
>time, i.e. days? Should I consider a much higher Ta in my calculations as
>the air inside the enclosure warms up? There will be lots of space above the
>regulator (within a 2u 19" rack).

Consider temperature difference from inside to outside your enclosure,
consider what happens if it is used in Barstow or Phoenix in mid-summer,
(or whatever the equivalent scenario is where you are), de-rate WELL from
Tj(max) for reliability, and also consider what happens if your
line voltage is 10% or more high. Also, at light loads compared to what the
transformer is rated for, it may put out significantly more voltage than
the spec'd voltage at full-load*. On for days may actually be better than
heating up and cooling down, the thermal cycling can eventually cause
failures in power semiconductors.

* If you had 10% high line, and 10% more voltage than at full load, your
   dissipation in the regulators would almost be *doubled*. Naturally this
   will also increase the temperature inside your enclosure eventually.

>Thanks in advance for any advice offered.

Can you button them down to the case, maybe using an overmold TO-220F case if
it needs to be isolated?  Assuming a metal case, of course. ;-)

Best regards,

Spehro Pefhany --"it's the network..."            "The Journey is the reward"
speff@interlog.com             Info for manufacturers: http://www.trexon.com
Embedded software/hardware/analog  Info for designers:  http://www.speff.com
9/11 United we Stand

--
http://www.piclist.com#nomail Going offline? Don't AutoReply us!
email listserv@mitvma.mit.edu with SET PICList DIGEST in the body