Back in the late 80's to early 90's when just starting out in=20
engineering at a tv station. My mentor, had this issue on many power=20
supplies, they designed and built, back then. All their engineering was=20
custom and done in house. He and his other colleagues solved this=20
plaguing issue in a very unconventional way, but very reliable way,=20
which I remember well.

 1. Soldered brass tubes to the legs, inserting the ceramic resistor leg
    into the tube and filling with solder;
 2. Tube diameter, was chosen, based on previous experimenting, not sure
    now, maybe 3mm to 6mm;
 3. Extended the legs if it was required;
 4. Moulded the resistor in either cement, concrete or other ceramics
    (unsure now), in a 2 or 4 litre ice cream container size (about
    that, sometimes bigger), with the brass tubes now the resistor legs,
    which then all mounted to the custom part of the PCB;
 5. Sometimes, they moulded a coiled brass or copper pipe inside, and
    pumped water through as well;

These systems worked commercially day and night for many years in the=20
stations environment, were never turned off, they just worked.

Because space was not an issue, they needed reliability. I even think=20
they reduced their breakdowns to nil for that particular section.

So may not be able to be implemented, but thought I'd share anyway.

Was very interesting!

Peter

------------------------------------------------------------------------
On 17/01/2018 6:15 AM, Bob Blick wrote:
> What techniques are most effective in keeping power resistors from desold=
ering themselves from PC boards? I'm thinking there are basically three met=
hods:
>
> 1. Don't use them at anywhere near their rated power. That's fine except =
if you need to tolerate the occasional overload condition.
>
> 2. Pour extra copper on and around the pads.
>
> 3. Use as much length of the wire leads as possible. That could really de=
crease the rigidity of the mounting.
>
> 4. ??
>
> Attached is a picture of the temperature chart. I'm using the 5 watt AC05=
.. Note the 260 degree K rise at rated power.
>
> If I fold or clinch the leads it will keep the resistor from falling out,=
 but the solder is going to melt. I imagine that won't give the greatest el=
ectrical connection :)
>
> Datasheet here:
>
> http://www.vishay.com/docs/28730/acac-at.pdf
>
> Thanks,
>
> Bob

--=20
http://www.piclist.com/techref/piclist PIC/SX FAQ & list archive
View/change your membership options at
http://mailman.mit.edu/mailman/listinfo/piclist
.