[sdiy] temperature testing

[Neil Johnson]

Hi Matt,

> Perfect, thanks Neil. I guess "Incubator" is the magic search term.
> definitely hobby budget! Hopefully I can find an equivalent in the states.
> So I'm guessing your process with the incubator is manually stepping up the
> temperature with a stabilizing period in between, with the thermocouple for
> the accurate temperature readout? Is the superglue for the thermocouple?

You can't rely on the thermostat on the oven - it's more of a rough
indicator, coupled with the time constant of the control loop.  So if,
for example, you're trying to determine the thermal properties of an
oscillator, say, then you need far better temperature measurement at
the device of interest.
Also, to minimize the effects of the test equipment on the experiment
you want to add minimal thermal mass to the device - small
thermocouples and tiny amounts of glue.

> I like the fridge idea for single point measurement. What would be the
> downside to say, unplugging the fridge and taking measurements while it
> warms up to ambient? Temp changing too fast to be sure the data is accurate?

Warm beer.

> Multiple thermocouples... to be certain you aren't being misled by hotspots
> on the PCB?

You want to know the chamber ambient, you want to know what
temperature key components are at (unless you like wasting lots of
time running a test with a single thermocouple at different points in
the circuit).

Cheers,
Neil
-- 
http://www.njohnson.co.uk