David said:
In order to keep the burden voltage as low as possible (which is the whole
point of the project), overload protection has been ommited.
Fair enough, but in that case it would be useful for the user to know
the level of overload which is considered safe.
If you have implemented the device with current sense resistors and a
voltage amplifier then it might make sense to add protection for the
amplifier input to handle the case where a massive overload fuses the
current sense resistor.
Yes, I considered that in various ways, e.g. a small PIC with a smart
push-button power switch or by some other means, but went with the KISS
principle in the end. Some people groan when they see a micro in such
projects, I didn't want hate mail :->
An optional auto-off feature might be a compromise.
Auto-off is of course a 'two-edged sword'. I have a battery powered
differential probe at work and find that its auto turn off time is
usually too short for the way I work... It's annoying, but probably
less so than continually replacing flat batteries.