AFAIR I used a photodiode - I can check.
I found that having an opaque cover that extended a substantial distance
beyond the sensor helped substantially. (Even though I used 'substantial'
twice, it seemed the best choice, Fowler notwithstanding).

If you wish to convert the flashing to a continuous analog voltage
indicating instantaneous Watts, it gets surprisingly hard to do well. Work
through what's involved and see if you agree or if I've missed a lightbulb
moment.

"Kill-a-Watt" LIKE (but far cheaper) power meters are now available at
lowish cost and contain ICs that are capable of good accuracy if adequate
sensors are used. A non contact current transformer can be added to your
switchboard cabling - legal here and may or may not be where you are and
Hall sensors are also available. Be aware that Hall ICs that pass the
current through a low resistance path in the module to allow a well defined
relationship between Hall cell and conductor, and which have good mains -
sense voltage isolation, may cease having such if the IC is overloaded and
catches fire - as can happen. In such a case a carbon and/or other
ionisation products bridge can form between high current circuit and logic
signals and supply. Ask me how I know. (In my case only 24V LiFePO4 supply
- damaging but not utterly fatal.)(YMMV).
I used two turns of the en-situ domestic cabling to increase the
sensitivity of the current transformer that I used for testing.

Russell
--=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
.