Oli Glaser talktalk.net> writes: >=20 > On 23/04/2011 05:07, Gerhard Fiedler wrote: > >> This was my point - I'm wondering how accurately they measure the > >> > instantaneous power. > >> > For instance, if the power was 10kW for a millisecond then drops to= 100W > >> > for 999ms, but you were only averaging the power over every second = then > Yes, it's a very basic point I was trying to make, about just what you=20 > mention - getting *less* data. This is not always the objective. > If your concern is about excessive data rates, your have cheap PICs with= =20 > up to 1Msps ADC capabilities, more than enough for very accurate peaks. > My point is simply what if someone wants to know their maximum=20 > instantaneous power attained? For some sports I imagine this may be=20 > important/useful to know and it might be good for sports equipment to=20 > sample it accurately. For this reason I imagine sampling at a higher=20 > rate would be better as it is more flexible - then you can keep the=20 > peaks for the true maximums and also average the data in firmware over a= =20 > longer period. If you filter first the peak data is lost forever. >=20 What about a pulse of 0.1 ms, or 0.1 ps? If the objective is to detect pea= k power, along with average use, one can use a peak detector. This will give= you that information without a significant increase in the data rate, although = at the cost of a second ADC channel. =20 Sergey Dryga http://beaglerobotics.com --=20 http://www.piclist.com PIC/SX FAQ & list archive View/change your membership options at http://mailman.mit.edu/mailman/listinfo/piclist .