hi all, if I correctly remember, to known the global error from the individual errors is somehow like computing the derivative of a function. where A and B are transfert functions with dA and dB their respective error. d(A+B) = dA + dB d(A-B) = dA + dB (always the worst case) d(A.B) = B.dA + A.dB d(A/B) = (B.dA + A.dB) / (A.B) these are the basic relations, just fancy how it quickly becomes quite heavy!!! regards Yann --- Johnathan Corgan <jcorgan@aeinet.com> a écrit : > Micah Stevens wrote: > > > If you have 10, 100 ohm resistors in series, and each is 1% tolarance, the > > tolarance of the total group is +- 10%.. That is indeed correct, and if > each > > is +-10%, then the total tolerance is +-100%.. odd as it may sound. > > Uh, no. > > Ten, 100 ohm, 10% tolerance resistors in series add to 1000 ohms. > > If they are all at the maximum of their range, 110 ohms, then their sum > is 1100 ohms, still +10% of 1000 ohms. > > Likewise, at the minimum of their range, 90 ohms, then their sum is 900 > ohms, or -10% of 1000 ohms. > > Same for 1% resistors. Same for parallel connection vs. series connection. > > -Johnathan > > > > > > Yahoo! Groups Links > > > > > > > ___________________________________________________________________________ Nouveau : téléphonez moins cher avec Yahoo! Messenger ! Découvez les tarifs exceptionnels pour appeler la France et l'international. Téléchargez sur http://fr.messenger.yahoo.com
Message
Re: [AVR-Chat] PWM
2006-01-10 by gouy yann
Attachments
- No local attachments were found for this message.