Jump to content

  • You cannot start a new topic
  • You cannot reply to this topic

errors

paddyb67

Posted 09 April 2007 - 12:13 PM

in my project im using nlambda = dsintheta
i measured theta using trig after having found the distance between the central maximum and other maxima in an interference pattern,as well as the distance form the slits to the place i measured n=0. im now writing my uncertainties but not sure what to say. i did 8 runs but changed slit separation and all other distances for each run so i cant really say the percentage error is the same in them all and so i cant really work it through to find uncertainty in lambda. is it ok to just guess? or to just do an approx rand uncertainty for the 8 values of lambda i got?

skint_student

Posted 12 April 2007 - 01:47 AM

ok this is what I would do -

1. Work out the uncertainty in the distances you measured (+/- half the least division of your ruler), then convert to a percentage. Then take the largest answer use as the % error in theta.
2. In the same way, use the largest % as your error for d.
3. Work out the random error in lambda from the 8 runs. Convert to a %.

Then add these 3 percentages together to get your total uncertainty for lambda.
The general rule of thumb is to ignore any error which is smaller than 1/3 of the largest error.

I would run this by your teacher before submitting your project, but this method should be ok as long as you make a few brief notes to justify each step. Remember this method gives the maximum error in your result. Then at the end say how the error could be decreased, for example by using more precise instuments to measure distance.
What ever you do don't just guess the error, you need to back up everything you say!

Hope that made sense, if not just ask and I'll try clear anything up.

  • You cannot start a new topic
  • You cannot reply to this topic