Errors

[Awi]

I am very bad when it comes to estimating errors and error calculations. For a voltmeter and ammeter what is the normal errors? Could I have 0.5%? And if I had that value, how would I do my error calculation when I take the average of two readings?

 

I appreciate your help!

[ibprincess]

I think for volt and ampmetres you would go with plus/minus the limit of reading which means it depends on the equipment you used (some equipment says the error on the back).

 

If this is for an IA, you can have 0.5% but you might have to explain how you found that value and why you chose to use it as a percentage error rather than a fractional or absolute value because from what I know, absolute uncertainties are the best for data tables.

 

When calculating the error when averaging the readings, I think, unless there were large fluctuations in your readings your error should be the same. If you measured both data points with the same equipment they should have the same uncertainty.

[Vioh]

For a voltmeter and ammeter what is the normal errors?

 

As mentioned by Lada, absolute uncertainty is dependent on the equipments. Uncertainty for a digital device usually has the same decimal places as the maximum places that the device can handle. In other words, if your voltmeter can handle to 2 decimal digits, then the uncertainty for each measurement is +/- 0.01V (which also has 2 d.p).

 

Could I have 0.5%?

 

You should not use the percentage error with raw data because it's not the same for every measurement. For example, if your voltmeter gives 2.04V for the first trial & 3.10V for 2nd trial, then the percentage error for first trial is 0.01 ÷ 2.04 × 100% = 0.49%, while it's 0.01 ÷ 3.10 × 100% = 0.32%. So the conclusion is that you should just stick with absolute uncertainties!!!

 

And if I had that value, how would I do my error calculation when I take the average of two readings?

 

There're 2 possible ways to calculate uncertainty for average of 2 readings. These are explained very carefully in Tsokos Physics textbook):

1. Use the absolute uncertainty of each measurement. For example, if your first measurement is 2.01 ± 0.01 and your second measurement is 2.03 ± 0.01; then your average is 2.02V, and your uncertainty is 0.01V (i.e. you can quote the result to be 2.01 ± 0.01)

1. Use the formula: Range ÷ 2 = (max - min) ÷ 2 --> This formula is used when the 2 readings are too far apart such that the uncertainty is even bigger than the absolute uncertainty of each measurement. For example, if your first measurement is 2.01 ± 0.01 and your second measurement is 3.10 ± 0.01; then your average is 2.56V, and your uncertainty is (max - min) ÷ 2 = (3.10 - 2.01) ÷ 2 = 0.55V. This is because 0.55V is much bigger than the absolute uncertainty of each measurement, which is 0.01V (i.e. it's because 0.55V > 0.01V). So for this case, you should quote your result to be 2.56 ± 0.55

Feel free to ask if there's anything unclear about my explanations above. Cheers

Edited November 11, 2014 by Vioh

[Awi]

Thank you both so much!

 

Vioh: Should I then put an error after each reading instead of putting it in the end of the table? (I hope my question makes sense)

Edited November 11, 2014 by Awi

[Vioh]

Thank you both so much!

 

Vioh: Should I then put an error after each reading instead of putting it in the end of the table? (I hope my question makes sense)

 

If the uncertainty is the same for all the readings, then put it on top (or in the end) of the table. If the uncertainty is different for different readings, then put the uncertainties next to the reading.

 

For example:

 

You see that the column to the left is the length measured in meters by some type of rulers, and therefore will always have the same uncertainty. Therefore, the error is put on the top.

 

On the other hand, column to the right is the resistance calculated by some formulas, and therefore will have propagated errors (i.e. uncertainty will be different for each measurement). Therefore, the error is put next to the reading.

[Awi]

Tack så mycket för hjälpen