2

What accounts for the slight difference in the theoretical and experimental values of the time constant in a simple RC filter?enter image description here

The slight difference was just (.5*10^-4) seconds

Sakamoto
  • 23
  • 1
  • 1
  • 5
  • Can you add a circuit diagram (edit your question, then hit CTRL-K)? Also be more specific about the amount of difference that you observe and how you measure it. – jippie Sep 22 '13 at 07:50
  • Okay. I've edited it. It's a simple circuit, sorry I'm kinda new. @jippie – Sakamoto Sep 22 '13 at 07:53
  • So what you are saying is you expect $1500 × 100 \cdot 10 ^{-9} = 150 \cdot 10^{-6} [\text{s}]$ and you somehow found a difference of $9 \cdot 10^{-6} [\text{s}]$, is that correct? Question still is how you measured this. – jippie Sep 22 '13 at 08:01
  • @jippie sorry! I've written the wrong values for the difference. I wasn't looking in the same circuit. I've edited it now. The difference was 5*10^-4 [s]. – Sakamoto Sep 22 '13 at 08:04
  • 3
    Now you confuse me, you expect $\tau = 1500 × 100 \cdot 10 ^{-9} = 150 \cdot 10^{-6} [\text{s}]$ and you call a difference of $ 500 \cdot 10^{-6} [\text{s}]$ "slight"?! That is 300% wrong. You should really be more specific what you are measuring and how. – jippie Sep 22 '13 at 08:13
  • Oh. @jippie. Didn't see the mistake. Sorry. The dot was missing in front. I got it now. The resistor's tolerance was to blame – Sakamoto Sep 22 '13 at 08:21
  • According to the information in the question (as of 3:22 pm 9/22/13 UTC), the expected time constant is 150 us, and the error is 50 us. That's a 33% error. Resistors with worse than 10% tolerance are really rare nowadays. Are you sure you aren't still missing something? For example, what was the capacitance of the scope probe you used to measure the circuit with? – The Photon Sep 22 '13 at 15:26
  • Another question, what measurement did you do to obtain a measured time constant of 200 (or 100) us? – The Photon Sep 22 '13 at 15:26

1 Answers1

2

No equation can be completely reciprocated by a Physical System of realizable complexity. However you measure it there always is an approximation; unless and otherwise you're simulating it, in which case the simulations are just equations.

For example in this circuit, how can you be 100% sure that the Resistor and the capacitor are exactly the value you want them to be? There's always a "tolerance" value for a Resistor which indicates the amount by which its resistance deviates from its marked value.

Sudarsan
  • 201
  • 1
  • 4
  • Beyond component tolerance (which you can decrease by measuring and selecting the pieces using a higher precision measurement device), there are other factors involved. Every component exhibits some parasitic properties, unlike idealistic components that exist only in simulators. For example, real capacitors have not only capacitance, but resistance and inductance as well. These parasitic properties may or may not have substantial influence on your measurements. – Laszlo Valko Sep 22 '13 at 14:42
  • Series parasitic resistance in smaller electrolytic capacitors can easily be several ohms. When employing smallish currents (on the order of microamps), parallel parasitic resistance (causing leakage in the capacitor) can also have a substial influence. – Laszlo Valko Sep 22 '13 at 14:46