Which one has the lower noise level？Wheatstone bridge or direct measurement?

[Chengjun Li]

Hi,

In the Whetstone bridge circuit:

If the circuit is balanced (and all resistances known), and we measure Vg as a function of time, then the fluctuations correspond to
4KbTR
the well-known thermal fluctuations. My question:1.what should be used for R here? 2.And if the Rx is function as a resistance temperature detector, its temperature is different from the three other dummy resistor, then what should be used for T here?

We can directly measure a resistance using the votlage divider circuit and a dummy resistor. 3. I wonder which method has a low noise level?

 

[Colin Mitchell]

The noise depends on the type of resistor, the current, the ambient temperature and the number of resistors. Also the noise from the power supply and the voltage of the supply as this affects the value of each resistor.

 

[shrtrnd]

I agree with Colin Michell's analysis.
As a general rule of thumb, the more components (and types of resistors), and the longer the leads, the more noise you're going to pick-up.
I am curious as to why you considered the possiblity that adding a Wheatstone Bridge would somehow reduce the noise level over direct measurement?

 

[Arouse1973]

The noise depends on the type of resistor, the current, the ambient temperature and the number of resistors. Also the noise from the power supply and the voltage of the supply as this affects the value of each resistor.

Hi Colin
If we are talking about purely thermal noise then I don't think it is related to the type of resistor only the value. Contact noise or also called current noise and shot noise are dependant on material type and the current in a resistor. The physical size (wattage) will also have a bearing on the later.
Thanks
Adam

 

[mofy]

I think you would use the thevenin equivalent resistance: i.e. R1||R2, R3||Rx. You would then RMS the noises together to get the final noise across Vg.