1
\$\begingroup\$

For a project I require simple temperature sensing. The specifications are as follows:

  • temperature range: 50-250 Celsius
  • costs resonably low (< 10 US)
  • accuracy +/- 2 degrees

for cost and availability reasons I selected the following RTD: (farnell)

The trick now is to get the temperature information into a micro controller. This will be accomplished using an Wheatstone bridge and an MCP6N11 (farnell). The gain of the amplifier has to be around 6. A proposed schematic can be found below (note that the RTD is placed between "R_uit" and "R_in").

Proposed circuit design

There rests only one question. How do I properly set Vref? Considering that the micro controller can measure between 0 and 5V, which I prefer to utilize fully. In the datasheet is stated that Vref "lifts" the output voltage. With that in mind, isn't it best to just ground Vref? It is provided that the MCP6N11 is a rail-to-rail device.

In the User guide of the demonstration board they talk about setting Vref around 1.8V when Vdd is 5V.

\$\endgroup\$

2 Answers 2

2
\$\begingroup\$

Vref of an amplifier determines its zero point; when the input is 0, the output will be equal to Vref. If the polarity to be measured is equal in both directions then Vref should be 2.5V for maximum range with a 5V ADC, otherwise it should be biased appropriately given the input.

\$\endgroup\$
2
  • \$\begingroup\$ using the circuit and RTD described above, the potential difference is between 0 and +0.5V. Given that, polarity only has to be measured in one direction. So grounding it should not be an issue right? \$\endgroup\$
    – Roel
    Jul 2, 2014 at 16:10
  • \$\begingroup\$ Assuming rail-to-rail input and output. Otherwise keep it a few hundred millivolts within the linear range. \$\endgroup\$ Jul 2, 2014 at 16:23
1
\$\begingroup\$

Yes, grounding it should be okay, since you don't need to read down to 0°C. With offset and such like, you might not be able to distinguish 1-2°C from less than that.

Note that you'll get significant self-heating with such an arrangement, I would recommend no more than 10% of the current you are using and up the gain of the inamp proportionally.

\$\endgroup\$

Your Answer

By clicking “Post Your Answer”, you agree to our terms of service and acknowledge you have read our privacy policy.

Not the answer you're looking for? Browse other questions tagged or ask your own question.