1
\$\begingroup\$

I have been doing research for the past week on setting up a 100 Ohm Platinum RTD to work with labjack U3 LV (I have to use labjack). I know I have to start off with a wheatstone bridge, and use a three wire connection since that is how many wires each RTD has.

For reference it is the labjack U3 LV. I could not post the link as the post would not let me.

This is what I thought the configuration would roughly look like. However I do not know if I need to add a differential amplifier, or an instrumentation-amplifier to obtain a good signal to labjack?

Below you'll see the typical InAmp with 3 op amps that I thought I would have to set up with my wheatstone bridge circuit.

enter image description here

However I don't even know how to begin choosing resistors for this, or how to properly set it up with the wheatstone bridge. The question remainds should I use this InAmp?

If neither of these would be suitable to obtain good readings from the RTDs then how should I set up my circuit? I am not very well versed in setting up circuits and much less using op amps. So if there is an alternative could you explain what it is, and why it should be used and how would I go about finding the resistors I would have to use for it to output a good reading, and the current that would run through it to obtain a good reading on the labjack? Actually I believe the current is 1 mA for most of the PT100's? Unfortunately I do not have the exact data sheets as this system has already been set up as is for years without use.

The version of the labjack that I have can handle 0 - 2.4 V and 0 - 3.6 V. (12 bit) Currently I am also not sure how I would power it, but I was thinking of using a DC Source like a small battery. I'd like to verify that I would need a voltage divider, and if so how would I go about incorporating it into the circuit?

I have been attempting to understand this set up, but the op amp part is what is confusing me the most. Since I know how to balance a normal wheatstone bridge, is there anything different I would do for this? Overall I'll be getting readings from 8 RTDs, but I figure once I figure out how to do one of them I can just do the rest of them the same way since they are all the same. For reference these RTDs are on heat pumps and will eventually be used in a GUI that will convert the voltage received on labjack to a temperature. All help is appreciated! Thanks!

\$\endgroup\$
  • \$\begingroup\$ I forgot to add that this is temporary, it is to test and make sure I can get proper readings from the RTDs and a schematic will be drawn up and sent to a manufacturer to make a more permanent board. \$\endgroup\$ – Kvothe Nov 29 '16 at 4:32
  • \$\begingroup\$ The easiest way is to buy a commercial signal conditioner for each 3-wire RTD if you actually want the line resistances to cancel and the output voltage to be accurate in absolute terms. \$\endgroup\$ – Spehro Pefhany Nov 29 '16 at 7:35
  • \$\begingroup\$ Resistance changes in an RTD should be plenty big enough to use without a wheatstone bridge. \$\endgroup\$ – Scott Seidman Nov 28 '17 at 22:19
1
\$\begingroup\$

Instrumentation amplifier is rather expensive. The input impedance is very huge, therefore it won't influence the wheatstone bridge. Further, what a RTD conditioner has to have is a low pass filter at input stage. You can place a RC filter on input, that again won't influence the bridge, since the input impedance of IA is unlimited. But there are ready made ADCs for RTD like ADS1248,...that are cheaper than a single IA.

\$\endgroup\$
  • \$\begingroup\$ I've used the AD1248. The matched current sources, negate any line resistance, so the only resistance measured with just the RTD. I didn't even need a bridge. I also just noticed that this post is a month old. So this comment may not be relevant to OP, but maybe to future reader. \$\endgroup\$ – efox29 Jan 2 '17 at 9:56
  • \$\begingroup\$ @efox29 Schematics of your work would help the community, as I didn't found a neat implementation for multi channel RTD, yet. \$\endgroup\$ – Marko Buršič Jan 5 '17 at 8:56
0
\$\begingroup\$

Your Wheatstone bridge turns resistance changes into small floating voltage changes, with a scale factor of the supply voltage. Make sure your bridge drive voltage is regulated and constant.

You would use your InAmp to convert this small floating voltage to a larger single-ended voltage for the converter. Connect the InAmp input to the bridge output, one input to each side of Vb. It doesn't matter which way round, though it affects the output polarity.

An InAmp has two stages of gain. First the fully differential stage. Then the differential to single-ended stage. I tend to assign the overall gain into sqrt(total) for each stage, to keep the individual stage gain as low as possible, but there's much flexibility.

The first stage gain is (2Rf+Rg)/Rg, if you look you can see that factor in the VO expression. If you omit Rg, the gain is unity. Reasonable resistor values would be in the 1k-100k range. Using 10k for Rf and 2.4k for Rg would give a gain of 10(ish) for the first stage.

The second stage gain is R4/R2, you can also see that factor in the VO expression. You must keep the R4/R2 ratio equal to the R3/R1 ratio, it's simplest to keep R3=R4 and R1=R2 as the data sheet suggests. The same reasonable range of resistor values applies.

If you want to adjust the overall gain, the simplest place to do it is the value of Rg, as it's a single resistor. All other resistors in the amplifier are equal-value pairs.

\$\endgroup\$
0
\$\begingroup\$

This looks like a circuit I used many years ago: It needs cleaning up with filter caps, etc. Should work with an 100 ohm RTD and a LT1001 should work well for the op-amp. If you search bing Images you should find variations and improvements on this. enter image description here

\$\endgroup\$

Your Answer

By clicking “Post Your Answer”, you agree to our terms of service, privacy policy and cookie policy

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