ADC Selection for Wheatstone bridge

Hi Doug,

This sounds like it might be a single-ended analog output sensor as opposed to true differential.  I've seen these used in pressure sensors (and others) containing an internal gain stage.  It is possible the output is gained with a differential stage, but I would need to know the common-mode of the sensor before I can fully answer your question. 

In many cases the output of the bridge sensor is rated as sensitivity of full-scale output based on excitation usually in mV/V and is often 1 to 3mV/V making the output very small.  This combination is often used for strain, load cell, pressure, etc..  So it would be helpful for me to know the true output of the sensor (differential or single-ended)? If the output is truly differential, what is the common-mode?  It would also be useful to know the voltage that powers the sensor?  If you have a model number that would be even better.

 A few devices I can recommend are the ADS1232 (unipolar supply only), the ADS1220 and the ADS124S06 (with the last two being powered by either a unipolar or bipolar supply).

Best regards,

Bob B

Hi Bob,

It's a raw (OEM) resistance sensor element that we have to add the bridge resistors to. It's a proprietary design so I can't publish the part number, but it's powered off 5VDC and has a variable resistance of about 400 to 6k ohms depending on pressure applied. With 6500 ohm balancing resistors in the Wheatstone bridge I am seeing about -35mVDC across the bridge with no pressure on sensor and we're only going up to about 2.4VDC at full pressure.

If you can send me a PM I can give you more information on the sensor itself. I think part of the issue is my lack of understand here, looking at the ADS1232 I am little puzzled that it has a range of +/-2.5VDC yet its unipolar... I normally deal with the kilovolt AC gear

Thanks,
Doug

Hi Doug,

So it sounds like the common-mode is about 2.5V (1/2 of 5V) and any of the ADCs I mentioned can work in unipolar supply mode.  The reason you can measure a +/- 2.5V is the measurement of the differential output is based on the voltage of AINP relative to AINN.  If AINP is greater in magnitude than AINN the result is positive.  If AINP is less than AINN then the result is negative.  As the sensor is powered from a unipolar supply and the sensor is a passive device, then the true output cannot be below ground and can be measured by an ADC powered with a unipolar supply.

If the ADC reference is powered from the same supply as the sensor supply, then the measurement can be made ratiometric limiting the noise and drift of the measurement.  The actual full-scale range differs somewhat for the suggested devices.  For the ADS1232 the full-scale range is +/- 2.5V for a 5V applied reference.  For the ADS1220 and ADS124S06 the full-scale range is +/- 5V for a 5V reference.  The advantage of the for the ADS1232 is slightly better resolution where as the other two devices offer high data rates.

Best regards,

Bob B

Hi Bob,

Thanks for the explanation. Are there any parts similar to the ADS1232 that have a "true" SPI interface? I certainly don't have an issue with the ADS1220 but I would like to take advantage of the better full scale range of the ADS1232.

Thanks,
Doug

Hi Doug,

The ADS1232 was meant to be a very easy to use pin controlled device basically clocking out the data.  For monitoring and calibration the ADS1232 does lend itself to using a bit-banging approach as compared to standard SPI communication.  The noise is better on the ADS124S06 compared to the ADS1220, so that may be a better choice if you need better resolution.  There are some other possible choices, such as the ADS1262, but I'm not sure how cost sensitive you are for this project.

Best regards,

Bob B

Hi Bob,

I do like the ADS124S06, do you have any opinion on the new ADS122U04? I realized it's still in preview, but a UART interface is practically cheating.

Thanks,
Doug

Hi Doug,

The ADS122U04 is basically the ADS1220 with a UART communication interface instead of the SPI interface.  The are a few other features that differ like the low-side switch has been removed and there is no option for an external clock input.  Otherwise the performance will be pretty much the same.  This device should move out of preview to fully active status very soon.  Which device you choose to use will really depend on the resolution required.

Best regards,

Bob B

Hi Bob,

Well I managed to get one of the EVMs for the ADS122U04 from the TI store and it's performance has been pretty nice in the few hours I was tinkering with it. Unfortunately I didn't buy the prototype ICs before they were pulled off the TI store. I am hoping the production ones are ready soon so I can move ahead with using them.

Thanks for the help,

Doug

Hi Doug,

we released the ADS122U04 officially to market today. That is why you do not see the prototype samples on ti.com anymore.
By later next week you should be able to order production samples of ADS122U04 in the TI Store. I hope this will still meet your timeline.

Regards,

Hi Joachim,

That's great to hear, I just managed to get a sample request in!

Thanks,
Doug