SparkFun Forums

[Vitaliy]

Hi all...

I need to measure 0-5V voltage as single-ended measurents.
I've an ADS1256 ADC (http://hippo.feld.cvut.cz/katalog/pdf/C ... ds1256.pdf).

This ADC have a AINCOM input. What is it used for? Where should I connect it?

Thanks.

[VStar650CL]

AINCOM is not an "input, it's the ground pin for the IC's analog section.

[manton]

AINCOM is not an "input, it's the ground pin for the IC's analog section.

Actually, it is not the ground for the analog section. It IS an input.

Where you connect this would be up to you. Usually this would be connected to a reference or ground that is used for single ended inputs. But, it is really just another input to the A/D. Take a look at the block diagram for the MUX input stage, and it should be pretty clear. In order to allow you to have 8 single ended inputs, they needed to give you a 9th input pin for the common, since this is a differential convertor.

Good luck,

Mike

[Vitaliy]

So if I have VREFP=5V VREFN=GND AINCOM=2.5V, when I set AIN0 to 2.5V ADC measurent wil be 0?

[VStar650CL]

Actually, it is not the ground for the analog section. It IS an input.

It is a reference, but not necessarily ground. Missed that in my first walk through the datasheet, thanks for correcting me.

So if I have VREFP=5V VREFN=GND AINCOM=2.5V, when I set AIN0 to 2.5V ADC measurent wil be 0?

Provided you're measuring single-ended, yes. From what I see, for differential measurement AINCOM isn't used and can be grounded or ignored.

[manton]

From what I see, for differential measurement AINCOM isn't used and can be grounded or ignored.

True, but, it is input the same at the others. It can be used as a differential input. Measuring single ended is really a differential measurement referenced to some other voltage.

This is a great convertor, but watch out for thermocouple effects, as they can be significant when measuring small signals like uninstrumented pressure sensors.

[VStar650CL]

It can be used as a differential input. Measuring single ended is really a differential measurement referenced to some other voltage.

I'll defer to your wisdom, since you're intimate with the part. Page 14 of the datasheet discommends using AINCOM that way, but I'm sure that doesn't mean it can't be done... most of us have made various IO pins do "discommended" things at one time or another. Is there perhaps something in the silicon that TI thinks makes it unwise?

[manton]

I'll defer to your wisdom, since you're intimate with the part. Page 14 of the datasheet discommends using AINCOM that way, but I'm sure that doesn't mean it can't be done... most of us have made various IO pins do "discommended" things at one time or another. Is there perhaps something in the silicon that TI thinks makes it unwise?

It is best if the pins are adjacent to one an other for the best performance. When measuring single ended, the pins are usually not adjacent to one another. If you were using 2 channels as single ended, you could use the pin in between the 2 channels as the common instead of AINCOM, which may give slightly better results than if AINCOM was used.

In general, for differential signals, you would want to run the two signals beside each other so that any induced noise shows up as a common mode signal, and is eliminated. This would force one to use adjacent pins.

[VStar650CL]

Gotcha, that makes sense. Thanks.