Reading analog input through E/IP

Hi all!

Is someone can help me to understand what they means ?

This is the only info I could find on my device :

With this module is possile to read two analogue inputs (voltage or current)
The inputs are sampled at 12 bit.
For practicality the sampled value is transmitted with 16 bit, of which the four less significant are fixed at zero.

So, what I understand is that the two inputs are 12 bits but send 16 bits through the E/IP network and the 4 LSB are not used. So, in my PLC i should read a value of 0 for 4mA and a value of 65520 for 20mA?

Is it correct?

I would love to set-up a bench test and test it, but I do not have this remote I/O in hand and need to help a buddy!

What you have stated sounds correct except I don't think you can be sure of the values you will receive from it. And you definitely can't scale it proportionally from zero to 65520.

If it doesn't state 4 to 20mA is the absolute limit of the device, then don't assume it. It might be 0 to 20mA or 0 to 21mA or 3.5 to 22mA. I have seen many variations and without good documentation, a bench test is called for. Hell even with good paperwork, a bench test is the only way to be sure of precision.

You could be exactly correct, but use adjustable variables for your scaling limits in any logic you set up for this. On a related note, you can shift the bits to the right 4 places by dividing the raw value by 16. Then scale proportionally from 0 to 4095 and you "should" get a valid result.

I agree, this sounds right UNLESS the analog input is bipolar instead of unipolar.

For practicality the sampled value is transmitted with 16 bit, of which the four less significant are fixed at zero.

Click to expand...

This is good and how I would have done it. If I had 14 bit of AtoD I would let the 2 lsb be 0. This way the software scaling doesn't need to change if the AtoD changes.

The way this works is that if your input is 0 to 10 volts and that represents 0 to 3000 psi
you multiply the AtoD input by 3000 and then use the value in the top 16 bits of the 32 bit result. The bottom 16 bits are a fraction of 65536. This also assumes you are using unsigned numbers.

This is the only info I could find on my device :

Click to expand...

I am looking both ways so no one hears me.

Can you tell us the complete catalog number of this device? And what PLC you are using?

I promise I won't tell a soul.

I see it a little differently
16 bits is 65,535 but that's not what you can use
the highest number you can use is 65,520 ( remember the lower 4 bits will be 0)
also the while a value of zero is possible with all "0" the lowest value that will be displayed is 16 (Bit 5)
so for scaling you will need to use 15 as 0 and 65,520 as full scale
As pointed out before you didn't say what the input range is 0-5VDC, 0-10VDC 0-20VDC
Most of those are 0-5VDC and for current you need to have a 250 OHM resistor on the input, pass the current through the resistor and measure the voltage across it.
just remember the scaling and you should be OK
Gary

First, thank you all for every answer!

I'm sorry, I forgot to mention that the input is 4 to 20mA.

On a related note, you can shift the bits to the right 4 places by dividing the raw value by 16. Then scale proportionally from 0 to 4095 and you "should" get a valid result.

Click to expand...

It sounds a very good and easy to do idea! I'll tell him to try it this way and keeps you all in touch.

Can you tell us the complete catalog number of this device? And what PLC you are using?

Click to expand...

I do not have any of those, he only sent me a picture of the doc and I know he do use A-B but no ideas of which PLC.