Analog Input problem

Ok people here is my problem....Rosemount presure tansmitter scaled 0-200psi reading 140psi at the insturment Via hart. This is about 15ma. Three local gauges confirm that this is the correct presure. The signal is going into a A/B 1794 analog input module, the reading at the card is 8ma. I have powered the loop with a fluke process meter and it reads 15ma. I have simulated the signal back to the DCS through the card and it reads correctly.... 8ma=50psi 12ma=100psi and so on. I have confirmed 24vdc source and even moved the input to an unused one, still have the differance. If I simulate with the hart it does not read correctly and has some bounce. I have checked the cable for a short and found none. I have even gone so far as to run cable outside the conduit between the transmitter and the card just to make sure it is NOT the cable. New cable, same result.The card is full and none of the other Inputs are reading wrong.

What am I missing here? HELP

Have you tried simulating 4-20mA to the card directly with nothing else in line?

how are you reading this 8 ma at the card? And with what?

this is what i would do.
1. draw out the circuit at your desk.
trace the wiring to see if your sketch matches your
wiring.
2. what is the instrumentation cable used? wire size?
3. is the shield properly grounded?
4. is your analog input card configured correctly?
by that i mean read the manual and write it down, then
verify with the program.

i have found many a mistakes by myself and others doing this.

regards,
james

I have confirmed 24vdc source

Click to expand...

The Rosemont transmitter could have its own 24 VDC power supply. The analog input module probably does have its own power supply. Two power supplies on same circuit = incorrect reading.

Lancie1 said:

The Rosemont transmitter could have its own 24 VDC power supply. The analog input module probably does have its own power supply. Two power supplies on same circuit = incorrect reading.

Click to expand...

Most Rosemount Pressure transmitters I am familiar with are loop powered.

I would guess this is not the OP's problem.

phuz said:

Have you tried simulating 4-20mA to the card directly with nothing else in line?

Click to expand...

Yes with a fluke process meter.........everything checked out perfect...

Mickey said:

how are you reading this 8 ma at the card? And with what?

Click to expand...

Same fluke meter in series with input

James Mcquade said:

this is what i would do.
1. draw out the circuit at your desk.
trace the wiring to see if your sketch matches your
wiring.
2. what is the instrumentation cable used? wire size?
3. is the shield properly grounded?
4. is your analog input card configured correctly?
by that i mean read the manual and write it down, then
verify with the program.

i have found many a mistakes by myself and others doing this.

regards,
james

Click to expand...

It is a two wire loop going from the voltage rail on the card to the transmitter and back to the input terminal with a jumper from the other input terminal to the common rail. The cable is the same as the other four hundred plus instruments and the shield is connected in the cabinet but not the instrument. The card is configured right also.

Is it possible the transmitter has been mistakenly wired to two input devices in parallel?

Edit - ok, forget that suggestion then.

Lancie1 said:

The Rosemont transmitter could have its own 24 VDC power supply. The analog input module probably does have its own power supply. Two power supplies on same circuit = incorrect reading.

Click to expand...

It is loop powered there are some for wire devices in the system but they terminate differently.......side by side instead of with a jumper to the common rail.

Mickey said:

Most Rosemount Pressure transmitters I am familiar with are loop powered.

I would guess this is not the OP's problem.

Click to expand...

You have guessed correctly.

This wouldn't be a 0-20/4-20 type issue, would it? If you simulated with a hart communicator then I would ohm the wire and scotch lock one end. Then maybe swap the analog card...if you are sure it is configured correctly and that you are looking at a raw value and not a scaled one.

I do this stuff for a living guys........if it was easy I would have fixed it already.

I'm starting to believe it has something to do with the place being built on an indian burial ground.

I is a noob said:

This wouldn't be a 0-20/4-20 type issue, would it? If you simulated with a hart communicator then I would ohm the wire and scotch lock one end. Then maybe swap the analog card...if you are sure it is configured correctly and that you are looking at a raw value and not a scaled one.

Click to expand...

No it is 4 to 20and on a running process so I can't just swap cards. I also mentioned that I already tried a new cable ........... The cable having a nick in it was my best guess.