Signal [mA] missing somewhere

Hey all.

I have a problem I seem to be unable to figure out by myself. I have been searching for a few days now.

The situation is as follows:
I have a PLC upstairs (a Saia PCD.M150, although I doubt the PLC type is of importance regarding this. If you feel otherwise, feel free to inform me) with a supply power of 24 [V].

This very same supply power is used to power a signal downstairs where we have a display displaying weight. This display has a card inside it that allows remote reading of the value via 4-20 [mA].

Now the power to that signal comes from the PLC-system upstairs (+24 [V]), goes through the car in the display and returns through a fuse into an input on the PLC.

That's the setup. Now, the problem is as follows:
I've tried using a loop calibrator to power and read the value of the display. It read correctly. Then I tried inserting a potentiometer instead of the display. The PLC read correctly. Then I tried plugging my PLC-signal into a display of same sorts that I know functioned correctly. It still read the correct values. But when I have the signal connected to the corresponding display, it shows either 0[mA] or 9,9[mA] (Two different displays).

Does any of this make sense to you? Are anyone able to figure out the cause of this problem, and the remedy?

Thanks a bunch in advance =D

Hei !

This is very a small sketch would be really helpful.

Here it is.

A few possible issues. I guess that the display also has a separate supply going to it, so make sure that it is a passive rather than active output. If it is an active output then you don't need the power supply in the loop. Next you can have issues with different grounds or negatives, can you tie the power supply negative and the display negative together? If you can't tie them together then you may have to use a 4-20mA loop isolator.

Right. The output is a passive output, isolated from the rest of the display, regarding power supply. But you may be on to something regarding the different ground. Now, since there's negative, it should be of no bigger task than connecting the com on the analog output to negative (a.k.a ground, right?) from the power supply, yeah?

Thanks for the respond =D

If the output from the display is definitely isolated then there shouldn't be a problem with unconnected negatives. Wouldn't connecting the Com on the analog output to negative disturb the 4-20mA loop. It is difficult to answer with knowing more detail. Whenever I have had problems with 4-20mA loops it has been because an output or input that I though was isolated, wasn't actually isolated. Or two negatives that should have been connected weren't or couldn't be. Generally a loop isolator has fixed any issues.

Why dont you tell us what make and type the "display" is ?
Is it secret ?

I agree with BryanG. Assuming that your power supply negative is grounded then your display 4-20mA signal should be floating otherwise you will have problems. You cant have two different points grounded in a series loop. I came across the same problem years ago.

No secret. It just didn't dawn on that such information could be useful. It's a Mettler Toledo IND331. Possibly 131, though.

So unless I find the fault, the suggestion is to use an loop isolator?

Is this the manual you have ?:
http://us.mt.com/us/en/home/support...le/file.res/64067481_R01_IND131-331_UG_EN.pdf

I can understand the confusion, because the analog output is not described at all, apart from that it is an isolated analog output.
edit: And it says "analog signal is isolated and requires that the connected device have an internal resistance of 500 ohms or greater", which cannot be right. It must be maximum 500 ohms.

That's the one ...

Even if the analog output is isolated, it may still be powered internally. Which means you shall NOT supply it with an external 24V.
Remove the external 24V and use a meter to measure if it sends 4-20 mA.
If yes, then you have to connect it to the PLCs analog input and ground.

This looks like a loop powered dispaly. Sometimes they use an internal diode to generate the display power, and at low current this looks like a very high impedence to the signal source. Check out the equivalent resistance - it may exceed the burden capability of the analog output.

It is Mettler Toledo weigh scale.

I suspect it is an isolated active output, powered by the Mettler itself. I'd follow Jespers' suggestion to remove the external 24V and use a meter to measure if it sends 4-20 mA.