Potential problems in analog control system

O

Thread Starter

Olle

I'm writing my master thesis and as a part of that I'm implementing a control system for a VA-process in a pilot plant.

The main "controller" is a PC running LabView with three external usb modules attached to it. Two analog input modules (USB 4718 from Advantech) and one analog output module (NI-9263 from National Instruments). 12 different current-signals from instruments are wired into the input modules. 4 different control-signals (0-10V) are generated on the output-module.

Electrical equipment:
sensors for pH, DO, NH4 etc. 4-20 mA
electrical stirrer with variable speed drive (reactive load)
pumps (reactive load?) 0-10 V control signal
PID-controller (0-10 V reference signal)
control-valves (4-20 mA control signal)
etc.

The electrical part of the plant is not made in an optimal way, for example it draws it's power from 4 different outlets (the reason for this, or so I'm told, is that the residual current device will break the power if one outlet is used). The system contains reactive loads (frequency transformers), that is plugged into the same system as some of the instrumentation.

The result seems to be that there is:
a, DC - potential differences between different parts of the control system (up to 40V).

b, AC - potential of 17 V rms as well betwen different parts.

Despite this, the Advantech modules have been working relatively well until I connected the 4 outgoing control signals. When I did, the advantech modules stopped working.

Questions:

How do I get rid of these strange potentials?

I have thought of some kind of isolators/zenerbarriers but I don't know what to choose. And will the AC-part of the problem still be there.

If I can, how should I separate the different parts in order to remove these potential differences?
 
I am not sure to understand each and every parts of your problem but if I understand well you seems to have power supply problem coming from four different sources of supply, is that so?

but If I still understand well this is only to give ONE source of supply to your device. I am still right?

In that case just a though, why don't you use a multi-input power transformer with 4 inputs (assumed 220vac) and one output (I guess 220vac).

this is of course just a suggestion as I told you again I am not sure I fully understood the problem.

If power source are DC it is even easier using power rectifiers (diodes)

Pascal
 
>The electrical part of the plant is not made in an optimal way, for example it draws it's power from 4 different outlets (the reason for this, or so I'm
>told, is that the residual current device will break the power if one outlet is used). The system contains reactive loads (frequency transformers), that is plugged into the same system as
>some of the instrumentation.

Well, there's your problem. Sort out your power supplies and separate your instrumentation power from your equipment power. Get your instrumentation onto a common earth to prevent ground loops. Search for Signal Conditioners and Isolators on google to find a local supplier.

Rob
 
Top