Hello.

I am facing a problem in my work. I have one Siemens S5 PLC cabinet to which i connect a number of pressure transmitters (2-wire loop powered). After some new projects the data from these transmitters were needed from another S7-300 PLC cabinet. I decided to use Analog outputs in S5 cabinet in order to reproduse these signals and send them with cables to S7-300 cabinet. I connect the shield only to one panel. After some time working i have some very strange variations in the pressure transmitters indication, not in S5 but in S7 panel (while working in 40% it goes suddenly to 100% and then to 10%, at that time the indication in S5 is 40%). Can you please give me some advice with this problem?

 

Sounds like you have lots of induced noise on your analog lines. Would it make more sense to run a (shielded) communications link between the PLCs to transfer the data?... save you lots of wiring and easily expandable. If not, ensure the relatively sensitive analog signal wiring is very well shielded and not routed between some 600/480V power or other higher voltage electrical wiring... especially VFDs.

Mark L.
ZTR Control Systems

 

Are the analogue systems isolated or do you have a common ground? If you are isolated, is it three way or only two way? Which end is the signal powered from?

Your systems may not be truly isolated and there may be a ground loop. Also, the analogue cards may not work properly if one side of the analogue signal is more than a certain amount above ground. I would suggest looking very
closely at the installation wiring, and reading the hardware specs in detail to see if you have done something which is not permitted. You may need to add loop isolators to cure the problem.

--

************************
Michael Griffin
London, Ont. Canada
************************

 

Hello:

It seems to be voltage reference problem.
Try:

1 - Earthing wire of generous section from one PLC to another (and a good ground connection)
2 - 0V wire from one PLC to another to equalize references.
3 - Shielded cable should be connected to ground each 2 metres aprox (cut the isolating cover at each grounding point and fix it to a metal surface).
4 - Shield connection should be done on the side of the supposed interference source, but you should determine it empirically (take care of Normative, in Spain, i.e., it is not allowed to left one of the sides of the shield unconnected, so you can find double shield cables, one to interferences and one to "normalize" installation)

Each option should be done separately to see which one works better (or not)

Hope useful.

Aquilino Rodriguez

 

Smells like a ground loop (ground or common potential difference between the two boxes) to me.

Meir

 

Yes, I second that. No amount of "EMC/Sheilding" will help you. You'll need to install a 4-20mA isolator/repeater near the S5 and power/ground the link to the S7 panel from the S7 panel.

- LynnL, www.digi.com

 

I am betting the problem was correctly diagnosed by Meir as a ground loop which is being equalized through the electronic earth of the PLC input card as there is inadequate equi-potential bonding. Isolation amplifier is good, big earth wire joining the power supply grounds and preferably another one joining the 0v signals together is better.

Donald P

 

Fisrt of all i want to thank you all for your time and your technical help. I also want to tell you that i have put in the terminals (not the card but the terminals where the cable from S5 is connected) of an analog input in S7-300 panel a non polarised capacitor 1 microfarad. After this i have not experienced any of the troubles that i mentioned earlier. Is there any opinion about that?

 

I would agree with the fact that you are certainly encoutering noise problems. The most successful manner that I have been able to solve this problem is by installing current loop isolators on all analog circuits comming in from instruments in the field and signal isolators on all communications, not only to eliminate noise but also affording PLC and comm port protection.

Also make sure that shields of shielded cables are adequately grounded as close as possible to ZERO potential to ground at the control source cabinet/enclosure and continueous throughout any daisy chains but floated (clean and isolated throughout the entire run and termination points in the field.

Shane F.Reynolds

 

I don't think it is an EMC problem. It happens because internally in the PLC there is an automatic diagnostic circuits that will stroke the output and sense it back to see whether the output still working or not. This is done periodically with a very high frequency pulse. The pulse has no harms to normal final elements. So, your S7 inputs actually picks these pulses periodically. Your capacitor helps in attenuating this pulses and the internal signal conditioning of S7 will filter the signal further.

 

There are 3 possibilities u need to check. i.e...

1. In S7 the respective channel should not be loop powered.
2. It should not be shielded in any panel.
3. Check for the channel working condition of both input, output channel of S7, S5 respectively.