Ground rod noise

[tom baker]

OK what type of equipment does the direct lines operate?
Is it 4-20 mA or audio tone?
Who owns the direct lines?
When the fuse blows you have 500 mA + current.
My audio tone circuits use a 1/16 amp fuse.
What type of cable is it? Telephone or shielded twisted pair?

 

[umagooma]

OK what type of equipment does the direct lines operate?
Is it 4-20 mA or audio tone?
Who owns the direct lines?
When the fuse blows you have 500 mA + current.
My audio tone circuits use a 1/16 amp fuse.
What type of cable is it? Telephone or shielded twisted pair?

Tom, I said direct line but this is a leased line from Embarq(Sprint). We are using 4-20mA on leased lines. We are using 1/8 amp fuses and it is shielded telephone wire, no twisted pair. Should both ends of the wire be bonded to ground?

 

[tom baker]

4-20 mA is typically immune to noise, as the noise is introduced equally into to each wire, as its twisted.
The cable shield should only be bonded at the end with the power supply. At the instrument end, bond the cable shield and instrument cable shield together.
Put a scope on the 4-20 MA and look for noise. You shouldn't see anything but DC. Go to a location that works and look at the noise there, to compare to the problem site.
Its possible that the problem is on the Telco side. Direct wire circuits like are hard for them to troubleshoot, they can't do it from the CO, and the techs probably won't know how, its simple but an uncommon circuit so most won't have seen it.
You may have to workk with the Telco to fix.
I always verify a telco problem exists before I call the Telco.
Telco tech stated recently that they don't bond the shield on drop wires for new installs. It may be as simple as an unbonded shield.
With 4-20 the shield is only bonded at one end, otherwise you get ground loops.
Also check the instrument. Is the 4-20 ma cable isolated or is one wire bonded at the instrument?

 

[dereckbc]

Should both ends of the wire be bonded to ground?

There is no clear cut answer to the question.

Typically if the lines are contained within the plant, you bond the shield on one end only at the terminal equipment. For example using a PLC the shield are terminated at the PLC and left open at the sensors.

However if coming from the outside as in your case it can work either way. I can tell you Sprint terminates their OSP cable at every pedestal if ran underground or every 3rh or 4th pole overhead. What I suggest is what I told you earlier. When the cable enters the building break and terminate the shield to ground. The with your in-house cable leave the shield open where it is X-connected to the OSP cable and terminate the shield at the equipment end. You can experiment and see what works best. If no difference leave both ends terminated.

 

[umagooma]

Thanks guys, You have been a world of help!!! I am on the right track now for sure! Will let you know how I fare the next time we have a good storm. Great website, love this place:grin:

 

[tom baker]

here is how I would trouble shoot this:
Install a spare level transmitter to see if its the transmitter or not.
See if the teleco can swap cable pairs.
Or take the transmitter from that site and try it somewhere else'

Of course if you have o-scope try looking at the cable and mA.
If the cable has a line protector, what kind is it? I have seen some old sites with carbons that were intermittent to ground. Those can be replaced with gas tubes.