How To Check Wiring Continuity.

Status
Not open for further replies.

goodoboy

Senior Member
Location
Houston
Hello,

I have a #14 one pair (red positive, black negative, and shield) cable is about 1000ft long connecting to an instrumentation in the field from a control systems sending 4-20mA (HART signal). I am getting intermittenly communication errors. I believe there is noise interfernce with the signal, but I want check the cable intergrity first.

How do I check the continutiy of the cable with ohm meter which such long cables from one end to the other?

Thank you
 

petersonra

Senior Member
Location
Northern illinois
Occupation
engineer
Hello,

I have a #14 one pair (red positive, black negative, and shield) cable is about 1000ft long connecting to an instrumentation in the field from a control systems sending 4-20mA (HART signal). I am getting intermittenly communication errors. I believe there is noise interfernce with the signal, but I want check the cable intergrity first.

How do I check the continutiy of the cable with ohm meter which such long cables from one end to the other?

Thank you

Disconnect cable at both ends. Short together at one end. Measure resistance at other end through both conductors in series. 2000 feet of #14 should be about 5 Ohms. Don't be surprised if your common VOM does not like doing this. It is using a very low voltage.

A better test is probably some kind of higher voltage resistance test but you will need a more expensive meter for that.

If you are getting any kind of signal, the cable might be OK.

Check any terminals along the way. Sometimes these things go through a bunch of terminations between the field device and the analog receiver and you might have an intermittent problem along the way.

What do you mean by "communication errors"? Are you getting actual Hart comm errors? The log files your DCS keeps can be helpful in running these kind of things down.

I find that usually when people blame "noise" or "bad grounds" it means they just have not been able to find the real problem.
 

kwired

Electron manager
Location
NE Nebraska
How is the shield bonded? Is it bonded on one end or both ends?

Is this one continuous length of cable between components?
 

don_resqcapt19

Moderator
Staff member
Location
Illinois
Occupation
retired electrician
Check your DC power supply for AC ripple. HART does not like it if there is much AC ripple on the circuit. I have run into this a couple of times where the filter capacitor on the DC power supply had failed.
 

goodoboy

Senior Member
Location
Houston
Disconnect cable at both ends. Short together at one end. Measure resistance at other end through both conductors in series. 2000 feet of #14 should be about 5 Ohms. Don't be surprised if your common VOM does not like doing this. It is using a very low voltage.

A better test is probably some kind of higher voltage resistance test but you will need a more expensive meter for that.

If you are getting any kind of signal, the cable might be OK.

Check any terminals along the way. Sometimes these things go through a bunch of terminations between the field device and the analog receiver and you might have an intermittent problem along the way.

What do you mean by "communication errors"? Are you getting actual Hart comm errors? The log files your DCS keeps can be helpful in running these kind of things down.

I find that usually when people blame "noise" or "bad grounds" it means they just have not been able to find the real problem.


Thank you petersonra,

I will use a megger to test the resistance.

Do I meg each terminals to make sure they are not shorted to each other?

The DCS shows intermittenly HART communications errors that occur about every 5-15 minutes. The valves still controls from the 4-20mA, but I lose the HART every 5 -10 minutes.

I want to completely make sure this cable has good intergrity. Because the shield to ground should isoloate any noise.
 

rlundsrud

Senior Member
Location
chicago, il, USA
I do a lot of temperature controls that use an rs485 Bacnet communication line. I made a simple time domain reflectometer to test the com loop before I turn it over to the tech for commissioning. If you Google building a TDR for testing cable, there are quite a few nice youtube videos. I made my TDR with a Schmitt trigger, it cost less than 10 bucks but you would need a scope to view the signal. I like using it because I can test cable impedance, verify that there are no T tap's and measure the length of the com loop. If there is a failure in the line I can tell how far it is to the failure to speed locating it. I know it is a bit technical, but if you can use a scope it is pretty handy.

Bob
 

goodoboy

Senior Member
Location
Houston
Check your DC power supply for AC ripple. HART does not like it if there is much AC ripple on the circuit. I have run into this a couple of times where the filter capacitor on the DC power supply had failed.

Thank you don_reqcapt19,

Glad to see I am not the only one. I will check the 24VDC that is used to power the DCS Anolog Output Card to make sure we have clean 24VDC signal used to power the HARD device.
 

goodoboy

Senior Member
Location
Houston
is the shield bonded to EGC/ other grounded components at one end or at both ends of the cable?

Thanks kwired,

The shield is only terminated to the ground bus at the DCS I/O cabinet. The shield is not grounded on the instrument side, it is wrapped and taped around the cable.
 

goodoboy

Senior Member
Location
Houston
I do a lot of temperature controls that use an rs485 Bacnet communication line. I made a simple time domain reflectometer to test the com loop before I turn it over to the tech for commissioning. If you Google building a TDR for testing cable, there are quite a few nice youtube videos. I made my TDR with a Schmitt trigger, it cost less than 10 bucks but you would need a scope to view the signal. I like using it because I can test cable impedance, verify that there are no T tap's and measure the length of the com loop. If there is a failure in the line I can tell how far it is to the failure to speed locating it. I know it is a bit technical, but if you can use a scope it is pretty handy.

Bob

Thank you very much,

I am renting an oscilloscope today to view the 4-20mA HART signal.
 

don_resqcapt19

Moderator
Staff member
Location
Illinois
Occupation
retired electrician
Given that the analog 4-20 mA control loop is functioning correctly and it is just the HART communications that is having the problem, I would not expect that you will find a problem with the cable itself.
 

goodoboy

Senior Member
Location
Houston
Given that the analog 4-20 mA control loop is functioning correctly and it is just the HART communications that is having the problem, I would not expect that you will find a problem with the cable itself.

Thank you don_resqcapt19,

The HART signals are very sentive to noise on the wire. Performing the cable test (megging the wires from end-to-end) I am checking to send if the sheild did not break off somewhere in the cable. If the shield is broken off somewhere then this could cause HART communication errors.

However, this errors are not constant, they come and go about every 10 minutes.
 

petersonra

Senior Member
Location
Northern illinois
Occupation
engineer
How do you calculate 5ohms? cable spec?

After disconnecting on both ends, can I check for open circuit between Red-Black, Red-Shield, and Black-Shield?

2000 ft of #14 is just about 5 ohms. there are charts and calculators on the Internet for such things.

you can check but my guess is you will find the cable is fine.

4-20mA is a very robust signal and highly resistant to noise.

Hart is a little less robust, but usually works pretty well. However, on long distances I have heard it can be affected by cable capacitance. I would not think 1000 feet would be enough to matter any.
 
Last edited:

goodoboy

Senior Member
Location
Houston
2000 ft of #14 is just about 5 ohms. there are charts and calculators on the Internet for such things.

you can check but my guess is you will find the cable is fine.

4-20mA is a very robust signal and highly resistant to noise.

Hart is a little less robust, but usually works pretty well. However, on long distances I have heard it can be affected by cable capacitance. I would not think 1000 feet would be enough to matter any.

Thank you.

I was thinking cable capacitance was building up on cable. Is there anyway I can confirm this capacitance on the cable?

I am reading the device manual in the field and here is what it says about Cable Requirements:

"This communication signal is superimposedon the 4-20 mA current signal. The two frequencies used by theHART protocol are 1200 Hz and 2200 Hz. In order to preventdistortion of the HART communication signal, cable capacitanceand cable length restrictions must be calculated. The cable lengthmust be limited if the capacitance is too high. Selecting a cablewith lower capacitance/foot rating will allow longer cable runs.In addition to the cable capacitance, the network resistance also affects the allowable cable length."
 

rlundsrud

Senior Member
Location
chicago, il, USA
If you rent a scope you will want one that has a bandwidth of at least 200 MHZ or you will limit the resolution of the signal reflection gap. If you can rent a function generator with the scope use a square wave of at least 5 kHz. the function generator should have a rise time of less than 2 ns or this will affect resolution as well. Do the test with the cable end open and measure from the leading edge of the wave to the point where the inflection/slope changes from positive to negative. For 1000 ft of cable I would expect that to be between 33 and 36 nano seconds. Next put a terminating resistor at the end of the cable that matches the cable impedance, I am guessing that it is between 100 and 120 ohms. If the cable is ok you should get a nice clean square wave that should look the same as the one being generated.

One other thing, you can purchase a USB oscilloscope that ranges from 100.00 and up. While these aren't fast enough to do TDR measurements without spending 1000.00 or more they can be very useful in looking at the signal to see if there is any degradation. They also have the advantage of being isolated from any system grounds so they can be used on active communication buses without crashin the data. Some will decode packet data and put it in a readable format. If you check out the USBEE scopes you will see what I am referring to. They are combination scopes, function generator, network analyzer, protocol decoders. They will even dice a salad nicely. If you have any questions feel free to message me and I will get back to you as soon as I can.


Bob
 
Status
Not open for further replies.
Top