Ground Fault, Yea!!!!

[gar]

100902-1949 EST

nhfire77:

If you have addressable devices this probably means there is some sort of data signal sent out on the signaling wires.

What does "monitored side of a addressable input module" mean? Is this the wire you said is the negative side of the data loop? How does this monitoring wire monitor whatever it monitors?

Keep in mind that my comments on the ground fault detection is only conjecture. The greatest use of this discussion may be to stimulate other ideas.

In your first post you said:

When metering the one side of the leg that I believe is the problem, no ground fault is detected (about 26 Meg Ohm actually, the other three legs are similar in readings not causing a GF, the threshold for GF in the panel is 40K Ohm)

So this was with the pair of signal wires to the loop that has a problem disconnected from the FACP. Did you try a normal DVM resistance check? This is not more than a few volts.

You went on to say:

When it is attached to the FACP, the GF condition returns. However, it randomly goes away, so right now its normal.

How long may the ground fault condition remain. A few seconds, minutes, or hours?

Next you said:

When this occurs and the GF condition is present on the panel but not on my meter, typically the GF is on the monitored side of a addressable input module (kinda like the load side of a circuit, but not.)

This might mean the megger voltage is sufficient to blow or heat away the ground fault.

.

 

[nhfire77]

100902-1949 EST

nhfire77:

If you have addressable devices this probably means there is some sort of data signal sent out on the signaling wires.

.

Yes it is a propriatary signaling system, by Hochiki, called the Hochiki Protocol, and customized for a Silent Knight Branded Panel

What does "monitored side of a addressable input module" mean? Is this the wire you said is the negative side of the data loop? How does this monitoring wire monitor whatever it monitors?


.

Poorly described by me. The monitor module has two sets of terminals. One is for the data loop so it can communicate with the fire alarm panel. The other set is for monitoring a dry set of form B contacts of another device, such as a switch on a tamper device. (normally open with an end of line resistor)

In past incidents, I have found that if I do not have a GF on the data loop that I can see with a DMM, but the panel shows GF when I connect the data loop, the GF is on the contact wiring of the monitored device.

Keep in mind that my comments on the ground fault detection is only conjecture. The greatest use of this discussion may be to stimulate other ideas.

.

I appreciate this, thank you.

In your first post you said:
So this was with the pair of signal wires to the loop that has a problem disconnected from the FACP. Did you try a normal DVM resistance check? This is not more than a few volts.

.

Yes, all the readings were normal across all the legs of the data loop, which leads me to believe it is on the dry contact wiring of a monitor module.

You went on to say:
How long may the ground fault condition remain. A few seconds, minutes, or hours?

.

Hours, normally. Of course, after I played around with the wiring in the control panel it went away on its own. I did check all the wiring in the panel and the adjacent conduit and wireways for potential wires rubbing. No go there.

Next you said:

This might mean the megger voltage is sufficient to blow or heat away the ground fault.

.

No, you said that (and I think you would be right). I believe you meant I said:

"When this occurs and the GF condition is present on the panel but not on my meter, typically the GF is on the monitored side of a addressable input module (kinda like the load side of a circuit, but not.) "


This was referring to the contacts (tamper switch) wiring being monitored.

Here is a link to the install instructions for the device I am talking about, its a small file and I am sure you will understand it:

http://www.silentknight.com/techdocs/SD500AIM_MIMis.pdf

I cannot see the GF at the panel with the meter, but the panel is showing a GF, and it appears it is on the data loop. The panel does not show what circuit has the GF, just that there is one somewhere.

 

[gadfly56]

I cannot see the GF at the panel with the meter, but the panel is showing a GF, and it appears it is on the data loop. The panel does not show what circuit has the GF, just that there is one somewhere.

So I'll repeat my question: Can you replicate the fault by deliberately grounding one of the terminals of another monitor module on the switch side of the module? Same message, same no GF measured on the SLC or panel wiring?

 

[gar]

100903-1207 EST

nhfire77:

I looked at your datasheet and also looked at the Honneywell site. I found no theory of operation anywhere. How are you to troubleshoot without a knowledge of the system theory of operation?

It appears that power, signaling, and supervision are all done over a single wire pair.

To supply power over the SLC pair there has to be a moderate voltage and by implication and previous comments this is DC. To signal over this power pair means the voltage has to be modulated by either the monitoring panel or one or more of the devices connected to the SLC depending upon who is the source of information.

To detect a ground fault requires both SLC wires to float off of ground. Then there has to be a sensing circuit between the floated SLC wires and ground.

On the SD500-AIM/MIM datasheet there is a comment "All Wiring is Supervised, Power Limited". Since only two wires go to each module this means that supervision is based on knowing the nominal steady-state current to the SLC wiring. A failure of a module or a broken wire or the two SLC wires shorted together would cause an abnormal current flow. Supervision would have no effect on ground fault detection.

If the short is a low enough resistance, then maybe TDR (Time Domain Reflectometry) might approximate the short point. See http://en.wikipedia.org/wiki/Time-domain_reflectometer
I took some CAT-5E cable and did a quick test with several hundred feet. 2000 ohms and above did not provide any useful reflection. 249 ohms did. There may be some area between 249 and 0 ohms where there won't be any useful information because of matching the line characteristic impedance. However, at 100 ohms, which is close the line impedance, I could see the reflection signal.

Note: the TDR would be done between ground and the signal wire with the ground fault.

.

 

[nhfire77]

I looked at your datasheet and also looked at the Honneywell site. I found no theory of operation anywhere. How are you to troubleshoot without a knowledge of the system theory of operation?

.

Theory of operation, from what I understand: An Signaling Line Circuit (SLC) is a polling loop. The panel is constantly requesting the status of the device. I believe the code allows upto a 60 second delay. The SLC uses numerical addressing for individual devices, there is maximum of 127 devices on any one loop. Addressing is binary with DIP switches, point 128 being an internal control point and Zero being null. The panel can poll many devices per second, so a delay of 5 seconds would be long

It appears that power, signaling, and supervision are all done over a single wire pair.

.

Yes

To supply power over the SLC pair there has to be a moderate voltage and by implication and previous comments this is DC. To signal over this power pair means the voltage has to be modulated by either the monitoring panel or one or more of the devices connected to the SLC depending upon who is the source of information.

To detect a ground fault requires both SLC wires to float off of ground. Then there has to be a sensing circuit between the floated SLC wires and ground.

.

Yes to all.

On the SD500-AIM/MIM datasheet there is a comment "All Wiring is Supervised, Power Limited". Since only two wires go to each module this means that supervision is based on knowing the nominal steady-state current to the SLC wiring. A failure of a module or a broken wire or the two SLC wires shorted together would cause an abnormal current flow. Supervision would have no effect on ground fault detection.

.

I believe so. If there is a serious enough GF, ground will draw so much current the loop will go down or the devices farthest from the panel will show missing due to the insufficient voltage. The wording you quoted tells me that both sides of the device are supervised for opens, shorts, GF etc.

Power limited lets me know what kind of wiring methods are permissible, and that its safe to mix with other power limited wiring.

If the short is a low enough resistance, then maybe TDR (Time Domain Reflectometry) might approximate the short point. See http://en.wikipedia.org/wiki/Time-domain_reflectometer
I took some CAT-5E cable and did a quick test with several hundred feet. 2000 ohms and above did not provide any useful reflection. 249 ohms did. There may be some area between 249 and 0 ohms where there won't be any useful information because of matching the line characteristic impedance. However, at 100 ohms, which is close the line impedance, I could see the reflection signal.

Note: the TDR would be done between ground and the signal wire with the ground fault.

.

I have a TDR, a simple one used for measuring opens and shorts on a un grounded cable. Are you saying if I had a control set up, say a specific length of wire I could then extrapolate the approximate distance to the GF. If, so how do you use the ground (earth) like that?

 

[gar]

100903-2048 EST

nhfire77:

Looking for a short with the TDR.

Disconnect the two SLC leads. If the TDR has two ungrounded terminals, then ground one. If it has one grounded terminal, then do a redundant ground. What should you ground to? I do not know. But probably building steel or something else that connects to earth close to you. The second TDR terminal goes to the SLC lead you think has the short to ground.

Try an experiment with a couple hundred foot wire across your yard, and not terminated at the far end.

At the speed of light a wire of about 500 ft will see a reflected signal in about 1 microsecond. Light travels about 982 ft/microsecond. In a wire or cable the velocity of propagation is about 60 to 70 % of the speed of light. I laid a length of wire, about 70 ft, on the basement floor. With the far end of the line open the reflection returned in about 200 nanoseconds. Some of the wire was not fully stretched out. With a shorted far end the signal and the ability to interpret the reflection was much more difficult. Coils in the wire mess up the signal. In any event by time correlation with the calculation the result is in the ball park.

I used a function generator and scope for the tests.

Your method of grounding the TDR may be important.

If you get a signal that indicates a possible location, then open the circuit closer to the expected point and run another test. Note: you have to calibrate the TDR by some shorted test points on your actual wire.

.