Simplexgrinnell 4010ES IDNET Ground Fault

I am a complete novice and am trying to locate a positive earth ground fault on an IDNET “loop” of a Simplex 4010ES FACP. I use the term loop lightly as this seems to be wired more like a tree than anything that could possibly be considered a loop.

After reading many guides on finding the fault, I started at one of the pull stations and checked wires for resistance to ground using a Multimeter. With the mindset that if the multimeter shows an open loop, it was not my problem. I did this through every device/junction box until i found what I thought was my culprit. I replaced the wire and still had the fault on the new wire. I backed the newly installed cable back out of the ceiling until I had an open loop. I then re pulled the wire a few feet away to limit interference from outside sources. When completed, both sides of the wire were open loop.

After connecting everything back together and powering the alarm back up, the fault was back. At the junction box with the new wire, I checked all 3 wires independently of each other and they are all open loop. If I reconnect them all together, I have a read of about 25 Ohms. This seems to challenge everything I know about wiring so I am hoping that someone here might have some experience that could explain/ point me in the right direction for troubleshooting this.

Other pieces of information:

  • around 75 devices (pull stations, HVAC cut offs, conveyor cut offs, smoke detectors, heat detectors, Sprinkler system detectors, and a few other things).
  • All wire splicing is done in junction boxes with wire nuts. The majority of the wire is about 40’ off of the ground and is not in conduit. Wire that is withing reasonable reach is in conduit.
  • Not sure what if anything this means but I did notice that If I check for resistance on the FACP while powered up between ground and the negative wire of the loop, the alert goes away.

When it comes to a commercial fire alarm system, troubleshooting should be left to a licensed, qualified professional as a host of legal issues can arise if work is performed incorrectly and the system fails to perform in or worse, causes, a fire.
That disclaimer aside, there’s a few factors that could be at play here.

First off, if the fault does not go away when you disconnect the loop at the panel, you’ve isolated the fault to the panel itself. Check for metal shavings or wire snippets touching the circuit board and/or chassis (with the power off of course). Seems like from your description, that this isn’t the issue.
In no particular order, these are a few potential causes/suggested troubleshooting steps.

Your multimeter likely is using a reference of a few volts (9-12?) when you’re checking resistance/continuity between a wire and ground. The FACP is using its internal 24-32V power supply to check the same. Should the fault be a high-resistance type (i.e. slightly flooded junction box allowing current leak from a wire; a slightly exposed, tarnished conductor just barely touching a grounded surface, etc), it’s possible that your multimeter’s internal voltage isn’t capable of pushing current through the fault connection and thus it is unable to pick up on a fault that the panel, using its 24-32V signal, can see.

You’re right about “loops” not being loops by the way - if this is an SLC, t-taps are allowed (so long as the circuit isn’t wired class-A), which is why lots of times Simplex refers to them as “IDNET ‘channels’” although I’m not too familiar with their terminology.

Sometimes, the best way to find the ground fault is to find a wiring diagram for the system, and disconnect sections of the loop until the fault disappears. Once you disconnect a branch of the circuit and the ground fault goes away, you know you’ve isolated the leg where the fault is. If you reconnect it and then a few more devices down the line, disconnect and check for the fault again, you can keep halving the list of potential culprits in this manner until you’ve found the device causing the problem. Also remember that the fault may not just be a wire-box or wire-wire connection issue. I have seen cases where an internally damaged pull station will leak current through the switch or terminal block through the metal backplate into ground (if it’s connected to a grounded junction box). Sometimes this will happen if a screw terminal or mounting screw has been over-tightened

Sounds like I need to just use the Control panel over testing it on its own. Thanks for the help, I truly appreciate it.

Theboginator is correct that we recommend using licensed and certified technicians when troubleshooting active fire alarm systems protecting life and property. However, I will make a few observations on your problem.

I am curious of your reading of 25 ohms in some condition. To the system 25 ohms from a conductor to ground is a “dead short.” The system is calibrated to indicate a ground fault from a conductor to ground at 50K ohms.

You did not mention the wiring type. If it is using twisted shielded cable there could be a short inside a cable from a conductor to the shield. If the shield drain wire is landed on ground at the panel that would cause the panel to show a ground fault.

Meters can be your best friend or worst enemy. In searching for a ground fault I often used my old analog meter instead of a digital meter. The typical digital meter uses a low power ohmmeter. They are designed to be able to correctly measure resistors in an electronic circuit. To do that they use a voltage below the forward junction potential of a diode or transistor. That is around 0.7 volts.

I checked two digital meters I have. One is an Ideal 61-340 which is about a $50 meter. The other is a Mastech M300 which can be found for as low as $5. Both have low power ohmmeters. The Ideal puts out 0.25 volt, and the Mastech 0.65 volt. Theboginator is correct that the meter may not find a ground the panel can find, especially a digital meter.

I like to know what I am looking for. I always measured the voltage to ground from the positive and negative side of the circuit first. The IDNet runs at 28 volts so the voltage should split about in half from each side to ground. A tilt in those voltage measurement will tell if the ground is positive or negative and how much it is grounded.

By the way, the IDNet term came from the devices have an identification bit in their digital communications with the panel. That is used to verify that the devices in the field match the panel programming. An address programmed to be an addressable pull station will give a trouble if any other hardware device is set to that address instead of a pull station.