Fault Finding on an IC-SNP / IC3-SNP and associated IAC controllers.
Satchwell MicroNet, Satchwell Sigma
How should testing be carried out where communications between an IC-SNP / IC3-SNP and IAC controllers is not possible.
The purpose of this article is to advise of the methods and techniques available when fault finding on an IC-SNP SubLAN.
Using the Alarm Manager on the Sigma default/backup Server.
- The Alarm Manager is a valuable source of information when investigating a problem SubLAN. All system messages generated by the controller will be displayed here.
- Check system messages (priorities 22 – 30) from the appropriate controller.
Unable to open gateway.txt file.
- This generally indicates that an error has been made in an entry in the gateway.txt file. When initially setting up the gateway.txt file add just one or two entries, as this cuts down the chances of making a mistake.
- Make sure the basic gateway.txt file is known to be good, or download it from Exchange-On-Line.
- If an error is reported, check the file for obvious mistakes, if none can be seen, then make a copy of the file.
- Working on the gateway.txt file, remove all entries, except for one. Download the file and then warm-start the controller. If the file is now processed correctly, then from the backup of gateway.txt, copy a small section of the entries back into the gateway.txt, download and warm-start again. Repeat this process until the problem is located.
Controller??? Is not responding
- A specific controller is not communicating. Investigate the controller locally. Will the controller communicate with Visisat via a LIB-4-485 converter, are the LAN connections correctly connected, does the controller have power?
Multiple controllers reported as not responding
- Dependent on the number of controllers involved, this may indicate a partial or total failure of the SubLAN. Investigation starting at the IC-SNP will be required.
Investigation at the IC-SNP
- The SNP SubLAN, channel 2 should have its BIAS jumper position set to D, for full biasing. The 120-ohm, end of line termination resistors should be fitted.
- If the controller is warm-started, then shortly afterwards the channel 2 TX LED should be seen to start flashing. This proves that the gateway.txt file is being processed. If the channel 2 RX LED flashes, then the IC is receiving replies to its requests.
- If the channel 2 RX LED remains permanently on, then this suggests that an IAC controller communications driver is faulty, and is stopping the LAN communications – see below.
IC-SNP Retry Mechanism
- Having completed a warm or cold-start, the IC-SNP will process the gateway.txt file. Assuming the file format is correct, then the IC-SNP will start to process the text entries.
- The IC-SNP will read each line of text in sequence and try to communicate with the appropriate IAC controller. If the IAC controller does not respond, then the IC-SNP will repeat the message three times before marking the IAC controller as off-line. It will also send an “Unable to Communicate with Controller ???” message to the default server. The IC-SNP will then try the next controller.
- Communications with the off-line controller is retried every 30 minutes, on the hour and half hour.
Testing – IC-SNP channel 2 RX LED is permanently on.
- Try to assess the mid point of the LAN, and instigate a binary chop by separating the twisted pair comm’s at the nearest controller.
- Check the IC-SNP, RX LED.
- If the LED has gone off warm-start the IC-SNP: If communications restarts to the remaining controllers, it suggests that the fault must be in the section of the LAN beyond the mid point, and further investigation must be pursued from there.
- If the LED remains on, it indicates that a fault is still present between the IC-SNP and the mid point of the LAN.
- In both of the above cases, the next action is to split a section of the LAN into two again, to be working on either 25% of the twisted pair, or by adding the first half of the remaining LAN to be working on 75% of the whole LAN.
- Using a voltmeter set to DC volts, measure the voltage on each LAN cable.
- Communications cable +ve and the controller 0volt terminal
- Communications cable -ve and the controller 0volt terminal
- Communications +ve and the Communications -ve
- Below are some example voltages, these are only as a guide, they will vary dependent on LAN conditions.
- Communications cable +ve and the controller 0volt terminal 2.5 volts
- Communications cable -ve and the controller 0volt terminal 0.15 volts
- Communications +ve and Communications –ve 2.31 volts
- When checking voltages, a minus value is a good indication that something is wrong. For example if testing at the 75% position, the fault could be in either direction. If when measuring the voltages, cable 1 gives a reading between +ve and –ve of 1.6 volts and cable 2 gives a reading of –2.4 volts, then cable 2 would be the cable to investigate further.
- The above process is continued until the faulty controller or controllers are located.
- Once it is believed that all problems have been resolved. The full gateway.txt file should be downloaded, and the IC-SNP should be warm started. Allow a few minutes and then check the Alarm manager for any new error messages.
Testing – IC-SNP channel 2 RX LED gives no indication at all
- Having warm-started the controller, channel 2 TX LED should be seen to start flashing. If at this stage channel 2 RX LED does not flash, then this suggests that there could be a problem with the IC-SNP, SNP SubLAN, or the controllers connected to it.
- At this stage there should be fair confidence that the IC itself is working correctly and therefore the controllers or the SubLAN would be suspect. The method for determining the problem will be the same in both cases.
- Identify the first controller on the SubLAN, and disconnect the twisted pair LAN going to the next controller at this point. Warm-start the IC and observe channel 2 RX LED to see if any flashing occurs. If communications with the controller is possible, then it will be necessary to proceed as in the previous examples, using the binary chop method to locate the problem. If the controller will does not communicate with the IC, then this could be a cable fault or the controller could be faulty.
- To test the cable, ensure that both cable ends are disconnected and that the cores are separated. Check the resistance between cores and each core to earth. All values should be infinity. And one cable end, short the cores together. At the other cable end, check the resistance between cores, should be almost zero, and each core to earth, should be infinity.
- To check the controller, connect to it using an LIB-4-485 converter and Visisat Engineering software.