Issue with Input Expander


Active Member
I have an input expander connected to an M1-DBH that is causing a lot of log issues with System Restart.  Note that the zones on the expander appear to function correctly but I get these in the logs all the time:
1367 = SYSTEM START UP Input Exp 3 this happens about every 3-5 minutes.
I have 4 devices plugged into the data bus with the terminating plugged into the last unused port.  Also J3 on the Elk is in place for termination.
No termination on the expanders, I measure 67ohms across A and B for the M1-DBH, Input Expander and Output Expander.  The other two devices on the bus are a wireless receiver and a keypad both are difficult to access to measure the resistance but those 2 have been in place for much longer with no issue.
So in reviewing everything over, I reviewed the wiring instructions again for the M1-DBH and noticed they tell you to use 568A which stood out cause I always wire 568B.  So I double checked and sure enough all of my CAT5 going into the M1-DBH is wired for 568B.  The effect of this is the A line and the B line is swapped on all the devices but since I am consistent I wonder if I could just swap the A and B connection between the M1 and the M1-DBH instead of correcting ALL of the other lines?
So this really should not matter since it just swaps them at one end the swapped lines are tied together at the device end so it just changes which wire is in the "return" loop but electrically should be identical.
If this matters I am surprised i did not have issues before I added the input expander.
EDIT: Reviewed closer and my initial though of A and B swapped was incorrect.
568A vs 568B pinout doesn't matter, as long as you are consistent at both ends of the cable.  It's just swapping the pairs of wires between pins 1&2 and 3&6.   The electrons don't know the color of the pair of wires they are travelling on (at least at RS-485 frequencies)  :)
67 ohms is the right resistance for the data bus.  Since the data bus is essentially one long pair of wires with a terminator at each far end, it should measure about the same no matter where you take the  reading.
The first thing I would do is reterminate all the RJ45s.   A less than perfect crimp is the most common source of problems.  Also reseat all the screw terminal connections. 
Thanks Ral, will work on terminating the RJ45s.  I also checked the voltage at the boards and it is at 13.85V so that is good and measures the same at each of them.
More info if useful.
While the boards are in a Input Ex and Output Ex are in a separate enclosure from the M1 it is right next to it and the data runs are less then a few feet.
Question?  does the Input Exp 3 on the log indicate that it was the cause of the restart or is what restarted?  Since I do not notice anything like keypads flickering or anything is it just the input expander board that is restarting? 
Also checked the System Diag.
Data Bus errors:
Retry T1A1: 00000
Retry T2A2: 00009
Retry T2A3: 00009
These do not appear to be changing.
It reports System Voltage at 13.8V and Current at 0.4A.
Replaced the RJ45 jacks and made sure the terminal blocks connections were secure.
Still have the issue.  I even rearranged the connections on the M1-DBH to see if the problem board would change (if it is not the Input Expander having a problem but another device ahead of it).  No changes.
Any other ideas?  How likely is it that I just have a defective board?
Other info - I noticed that ElkRP while it sees the correct voltages for my zones connected to the Input Expander board, it takes a few minutes before it registers the STATE of the zones after a restart.  Is this normal behavior?
I haven't seen a clear answer from Elk about the meaning of the System start up message, but it seems to suggest that the expansion board was the cause. 
I don't think it should take several minutes for it to see the state of the zones on the expansion board.  A long time lag is normal for wireless zones, but I don't recall seeing it on a hardwired expansion board.  But I can't say I looked for that, either. 
You said your voltages seem ok.  One thing to check is to measure the supply voltage directly on the expansion board with a multimeter.  A low voltage at the board can cause communication problems over the data bus, which might also explain the time lag on the zone status. 
Have you ever done a power supply load calculation for the M1 to make sure you are not overloading the power supply? 
What firmware level is your M1 running?  Some older firmware levels had issues with system restarts. 
Using 5.3.10 firmware.
I did measure the voltages AT the expansion boards
I have not done a manual calculation however with the system board reading 0.4A that seems to be well within the limits.
I did start reading more of the Elk Wireless expansion and noticed some intersting requirements/recommendations they have.
1 - the 1st wireless board MUST be on address 2 - check that is what I have.
2 - Not clear but it sounds like they do not want you to have an Input Expander on address 3 even if you dont plan to use address 3 for other wireless.  Odd, and mine is on address 3 so I am going to move it out to address 11 to also make sure its zones do not overlap with any potential wireless zone.
Other thing I noticed and not sure why i did this (setup wireless about 4 years ago) is my wireless zone is in Group 4 not Group 2, note that my input expander is in Group 3 (address 3).  So maybe this is a source of my issues.
Will get back once I change the input expander address.
How many wireless zones are you actually using?  Elk wants any M1XIN zones to be above the last wireless zone that is in use.  So if you have a wireless zone in group 4, I'm not sure what would happen with the M1XIN at address group 3.  As long as you don't also have a wireless zones in that group, it might work, but better not to overlap things. 
As long as you are changing addresses, I'd recommend moving the M1XIN up to the high end (e.g. address group 12) and leave a gap in the middle.  That way, it allows you to expand the wireless zones up and the hardwired zones down without having to reassign existing zones if you ever add more zones.
Just before I read you message I moved it to Address 11 and since I only have wireless in Group 4 I should be fine.
However it does not appear to help anything.  I have received 3 1367 logs since I restarted the panel 10 minutes ago.
I guess I would next go though a process of elimination to try and narrow down the cause. 
Try connecting the zone expander directly to the M1 and bypass the DBH.  You'll need to change your terminator placement for this. 
If the zone expander is close to the M1 (or it's not too difficult to run a long, temporary cable), try replacing the cable. 
If you still get restarts, remove the zone expander from the data bus and see if the restarts disappear.  If so, then I'd suspect the zone expander as the bad guy. 
So for the first step I could just move the Input Expander from the DBH to be in parallel with the DBH and add a termination jumper to the Zone Expander correct?  This should not be hard to do they are not far from each other.
hagak said:
So for the first step I could just move the Input Expander from the DBH to be in parallel with the DBH and add a termination jumper to the Zone Expander correct?  This should not be hard to do they are not far from each other.
I'm assuming you have all your devices connected through the DBH, which forms a single branch of the data bus.  When you connect the zone expander directly to the M1, you will then have 2 branches on the data bus. The first branch is to the DBH and and the second branch is to the zone expander.   Each branch needs a terminator, so you would move the terminator from JP3 on the M1 to the zone expander. 
Ok so I am getting close to just giving up on this, currently the only issue I can see is it takes a few minutes for the zones in the input expander to be "seen" and I get those log messages.  But other then that things function.
I did put the input expander out on its own branch as describe, problem continued. 
I moved the output expander from the DBH to the branch and the problem still continued. 
I removed the input expander leaving the output expander and no more issue.
I removed the output expander but left the input expander - issue still happens
I am reluctant to remove the wireless and keypad from DBH since I do not have an easy way to re-wire them.  However in debugging this I did notice that the wireless receiver was showing a few bus errors so I did double check its wiring and noticed that the B line Return wire was loose, so I corrected this.  The bus errors have gone away but still have the System Restart on the Input Expander stays!
Seeing that the only time this log occurs is if the input expander is connected, that leads me to think the issue is with the Input expander.  Note because of moving it to a branch circuit I also went to a different cable for it so it has had 2 different cables entirely. 

Seems like if using the DBH it would be really handy to have an RJ45 socket to terminal block so that you can just have 4 properly guaged wires for the terminal blocks on the different devices. CAT5 and CAT6 wire is not really the correct wire for the terminal blocks they have and it is even worse that you need to have 2 wires in the A and B terminals.

Found this: still have to tie the A/B and A1/B1 lines together (would be nice if elk supplied one just for this purpose).