Data Bus Errors in output expander

hgupta1

Active Member
I've been getting data bus errors for my output expander located in the garage, and I am at wits end trying to figure out the cause.  I suspect that is why my system has randomly rebooting, sometimes a couple times a day.
 
 
The detached garage is about 150 feet from my house, and the enclosure in the garage houses one input expander, one output expander, and auxilliary power.  I have direct burial cat6 cable running between garage enclosure and the ELK M1 control panel.  
 
Currently, the wires look like this:
 
green (data A) and white (data B) and black (negative)  go out to garage each using a strand of the cat6 cable from the Elk M1DBHR and connect directly to the Elk M1XOVER.  Red (12v) and negative from the AUX battery also connect to the Elk M1XOVR.  The screw down terminals for the ELK M1XOVR also have a short  4 conductor wire that takes the red, green, white, and black from the Elk M1XOVR and brings each one to the Elk M1XIN in its respective spot.
 
 The jumper is terminated on the ELK M1XIN but not the Elk M1XOVR. If that seems confusing, I can take a picture.
 
I am getting about 100 data bus errors per hour on the Elk M1XOVR, but none on the ELK M1XIN.  Is there something I can check to fix these errors?  Or is there a better way to wire these? Unfortunately, I don't think ELk did a good job explaining how this should look.
 
 
 
Debugging data bus issues can be a challenge.
 
A few questions...
 
You said you have a terminator on the M1XIN.  Do you also have a second terminator jumper on the M1DBHR for this branch?  Or does the M1DBHR have a second leg on that branch that is also terminated?
 
Are you using only 4 wires in the Cat6 cable?  Or did you connect up 6 of the wires to the M1XOVR, as you would do if you had a M1DBH instead of a M1DBHR?
 
How deep is the Cat6 cable buried?  Is it in conduit, or directly buried in the ground?  Buried in a bed of sand?  Often, rocks can move around over time and cut in to a cable.
 
You mentioned an aux battery.  Do you mean an aux power supply that has its own battery?  
 
Have you checked the voltage on the +12V terminals to make sure the power is ok?  Hopefully, you've got about 13.8V there.
 
The first thing I would check is all the data bus connections.  It sounds like you have 2 wires under each data bus connector screw on the M1XOVR.   It can be difficult to have both wires make good contact.   You could instead connect the wires together with a wire nut along with a pigtail wire that connects to the M1XOVR terminals.
 
If the connections all seem solid, I would try disconnecting one of the expansion boards and test each one connected to the Cat6 cable individually.   If the M1XOVR still gives errors that way, try moving it back to the house and connect it directly to the M1 there.  If you still get errors, then there is probably something wrong with the M1XOVR. If no errors, then the wiring would seem to be part of the problem.

One other thing to try is moving the branch that goes to the garage to a different branch connector on the M1DBHR and see if the errors stay with the garage branch or stay with the one branch connector.
 
RAL said:
Debugging data bus issues can be a challenge.
 
A few questions...
 
You said you have a terminator on the M1XIN.  Do you also have a second terminator jumper on the M1DBHR for this branch?  Or does the M1DBHR have a second leg on that branch that is also terminated?
 
Are you using only 4 wires in the Cat6 cable?  Or did you connect up 6 of the wires to the M1XOVR, as you would do if you had a M1DBH instead of a M1DBHR?
 
How deep is the Cat6 cable buried?  Is it in conduit, or directly buried in the ground?  Buried in a bed of sand?  Often, rocks can move around over time and cut in to a cable.
 
You mentioned an aux battery.  Do you mean an aux power supply that has its own battery?  
 
Have you checked the voltage on the +12V terminals to make sure the power is ok?  Hopefully, you've got about 13.8V there.
 
The first thing I would check is all the data bus connections.  It sounds like you have 2 wires under each data bus connector screw on the M1XOVR.   It can be difficult to have both wires make good contact.   You could instead connect the wires together with a wire nut along with a pigtail wire that connects to the M1XOVR terminals.
 
If the connections all seem solid, I would try disconnecting one of the expansion boards and test each one connected to the Cat6 cable individually.   If the M1XOVR still gives errors that way, try moving it back to the house and connect it directly to the M1 there.  If you still get errors, then there is probably something wrong with the M1XOVR. If no errors, then the wiring would seem to be part of the problem.

One other thing to try is moving the branch that goes to the garage to a different branch connector on the M1DBHR and see if the errors stay with the garage branch or stay with the one branch connector.
 
yes, second terminator jumper is on the M1DBHR for this branch
 
I am using just three of the wires out of 8 in the cat6 cable (orange (Data a), orange and white (Data B), and brown for Neg).   I am not running power out there and instead use an AUX power supply with its own battery.  (Elk P1812)  The cat 6 is not in conduit, just buried a few inches in dirt.  It's possible that it now has a nick in it somewhere.   I am getting my electrical service buried soon and the directional bore guy said that I can add additional conduit to the bore.  I am getting fiber run in the adjacent conduit,  but I am afraid that AC current would wreak havoc in the RS485 bus if I pulled wire for the Elk in an adjacent conduit.
 
The voltage on the 12v terminals is fine.  
 
I will try your suggestions and see if I can reduce some of the errors.  
 
Thanks!
 
A few things.
1) If you used unshielded cable, then there is potential for interference
2) Power is being supplied by two different sources; there is a possibility of voltage differential
 
In the Interlogix world, they have a bus extender and the bus extender has its own power supply but that is only used for the devices connected to the extension bus.  The actual card is powered by the bus from the alarm panel.  This is done to prevent voltage differentials.  It is far easier to have a bridge between the two data buses but they are electrically isolated to prevent errors.  So I'm wondering if this is part of your problem.  A common ground is used to prevent voltage differential.  I don't believe Elk uses a ground in their data bus though or at least it is not explicitly called a ground.  I just see Data A/B and then the third wire is negative.
 
lanbrown said:
A few things.
1) If you used unshielded cable, then there is potential for interference
2) Power is being supplied by two different sources; there is a possibility of voltage differential
 
In the Interlogix world, they have a bus extender and the bus extender has its own power supply but that is only used for the devices connected to the extension bus.  The actual card is powered by the bus from the alarm panel.  This is done to prevent voltage differentials.  It is far easier to have a bridge between the two data buses but they are electrically isolated to prevent errors.  So I'm wondering if this is part of your problem.  A common ground is used to prevent voltage differential.  I don't believe Elk uses a ground in their data bus though or at least it is not explicitly called a ground.  I just see Data A/B and then the third wire is negative.
The RS485 data bus is a differential bus, which helps improve noise immunity.  Shielded cable would help further, and that's something to consider for future replacement. 
 
Having two different power supplies can create common mode voltage problems, but properly connecting the negative sides of the two power supplies will help eliminate or at least reduce that problem, and I think he's got that.
 
Back
Top