Controller stops responding when another controller is plugged in

[MSB]

We've been doing a light show for the last 10 years and this year has me baffled. Our overall layout is pretty much the same but we did add additional network cabling this year to allow us to create auxiliary enhanced networks to take advantage of the S5 programming features. Also we are running our of UID's due to the additional Pixie II, 8, and 16 controllers we've added. Most of our "dumb" controllers are the 16 channel Gen3 pro version. Last year we had a successful year learning S5 and we're now on 5.6.6 at the Pro level. one display area that we are finishing plugging in for this season has been there for 4 or 5 years. Pretty simple. On one side of the road there are LOR160xW g3 controllers 6A, 6B, 6D and on the other side of the road there are two more 6E & 6F. They control tree wraps on 80 trees made from led string lights with 4 colors Red, Green, Blue, and Yellow.  with 500 lights per prop and this has not changed. In the past, we repeated the regular (not enhanced) network on both sides of the road and things worked just fine. This year we buried 2 CAT 6 cables and now there is an enhanced Aux C network at 500k on once side of the road and an Enhanced Aux D network on the other side of the road. I have a preview created just for this area for testing. All of the props are numbered correctly and I's still using the same UID's. In the s5 preview I created a horizonal stack align bottoms group for each color. I also createdb2 groups (one for each side of the road) with a horizonal stack align bottoms. Lastly I created the last group as a vertical stack using each side of the road (top and bottom).

We were plugging in the controllers tonight and something very weird happened. On one side of the road is Aux D. The first controller is 6C and that lit up with the preview running. We moved to the next controller, 6B and plugged that one in and 6C stopped working and nothing in section 6B lit up. We then moved to 6A and plugged that in and got the same results as 6B. We went back to controller 6C and there was a no conn message. We placed a coupler at 6C and sections B and A started to light up but there was a big delay. and definitely was not running the sequence properly.

ALL network cables including patch cables have been tested and qualified with a very expensive Fluke Cable IQ tester. Unplugging the 6B controller and Section 6C started kinda working again and the no conn went away on controller 6C.

We moved to the other side of the road and plugged in 6E and that section lit up. We then moved to controller 6D and plugged it into network and 6E stopped responding. Both controllers on this side had connectivity.

Move to the PC and did a hardware scan on each network and found all 5 controllers including the controller showing No conn on the display.. 'm at a loss at what to do at this point. Tomorrow I will try changing the preview and switching everything back to the regular network. any Ideas?

[k6ccc]

Real quick answer because I need to get to bed.  Since you have verified the cables as OK, very carefully inspect the RJ-45 connectors on the controllers.  What you are looking for is pins bent or crossed over.  If you find that, use a very small hook to move the pins back in place.  If your Fluke tester can do this (and it should), test for proper connectivity between the two RJ-45 connectors on each controller.  For at least the first side of the road, 6B is my suspect.  I'm too tired to follow the other side...

If that does not help, longer answers available when awake...

 

[TheDucks]

Fluke makes some nice equipment. that said, it also implies the run is under 100M (300') if it 'qualified'. well under 4000'  You might Terminate the last controller with 120 ohms (pin 4 to 5). (Gen 2 Pixies have a jumper to do this, but AFAIK, those are the only ones)

Buried cable and Lightning can be problematic due to massive ground (in earth) currents nearby. It does not need a direct strike to the affected equipment. You may have a BAD com chip that won't release the line after transmitting (RS485 is a party line type protocol)  (I was going to guess 6B, but then you said 6A did the same )

Lost the BiSync (RS232) board  in 2 different stores when we had a (rare for the SF area) storm with nearby strike. In both cases, the cable was under the slab.

Another case I was wearing a field telephone headset at a sports car race and we were getting shocks from strikes that were not even anywhere near the course

ELK Products sells Surge protectors for their Fire/Security systems (alarms use 8 pin jacks, wired RJ31X). Those provide surge protection on what would be Line 1 (standard landline) which is the same pair as used by LOR. The Green wire out the side needs to go to a Ground rod or metal pipe (same as any 'surge protector', which used the round pin to dump the spike)

[MSB]

Thanks to all for the replies. I hope this link will work. This is how this problem area is wired. All underground wiring is 23 gauge Cat 6 rated for direct burial. Our display is in a public park and we have to bury the cables to prevent vandalism and critters. All junction boxes are weatherproof.  The only difference this year is we took the Gen3 A/C controllers off the Regular network and put them onto and enhanced auxiliary network since we added a Pixie 16 upstream and needed the enhanced network to do so. This is where we found things acting very strange with the 16 channel controllers. The RGB using a Pixie 16 is behaving as expected on Aux C. I will try last year's preview using only the regular network and see if that makes a difference this afternoon.

[TheDucks]

1 hour ago, MSB said:

Thanks to all for the replies. I hope this link will work. This is how this problem area is wired. All underground wiring is 23 gauge Cat 6 rated for direct burial. Our display is in a public park and we have to bury the cables to prevent vandalism and critters. All junction boxes are weatherproof.  The only difference this year is we took the Gen3 A/C controllers off the Regular network and put them onto and enhanced auxiliary network since we added a Pixie 16 upstream and needed the enhanced network to do so. This is where we found things acting very strange with the 16 channel controllers. The RGB using a Pixie 16 is behaving as expected on Aux C. I will try last year's preview using only the regular network and see if that makes a difference this afternoon.

You need to 'Share' that image

[MSB]

I did share the link but I've been having issues with google on our gsuite domain. Try this one

[dibblejr]

4 hours ago, MSB said:

Thanks to all for the replies. I hope this link will work. This is how this problem area is wired. All underground wiring is 23 gauge Cat 6 rated for direct burial. Our display is in a public park and we have to bury the cables to prevent vandalism and critters. All junction boxes are weatherproof.  The only difference this year is we took the Gen3 A/C controllers off the Regular network and put them onto and enhanced auxiliary network since we added a Pixie 16 upstream and needed the enhanced network to do so. This is where we found things acting very strange with the 16 channel controllers. The RGB using a Pixie 16 is behaving as expected on Aux C. I will try last year's preview using only the regular network and see if that makes a difference this afternoon.

Double check those AC controllers on the same NW as the pixie16. If any of them are not G3 this can create problems. Why did you change the G3 controllers from the regular network they were working on? You could have added the pixie16 to the reg NW at 500K unless I missed something.

Sounds like you may have some serious cabling issues or unit ID sharing problems. You may need to start all over again disconnecting and reconnecting equipment and props to figure it out.

JR

[TheDucks]

Why so many repeaters and how are the getting power?

RS485 can do 4000' when terminated at each of the ends  (do NOT terminate that Pixie 16, it is in the middle of the chain.)

[jfuller8400]

I would take out the repeaters as like @TheDucks pointed out, you can run up to 4000' and you are nowhere near that.  For troubleshooting I would also try connecting a controller (or two) close to the repeater (like a foot or two) and see what that does.

[TheDucks]

The alternate is to power there repeaters from local (to repeater) supply.

The only necessary one is where you split (the black) into 2 directions

[Friday]

On 11/10/2021 at 2:05 PM, dibblejr said:

Double check those AC controllers on the same NW as the pixie16. If any of them are not G3 this can create problems. Why did you change the G3 controllers from the regular network they were working on? You could have added the pixie16 to the reg NW at 500K unless I missed something.

Sounds like you may have some serious cabling issues or unit ID sharing problems. You may need to start all over again disconnecting and reconnecting equipment and props to figure it out.

JR

Jr I was trying to send you message are you still helping individuals with their Pixcon?

[dibblejr]

2 hours ago, Friday said:

Jr I was trying to send you message are you still helping individuals with their Pixcon?

Ive only helped one member with a Pixcon16 many several years ago. I help with pixie controllers.

JR

[MSB]

Yes we know how the repeaters work and their function. They get their power from the controllers. However, the show pc  in the "Command Center" is in the center of the park. The show goes around a 1.1 mile loop and we have a star configuration suppling the data to the various "stages" . We tested 8.8 miles of buried cable and all major links tested good. I've spent the last two days with another other team member carrying two new g3 pro controllers testing links, cables, couplers, and had the same problem. Plug in one controller and it works, plug in the next controller downstream and the one before it stops working. When we just plugged in a cable to the "in link" the upstream controller quit. I was just about to give up all hope that the Food Pantry Run-Walk scheduled for Nov 20th would be able to go on with our lights and might have to be canceled. We abandoned the newly installed underground cable and ran two lines over the ground and things started working again. My team mate suggested trying a terminating resistor.

So today I made a one using a 100 ohm resistor and tested the same way we did last night and the problem went away. After 9 years of not needing them it was a shot in the dark. Our hypothesis is that the new cable is a much better conductor and the reflections were the issue all the time. Tomorrow is another day of plugging in and more testing but I'm much more confident that we will be able to have a show this year. Tomorrow is trying our a new display using DMX over E3.31 for a 48 x 100 ribbon 360 degree mega tree upgrade.

Thanks for all the suggestions. Please let me know if these links work or helpful.

Greg

 

[k6ccc]

Glad you got it figured out.  Had you by chance increased network speed this year?  That will make termination more picky.

On 11/9/2021 at 9:02 PM, MSB said:

Also we are running our of UID's due to the additional Pixie II, 8, and 16 controllers we've added.

Although it can be more confusing for us humans, you CAN reuse controller IDs on different LOR networks.  For example, Regular network Unit ID 01 is NOT the same as AuxA network Unit ID 01.

Yes, I use separate ID ranges for each of my three LOR networks, but that is not required.