How many use shielded ethernet cable

[Coop]

I have been having issues with phantom lights across all my controllers and channels. I have opened a ticket and LOR thinks its noise/interference. I haven't been able to isolate the cause yet (Mother Nature hasn't been cooperating here in Minnesota either) I have seperated as many AC cords from ethernet as I can with no change. My next thought is shielded ethernet cable. Does anyone out there use shielded network cables?

[rbmurray210]

i use shielded 18/4 security cable, just couldn't get myself to use Ethernet cable.

[k6ccc]

I do not use shielded cable for either the LOR network (what I assume Coop was asking about) or for my light cables.  The only thing I have shielded is the audio cables between the show computer to the mixer and from the mixer to the amplifiers.  The speaker cables from the amps to the speakers are not shielded.

[Paul R]

Wahta re you currently using. Standard Cat5 cable is not very well shielded, while Cat5e has some internal shielding. If you will notice, LoR only sells 5e...that tells me something.

 

-Paul

[Santas Helper]

I don't think I've ever used shielded cables and they run along with power cables.

Maybe it's mother nature  since you said she isn't cooperating.

[Steven]

I have some areas where I run network and power through the same conduit. For those runs, I use Cat5-shielded. It's expensive, and not flexible, but it seems to work.

 

I also recommend a RS485 terminator on both ends of your network. This has solved network problems for me.

[-klb-]

Wahta re you currently using. Standard Cat5 cable is not very well shielded, while Cat5e has some internal shielding. If you will notice, LoR only sells 5e...that tells me something.

 

-Paul

Um, no. Unless you specifically have shielded cable, cat 5, 5e, and 6 do not contain any shielding.

The difference is the twist rates, required differences on twist rates between pairs, precision of twist rates, and allowable untwisting at terminations.

[Coop]

I installed shielded Cat 5e with no change. Today I disconnected the ethernet cable from USB485B and the phantom lights stopped. I shutdown the laptop and phantom stopped. I swaped out the 485b and filtered USB cable (no change) and I swapped the laptop out (still no change). As soon as USB port gets power the phantom lights begin. I created a blank animation sequence thinking that an "off command" might supress the phantom. I'll have to wait until after the show tonight to test

 

If anyone has any other suggestions I'm all ears eyes. 

To clarify the shielded cat 5e connects all controllers and the USb485b to the network. Could changing the network speed help?

[dgrant]

I use Cat6 wire for all the communications and at the normal 59k speed. Seems to work for me but I'm only running 7 standard controllers and one CCR controller.

[Steven]

Try an RS485 terminator. It worked for me. Here are some instructions on how to make one.

[Ken Benedict]

How about these that are already made up?: http://www.ebay.com/itm/2-NEW-IBM-59P2950-59P2951-Interconnect-Terminators-RJ45-/270758786306?pt=US_Server_Disk_Controllers_RAID_Cards&hash=item3f0a7b3502

[stachows]

I used to have a big problem with phantom lights. One year it was a bad network cable in my setup. It only took one bad cable out of 20.

Another year it was a bad DMX (mic) cable connected to my DMX lights through a IDMX 1000.

Here is how I found it.

1. Start by making sure you can duplicate the problem. If not it is much harder to identify.

2. Keep running the sequence, or section that is the worst.

3. Start at controller 1 and disconnect all controllers beyond it. If problems do not occur, then the cable from the USB dongle to controller 1 is probably good.

4. Move on the the next controller. Unplug everything downstream and look for issues.

5. Repeat moving downstream one controller at a time. Once the issues occur again, replace the cable between the last two controllers you tested, and try again...

Couple notes: One year had to plug the whole network in reverse to find out it was a bad cable between controller 2&3. This was because the phantom lights only occurred on lights plugged into controllers 5,6 and 7. (Which obviously we're not connectd when tested as listed above.).

The DMX cable issue was easy. The problems stopped when I unplugged the IDMX.

Also, check your CCR's and on their devices that are capable of standalone sequences. Make sure they are not storing any. This can cause phantom lights to com on.

I have since switched to shielded gigaspeed cable and good quality DMX cables, and now have no issues at all. With 20 controllers, I have never used a terminator for what it is worth.

Let me know if you have questions,

Scott

[kzaas]

Last year I had a huge problem with phantom lights so this year I switched to shielded CAT5e and no problems until ran a subwoofer to my porch to enhance my sound. Then POW major phantoms, turns out I ran the subwoofer cable to close to the USB to RS485 cable. I moved the sub cable one foot away and no more phantoms.

 

It will not do any good to run shielded cable unless you connect one end of the shield to ground( do not connect both ends to ground, that can cause a whole other set of problems). The RJ45 connectors on the LOR boards do not have a ground for the shield and neither does the USB adapter, I ran the shield wire outside of the RJ45 connector and connected it to a ground on the controller box. each cable is grounded at one end and it is working perfectly this year.

[ks4an]

What is a RS485 & a RJ45 terminator used for?

[dgrant]

RS485 communication requires a termination to function normally. I would imagine that its handled in the controllers themselves as LOR would have already taken this into consideration in their designs. I'm not an expert but have used RS485 in a laboratory environment and if the comm line didn't have the terminator on it, spurious things would happen or it not work at all. RS485 is a bi-directional serial data comm path/type. Just imagine RS232 which is a single device, RS485 allows for 64 devices on one cable. The TX+/- wires and the RX +/- wires are suppose to be twisted pairs and invidually shielded, plus shielding around the whole cable. LOR uses Cat5 type wire due to the costs are considerably less expensive, for us!!! The distances involved with cat5 cables are less than RS485 cables because of this and other factors come into play as well without buying expensive extenders, adapters and so on. My prediction here is that LOR will eventually have to switch to TCPIP communications with our controllers simply due to the number of controllers being placed in a system and its a whole lot faster. Now I've been reamed in here before for saying that, people complaining that their Christmas show will get hacked, viruses and etc...all highly doubtful if the normal precautions are used or disconnected from the outside world. Anyway, we're not at that point yet but I think it'll have to come.

[Steven]

RS485 communication requires a termination to function normally. I would imagine that its handled in the controllers themselves as LOR would have already taken this into consideration in their designs.

 

Yes, in theory (and in the standards) RS485 requires a terminator, but in practice (e.g. LOR) it works without. There is actually no termination in the controllers themselves, but the relatively low speeds, relatively short distances, and losses in the cables and connectors let it work without termination.

 

But, a terminator can't hurt, and can solve some problems. It's also easy.

 

My prediction here is that LOR will eventually have to switch to TCPIP communications with our controllers simply due to the number of controllers being placed in a system and its a whole lot faster.

 

The LOR protocol is very efficient (compared to DMX, for example) so they are able to get away with lower speeds. New controllers can work at 500kbps. The only way to go faster over Cat5 would be to use 100baseT (or 10000baseT), but the big disadvantage is that 100baseT is point-to-point, so in order to daily chain, you would need the equivalent of a 3-port Ethernet hub at each controller. This would be much more expensive, and it would require that all controllers in the chain be powered, which means 1 controller failure could take down the entire network.

[aussiephil]

My prediction here is that LOR will eventually have to switch to TCPIP communications with our controllers simply due to the number of controllers being placed in a system and its a whole lot faster. Now I've been reamed in here before for saying that, people complaining that their Christmas show will get hacked, viruses and etc...all highly doubtful if the normal precautions are used or disconnected from the outside world. Anyway, we're not at that point yet but I think it'll have to come.

 

Oh it's already well and truly here for a lot of people, I have 5 DMX based controllers and ELEVEN Ethernet connected controllers including the Ethernet to DMX bridges. It's all on a seperate secured VLAN and whilst a determined hacker may get in the likelyhood is so low that it's just not worth thinking about.

Those Ethenet based units are driving over 18,000 channels this year, and that is seeming to be quite unremarkable these days

[-klb-]

LOR has even talked about a E1.31 to LOR network bridge. Probably a bit of a stretch to the standard, but could offload some of the USB adapters, and shorten some of the runs.

[jeffl]

It's interesting reading this. I can remember searching for Steven's post on these topics all the time when I ran all LOR protocol with sticking channels and such.  Ever since I switched over to DMX I have not had any issues related to comm cables.

[k6ccc]

LOR has even talked about a E1.31 to LOR network bridge. Probably a bit of a stretch to the standard, but could offload some of the USB adapters, and shorten some of the runs.

 

Now that I would love to see.  I'm already thinking about running all my LOR controllers on DMX downstream of SanDevices E1.31 controllers.