This year is just HoHoHoting me off!

[scamper]

I do actually have a question... Which I am sure I will end up figuring out, but I have had one problem after another this year and the latest one is doing my head in.

I have 3 falcon controllers and a Pi and all are on the same network and wired in.
I have a switch that they all go through and all the lights on the switch are flashing happily. Also, after looking for all sorts of connection problems, I find that in fact all of the controllers are working as they should as I can run the Pi in test mode and all the lights do as they should.
But, the falcon controller in question does not come up in my browser.
I can log into 2 of the falcons and the Pi and do everything I need, I can look at the IP address on the controller but it just comes up as took too long to respond, I have shut it down, my internet browser done but it still does not come up.

Is there something that someone can recommend that I may have overlooked?
Oh, and this only started after I got a problem with a supply cable that runs to the box it is in, as last week I was able to log in and do everything I needed to reassign ports etc.

 

[bassowl]

Could it be the network cable? Have you tried swapping the falcon for one of the other devices?

 

[Johnnyboy]

I had a dodgy network cable that was making my display stutter.
The cable worked, and could do controller software updates but just not run a sequence without micro stutters. So it worked, just not to the right level. I replaced it and now works perfectly
Try new cables if you can

 

[scamper]

I had a dodgy network cable that was making my display stutter.
The cable worked, and could do controller software updates but just not run a sequence without micro stutters. So it worked, just not to the right level. I replaced it and now works perfectly
Try new cables if you can

Yeah thanks. I did think of this, and when it got dark I did notice a lot of flickering in a 2 different props connected to this controller, so may look at this today. Unfortunately, this is one that is buried in a conduit that runs out to the front yard, so am hoping a bit of crc on the connectors may solve it.
I may double check the power also, but less likely as the 2 props that were playing up are different supplies and even different voltage

 

[Indigogyre]

Double check the network subnet mask on the flaky controller. most people look at the ip and vague at the subnet mask and it can be ever so slightly off and not be noticeable.

 

[scamper]

Double check the network subnet mask on the flaky controller. most people look at the ip and vague at the subnet mask and it can be ever so slightly off and not be noticeable.

I will have another look, but it has the same network it has been running for 4 or 5 years. And it was actually running about a week ago, but then I had a breaker trip and thought I had a supply problem, but it turned out to be a mains extension lead that had moisture in it.
I am also going to hook up the next box which I run downstream from this just in case the missing box is causing it not to have an end termination to the network run.

 

[AAH]

Yeah thanks. I did think of this, and when it got dark I did notice a lot of flickering in a 2 different props connected to this controller, so may look at this today. Unfortunately, this is one that is buried in a conduit that runs out to the front yard, so am hoping a bit of crc on the connectors may solve it.
I may double check the power also, but less likely as the 2 props that were playing up are different supplies and even different voltage

IF the cable has water around it due to the conduit now being filled with water that could do awful things to the signal integrity.

 

[scamper]

IF the cable has water around it due to the conduit now being filled with water that could do awful things to the signal integrity.

No, the conduit is sealed all year round and then has candy cane tops on when I open it for the year. The power cable is run separately and is where the moisture problem was.

On a side note, I think I should change my avatar

 

[Kimbo3000]

Hey..
I feel your pain. This year we have added probably the least number of new lights to our display and had the most problems in total (some I will admit were our doing).
Last night was our first night this year for blinky and 1 hour before turn on I had the same problem.. I couldnt access the controller either.. We also had troubles with this same controller earlier but started working after wiggling some cables and worked since.. so i thought it was sorted.
Finally worked out that it was the waterproof RJ45 connector on the controller box was dodgy. Removed it and just ran the cable through the 20mm hole in teh bottom of the cabinet and put some tape over it.. problem solved!

So maybe check any connectors or joins you have along the way in the cabling..

Kimbo

 

[BAZMick]

Oh, and this only started after I got a problem with a supply cable that runs to the box it is in

You didn't happen to move the mains cable next to and parallel with the network cable? The mains cable can induce voltage into the network cable and mess up the data signals

 

[scamper]

You didn't happen to move the mains cable next to and parallel with the network cable? The mains cable can induce voltage into the network cable and mess up the data signals

Don't think so, I am using cat5 cable and haven't had any problems in the past, and in the box over the other side of the yard that is working fine I do actually have the mains cable and the network cable run through the same conduit. Worth a look though.

 

[BAZMick]

mains cable and the network cable run through the same conduit.

That'll do it. Back when we used to install office partition screens the specs say to keep data cable min 150mm away from mains. Or run in seperate conduit.
And from personal experience I've had to re-route a data cable inside an industrial machine away from the mains supply to solve a gremlin.
Try running another extension lead or data cable (which ever is easier for you to find) over the ground and disconnect the in conduit one

 

[battle79]

Standard networking signals on ethernet cable should not get interference from power. They specifically don't use frequencies that are likely to be affected by AC power. The segregation is for safety to prevent AC power transferring to ethernet cables if something ever gets cut.

The differential signals we send out to receivers on the other hand. I have no idea what frequency they use and if AC power would affect it. What I can say is I'd treat it as if it can be affected, and separate as it much as possible. I believe I've seen this issue myself, but honestly could of been many other things.

 

[Skymaster]

Differential pixel data is transmitted at the same frequency as the regular pixel data just with the line balanced, rather than single ended. This is at 800kHz.

Noise is introduced onto both legs of the balanced signal equally, so they becomes A+N and B+N.
The balancing means any induced noise is removed at the far end when it feeds into the differential line receiver chip. Here it is doing a subtraction of the signal which equals (A+N) - (B+N) = A-B
With B being the inverse of A, this then becomes A -(-A) = 2A ..... So all the noise is removed.

This is also how ethernet operates, with a differential pair for Rx and Tx (10/100), or 4 pairs for either direction (>=1G)

Now, the biggest issue you get is that A and B above are still ground referenced. If both lines shift too high on respect to ground, the receiver can't decode the signal. This could be caused by DC on the lines, or induced voltage at a crazy high level. Running parallel to power for a long run could potentially cause this, but it'd be unlikely as you need to get something like 7 volts induced in respect too ground on both legs.

 

[scamper]

That'll do it. Back when we used to install office partition screens the specs say to keep data cable min 150mm away from mains. Or run in seperate conduit.
And from personal experience I've had to re-route a data cable inside an industrial machine away from the mains supply to solve a gremlin.
Try running another extension lead or data cable (which ever is easier for you to find) over the ground and disconnect the in conduit one

If you read it again, you will notice that I said, the run with the network and mains in the same conduit is on the other side of the yard and is in fact working fine. The problem child is the one where the power is above ground and the network cable is a permanent below ground fixture.