Answered Network cable length?

scamper

Dedicated elf
Joined
Jan 5, 2014
Messages
1,225
Location
collie
I have just wired up my controllers outside for my display and discovered I now have a bit of a problem.
I have 2 falcon f16 controllers and wanted to use them to control different sections in my front yard.
I set them up and tested them in the shed no problem.
I have a cheap switch (which I am thinking could be the problem) I run a network cable from my computer into that, then out of that into the 2 f16's (1 cable to each)
The problem is, I cannot communicate with the furthest one through the switch.
If I run he cable direct, it works ok.
Could it be that the switch is too cheap and cannot transmit the data that far?
any advice is welcome.
 

David_AVD

Grandpa Elf
Community project designer
Generous elf
Joined
Jun 12, 2010
Messages
4,681
Location
Victoria Point (Brisbane)
Ethernet is good to 100 metres. What length are the cables?

Make sure you have the wire pairs in the correct positions on the RJ45 plugs.

A normal network cable tester will not show up split pair mistakes.
 

i13

Dedicated elf
Joined
Jul 5, 2013
Messages
1,172
I'm wondering if the switch has a dud port or your longer cable is bad. It should work fine for at least 50 metres between the switch and controller. These wouldn't be hard to check - try using the longer cable to the other F16v2 controller. Are you using the same long cable when you plug directly into the computer? Try plugging the working F16v2 into the non-working port on the switch and then turning the switch off and on.

This might be worth knowing if you still can't get things talking
https://auschristmaslighting.com/threads/network-switch-vs-usb-hub-what-does-what.8041/
 

scamper

Dedicated elf
Joined
Jan 5, 2014
Messages
1,225
Location
collie
I used the same cable that I used for the long run. I did it as a test as that was my first thought.
The tester showed it working and there was data being transmitted as the lights were flashing so that is when I plugged it directly into the computer and it worked fine, I setup what I needed and run a test sequence without any problem.
I then plug it back into the switch and it won't find the f16.
I swapped boards, swapped outputs on the switch, still no go.
The only thing I could think was the switch was clipping the signal or something.
but not being a full bottle on networks, I really am only guessing.
The cable is only about 15 meters
 

B_P_J

Full time elf
Joined
Jan 6, 2014
Messages
355
Location
Wadalba
What kind of switch is it Scamper? is it pre-loved? could be a setting on the switch (depending on type) if you can see it when directly connected to the laptop, do you have another switch anywhere you could try? or just another thought, are the controller IP addresses different? (just making sure)
 

scamper

Dedicated elf
Joined
Jan 5, 2014
Messages
1,225
Location
collie
bpj1980 said:
What kind of switch is it Scamper? is it pre-loved? could be a setting on the switch (depending on type) if you can see it when directly connected to the laptop, do you have another switch anywhere you could try? or just another thought, are the controller IP addresses different? (just making sure)

It is a cheap one I bought from china. (new) I didn't think it would make a difference. I got it as it was small enough to fit into the box with the rest of the gear and had a usb power lead, so no messy plug pack.
The ip addresses etc are all good, like I said, I had I all working in the shed albeit with short cables.
The other thing I thought, When I made the cable I just ran the wires in parallel, meaning I didn't worry too much about colour order except they were the same each end, I cannot believe that would make a difference, but is it possible?
 

David_AVD

Grandpa Elf
Community project designer
Generous elf
Joined
Jun 12, 2010
Messages
4,681
Location
Victoria Point (Brisbane)
scamper said:
The other thing I thought, When I made the cable I just ran the wires in parallel, meaning I didn't worry too much about colour order except they were the same each end, I cannot believe that would make a difference, but is it possible?

That's EXACTLY the issue I was talking about. You NEED to wired the ends correctly.
 

AAH

I love blinky lights :)
Community project designer
Joined
Dec 27, 2010
Messages
4,188
Location
Eaglehawk
Unless the wires are in the right holes they won't be using the twisted pairs of the cat 5 cable and you will pretty much always get failure in the data. There's 2 different ways of pairing up the wires and both work.
http://www.cableorganizer.com/articles/difference-between-T568A-and-T568B.html
You end up with matching wire colours on both ends of the cable but there is no apparent logical order to the wires if you just look at them.
image010.jpg image013.jpg
 

multicast

Senior elf
Joined
Jul 13, 2013
Messages
715
scamper said:
When I made the cable I just ran the wires in parallel, meaning I didn't worry too much about colour order except they were the same each end, I cannot believe that would make a difference, but is it possible?

Absolutely.. And as your cable gets longer this becomes really really important. It might work over a few meters but certainly not at distance.


Ethernet signals are "differential". That menas two wires ( a pair ) are used to transmit the same signal. On one wire, the signal is sent with a + sense and on the other its sent with a - sense. At the receiver end, any noise that is induced onto the cable can be cancelled out by "adding" the + and - signals together. ( wiki differential signalling for a more detailed descrption ). It works because the two pairs are tightly coyupled ( thats why they are twisted in pairs.. and the main differernce between Cat5 and Cat6 is how regular that twist i..

THe TX+ and TX- signals ( and RX+ and RX- ) MUST always be on a pair.
 

scamper

Dedicated elf
Joined
Jan 5, 2014
Messages
1,225
Location
collie
Thanks for that, I have now read the wiki article and it makes more sense.
I couldn't work out why there would be a difference in the colours as they are twisted in 4 groups.
But it appears the twisting is at different rates so as to not induce any emf or harmonics from one to the other.

I have now corrected my mistake.
And it all works.
Thanks for being so helpful and nice about it. *** I love this forum***
 
Top