Netonix Forums

How accurate is the cable length test? I know it says:

Lengths are accurate to +/- 3.28 feet / 1 meter

Code: Select all

Lengths are accurate to +/- 3.28 feet / 1 meterRun

However, I have a 3 foot cable plugged into a port, when I run the cable test it says:

Code: Select all


Pair 1 Length
9.84
 
ft /
3
 
m


Pair 2 Length
9.84
 
ft /
3
 
m


Pair 3 Length
9.84
 
ft /
3
 
m


Pair 4 Length
9.84
 
ft /
3
 
m





So the cable test routine is built into the core, we do not write it nor have the ability to modify it.

We simply request it to run and the core returns values we display.

Normally the cable test length is pretty accurate but external things can affect it such as:

A damaged Ethernet Transformer. I have seen Transformers that were carrying AC ground current all the time and get hot but do not fry and it changes the attenuation on the magnetic coupling. Also would introduce errors and shorten the cable length that achieves 1G connections.

A damaged ESD protection DIODE - CAN HAPPEN BUT NOT COMMON

Use of Ethernet Surge Protectors - varied results depending on make

Bad or damaged (kinked) cable - attenuation (but normally shows differnt lengths per pair)

RF Noise on cable

If you suspect a port is damaged test the same cable on multiple ports and compare answers.

Also try a longer cable, short cables are inaccurate.

Try the test multiple times, the way it works a bad answer can occur.

mhoppes wrote:I have a 3 foot cable...

I cannot help but wonder why you need to run the cable length test if you know the length but my guess is you want to show it is inaccurate. I would not expect a very short length to be measured accurately. Have you compared results with longer known length cables?

We have a Ws-6-mini that’s in orbit. Pretty much inaccessible. A device connected went offline with no Ethernet connection.

I was trying to determine if the cable was damaged or what may have happened.

Someone will get to it eventually - but it requires a very specialized trip.