Ethernet cable resistance

[Stonent]

Anyone know off the top of their head what the maximum resistance should be on about 100M of a Cat5e Ethernet cable pair?

Or I guess even better would be if I put a 9V battery across one pair, what kind of voltage drop should I expect on the other side? (I guess technically in that case it would be 200M since it would be in a loop.) Testing resistance on installed cable might be nearly impossible directly but sending a small voltage down a pair and testing it would probably be easier.

[AndyC_772]

It's normally of the order of 1 Ohm for every 10 metres, so your 100m cable will be about 10 Ohms end-to-end (20 Ohms round trip). The voltage dropped across the cable will depend on the current you draw, of course.

[Dave]

Testing resistance on installed cable might be nearly impossible directly but sending a small voltage down a pair and testing it would probably be easier.

You could short a pair at one end and hook up a power supply and two multimeters to the other. You send a constant current through that pair, and measure both the voltage drop and the exact current. Apply Ohm's law and you have the resistance. Don't forget to divide it by 2 to get the resistance of a single wire.

[Bertho]

Anyone know off the top of their head what the maximum resistance should be on about 100M of a Cat5e Ethernet cable pair?

Or I guess even better would be if I put a 9V battery across one pair, what kind of voltage drop should I expect on the other side? (I guess technically in that case it would be 200M since it would be in a loop.) Testing resistance on installed cable might be nearly impossible directly but sending a small voltage down a pair and testing it would probably be easier.

Cat 5e has a loop-resistance of less than 0.188 Ohm/m. So, pulling 1A through one pair for a cable of 1m drops ~0.188V (2m -> ~0.376V, etc). If it is only DC, then this will hold. If you put AC or AC superpositioned on DC, then the AC part will see the impedance of the cable (depending frequency). For frequencies from ~100kHz you need to take account for the characteristic impedance (and that means proper termination). If you put a digital signal on it, you will see the impedance almost immediately (square-wave contains higher harmonic frequencies).

See also: https://en.wikipedia.org/wiki/CAT_5#Characteristics

[digsys]

I had that exact situation last week. 12R drop on 120M of cable, one way. Matched exactly the specs of the reel.
NOTE: If you have a S/Mode at the other end though ie POE, it may not "start up", if it isn't a soft start type.
Had that too. A 1,000uF cap fixed it, messy but had to happen.

[Stonent]

Our network guy at work has had some of his duties transferred to me and someone else. So we'll be responsible for troubleshooting cable issues and putting RJ45 ends on pulled cable. I've got a Fluke Etherscope II and a Fluke tone generator and I was troubleshooting a newly ran cable issue where the Etherscope was saying the jack was 100% dead with no wires connected.  It said when I went on one end of the cable that it was 189 feet long and with nothing on the other side.  If I did it going in the other direction it said the cable was 4 feet long. And I've found the tone generator seems to spill signal over to other cables which was annoying when tracing this wire.  I finally thought, damn, it would be so much easier to just put a 9V battery on the end of each pair and test the other end for voltage.

[Bertho]

You cannot test the performance of a cat 5e cable with DC. You need to test the characteristics of the cable. DC is for the most part unrelated to the actual state of the cable, unless you have a short or break.

[Stonent]

You cannot test the performance of a cat 5e cable with DC. You need to test the characteristics of the cable. DC is for the most part unrelated to the actual state of the cable, unless you have a short or break.

I'm mainly concerned with continuity right now. I'm sure the cable quality is ok since it's the same cable we always use, but I really want to check for breaks in the line.

Even if the cable was wired completely wrong, the Etherscope should at least see something. I think if there were a continuity break the tone generator signal might be picked up and carried down some ways.

Now that I think about it, if one end of the cable says 4 feet long and the other 189 feet long, then maybe there is a break 4 feet away from the wall jack. I guess I'll find out on Monday.

[MacAttak]

Now that I think about it, if one end of the cable says 4 feet long and the other 189 feet long, then maybe there is a break 4 feet away from the wall jack. I guess I'll find out on Monday.

Or the jack itself is damaged or incorrectly installed. 4 feet is close enough to zero feet that it could be within the margin of error.

I'll presume that proper punch-down tools are being used to install the jacks, along with properly tensioned impact bits... so good chance it's just a bad jack on that end.

[Stonent]

Now that I think about it, if one end of the cable says 4 feet long and the other 189 feet long, then maybe there is a break 4 feet away from the wall jack. I guess I'll find out on Monday.

Or the jack itself is damaged or incorrectly installed. 4 feet is close enough to zero feet that it could be within the margin of error.

I'll presume that proper punch-down tools are being used to install the jacks, along with properly tensioned impact bits... so good chance it's just a bad jack on that end.

Yeah we didn't get nearly the amount of time to troubleshoot this as we wanted.  As we were headed back to get some tools someone stopped us and said that all outbound long distance calls were failing so we had to contact our corporate telecom group so they could figure out what was going on. I spent the rest of my evening working with them and running tests.

[Psi]

Really cheap cat5 is made of another metal that's been copper plated, where as good cat5 is solid copper.

So you may find variations depending on the cat5 you have.

[Stonent]

Really cheap cat5 is made of another metal that's been copper plated, where as good cat5 is solid copper.

So you may find variations depending on the cat5 you have.

It's solid copper all the way through, I've looked at the ends of the wire.

[G7PSK]

Could be the cable core is broken 4 feet from one end and your meter is not telling you porkies.