|
|
Hi,
First post and hopeful someone out there can shed some light.
I've put a garden office outside, it's approx 35m from the house. The buidling has power and as of a few days ago I have run a weatherproof Cat5 cable to the office.
I've connected to either end a wall mounted ethernet socket to allow me to plug in to. So far not got past plugging directly into this socket.
At the router end(testing the ethernet cable that is plugged directly into the router to the new socket into a laptop) I am getting c30-35M speed down and c7-8M up. As I would expect.
At the other end I'm getting 9M down and 3M Up. So from the socket the other end of the 35meter cable I am getting.
As I see it it has to be one of the following:
- Sockets newly installed are slowing things down (could be one or the other i.e. either end).
- 35meter cable is causing the issue(it was a cable from a nationwide electrical retailer).
- would proximity to the powered armoured cable(power supply to the office) have an effect on speed or would it be more stability?
I've read some stuff as you do and am led to believe a cable of this length 'shouldn't' cause an issue?
Also my other question is with the sockets and the wiring of them can they be right or wrong or can they be 'part right' thus causing a speed issue.
Bit baffled, aware where in the connection the speed changes or drops(somewhere from new socket in house/cable/new socket in office), but wanted to see if anyone has any ideas before I rip out the cable that took me hours to thread along the fence line and start trying to fiddle with sockets. Hoping in the sockets case they are either right or wrong.
Is this just something I will have to live with and aside from getting a dedicated line installed which I don't want to do what is the best alternate way of doing this? Did look at point to point but don't want to throw however much at it if it's likely to be the same.
Any help, pointers, advice etc would be greatly apprecaited. I'm not overly technical but am willing to have a go(could be the issue perhaps).
Thanks in advance.
Edited by deleted (Tue 17-May-16 09:49:13)
|
|
|
35m of cable should not cause any issues.
The power cable will not cause a problem with Ethernet signals (normally) however, if you had VDSL running down there it might..
First question:
Can you open up both sockets and list which colour wires are connected to which pins - also take a picture of each and host that somewhere and then post a link here.
~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
M H C
taurus excreta cerebrum vincit
|
|
|
Good photos so that we can identify what you've done with the wiring will help.
If you don't have one, then a cable tester and simple ones are cheap £5 on Amazon can save lots of bother, e.g. it be may you have not got good contact on wire of a pair.
For quick and easy rigging up I'd recommend No tool RJ45 kit there are a variety available, useful if you are someone who never gets on with crimping.
|
|
The author of the above post is a thinkbroadband staff member. It may not constitute an official statement on behalf of thinkbroadband.
|
|
Register (or login) on our website and you will not see this ad.
|
|
|
Thanks for your quick reply. Both sockets are wired per the attached photographs/
https://www.dropbox.com/sh/qxfppkpy67ihy1y/AAAJSRQ-G...
First time i've used this so hope it works. Thanks.
|
|
|
I've just uploaded two photographs. One on the left is the house end, the other is the garden office end.
Is this where I'm made to look stupid  I followed a print out I found online.
|
|
|
|
I've done it. Thanks.
|
|
|
The first thing I would do is double check the connections including a full continuity check on all pairs.
Note that cat5 is not compatible with gigabit operation (for which you need cat5e cable). I'm not quite sure what happens if you connect cat5 cable over all four pairs to two gigabit ports. It will presumably auto-negotiate it down, but I wonder if you'll just end up with 10BaseT speeds (the default is often to run at the lowest speed, and possibly not even in full duplex mode).
If you are using cat5 (not cat5e cable) then use the 100base-TX wiring mode (that is use just two pairs - 2 (1/2 and 3/6).
http://www.zytrax.com/tech/layer_1/cables/tech_lan.h...
|
|
|
Looking at those, my first comment is that you may have the brown and orange pairs reversed and the brown pair is itself crossed. Also, the Blue pair are incorrect too, zooming in it shows Blue on 3 whereas on T568B (and A too) it should be 4.
1 Wh/Or
2 Or
3 Wh/Gr
4 Bl
5 Wh/Bl
6 Gr
7 Wh/Bn
8 Bn
You will need to pull them all out and re-terminate
~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
M H C
taurus excreta cerebrum vincit
Edited by MHC (Tue 17-May-16 11:00:46)
|
|
|
|
Sorry. To be clear I am using CAT5E cable.
Learning a lot as I go here(I now know there is a difference).
|
|
|
|
|
|
The author of the above post is a thinkbroadband staff member. It may not constitute an official statement on behalf of thinkbroadband.
|
|
|
|
Is there an issue if cat5 (not cat5e) cable is used to connect two gigabit ports over 35m of cable? I assume it will auto-negotiate down, but to what speed (auto-negotiation can be tricky - you might end up with 10BaseT speeds).
|
|
|
And not a bad attempt at wiring ... a couple of pointers.
Cut off the stripping thread once you have opened the cable.
Keep the pairs twisted together as much as possible and avoid the little loops.
A couple of the IDC connections look as though the wire may not be fully home. I will insert the wire, press down to get the click and the surplus trimmed and then, without removing the tool, one or two more depressions - just to make sure.
Good to hear you used External cable - it will last.
~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
M H C
taurus excreta cerebrum vincit
Edited by MHC (Tue 17-May-16 10:55:48)
|
|
|
Not an issue as the OP has confirmed Cat5e.
~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
M H C
taurus excreta cerebrum vincit
|
|
|
|
Thanks, I see the error. I have copied a diagram in the sense of colours connecting where but seemingly ignored the numbers behind.
I'll give it a rewire and let you all know.
Much appreciated.
|
|
|
It's rather tricky as we can't see all the pin numbering. I have found a similar layout on another socket which matches the numbers I can see. If the pin numbers are arranged as
2 8
1 7
6 4
3 5
then I can see a problem. 3 & 6 should be a pair as should 4 & 5. If the numbers are arranged as above, then they aren't paired on your connector.
In any event, I'd strongly suggest you wire the connectors using the 568B systems as it makes diagnosis easier. If that doesn't work (because of an incompatibility between cat5 cables and 1000BaseT ports) then try connecting just 1/2 & 3/6 for the 100Base-TX format.
http://www.zytrax.com/tech/layer_1/cables/tech_lan.h...
|
|
|
You have hit a well known issue - different connector manufacturers put their IDC connections in different places for each pin. IT is not unknown for an experienced technician to wire up a large number of RJ45s to a specific scheme, following the correct pin out - then suddenly one set of outlets is a different brand and he carries on as before, totally forgetting to check pin numbers!
Hopefully you have read my hints too ...
And take a picture of the connector without wires and then the reworked connectors - will not take us long to check.
~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
M H C
taurus excreta cerebrum vincit
|
|
|
|
OK - I just looked at the original post which said cat5. Having looked at this some more (and not being able to see the full numbering scheme) I've found a similar type of connector and if the pin sequence echoes that, he has got g & b/w on one pair and b & g/w on another. If so, that's not good.
|
|
|
See my other post, The OP seems to have followed a diagram rather than the actual pin numbers. And yes, it is a case of swapped pairs, inverted pair, and split pair! At least the two ends are the same so the signals were appearing in the right place albeit with significant crosstalk.
~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
M H C
taurus excreta cerebrum vincit
|
|
|
An Ethernet cable tester from www.millsltd.co.uk is inexpensive and will highlight any issues and pinpoints the actual wires requiring attention. It is all too easy to wire up one end as 568A and the other as 568B. I think this thread said you used exterior grade cat5e but was it also armoured that is earthed? My garden SOHO office is about a 50M run laid in the same trench as the power supply but separated apart as far as possible. There is hardly any speed drop at the remote end.
|
|
|
An Ethernet cable tester from www.millsltd.co.uk is inexpensive and will highlight any issues and pinpoints the actual wires requiring attention. It is all too easy to wire up one end as 568A and the other as 568B. I think this thread said you used exterior grade cat5e but was it also armoured that is earthed? My garden SOHO office is about a 50M run laid in the same trench as the power supply but separated apart as far as possible. There is hardly any speed drop at the remote end.
URL given is out of date. Go to http://www.millsltd.com/mills-cable-tester.html instead!
|
|
|
|
And at 35m you would probably be fine with plain Cat5. Remember the system is specified for a full 100m run so as you back off the distance things get a whole bunch more forgiving.
That said I don't think you have been able to buy plain Cat5 cable for a very long time now. It's been all Cat5e for over a decade.
|
|
|
DELETED
Beaten by Mr Saffrons advice of £5.00 testers being available!
Edited by Rastus (Tue 17-May-16 16:12:24)
|
|
|
DELETED
Beaten by Mr Saffrons advice of £5.00 testers being available!
There are differences. The £5 tester just checks continuity 1-1, 2-2 &c whereas the more expensive will look for reversed and crossed pairs too.
And using the £5 tester, everything may have looked fine, in this case.
~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
M H C
taurus excreta cerebrum vincit
Edited by MHC (Tue 17-May-16 16:19:40)
|
|
|
There are differences. The £5 tester just checks continuity 1-1, 2-2 &c whereas the more expensive will look for reversed and crossed pairs too.
And using the £5 tester, everything may have looked fine, in this case.
You're probably right, I was assuming all cheap testers were capable of performing all those tests as standard.
I can't remember how much I paid for my Velleman VTLAN3 about 2-3 years ago from eBay (UK) but I'm fairly certain it was under £10.
Just had a quick look and I can't find it on the UK site but it's still available on eBay.com only at a higher price than I paid, but it proves the point that it's not necessary to fork out over £70 for something which will probably only be used occasionally.
|
|
|
Any progress?
~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
M H C
taurus excreta cerebrum vincit
|
|
|
Hi,
Yes complete success thank you. I followed the link within the post and followed the 'B' plan and low and behold. Full speed.
I did post this I 'think' in response to the poster who put the link up but not sure how the forum works.
Very much appreciate your quick response. Had it dragged on I would have likely ended up hiring someone to 'fix' my erroor.
I write this sat in my lovely new office
|