Ive been networking around where Im currently working, it is not your normal environment to say the least...lets say there is a LOT of EM waves going around here. Depending on where i run the cable, and how long, my connections are intermittent, and I generally have to make shorter cables to get things done, needless to say, I dont have the hardware to go hub-hub here, unfortunately, and 3coms claims to work are obsurd...not sure if its caused by the environment here, but its happening, Im a long time networker, have been networking for about 6-8 years now, give or take, I have 2 reels of cable here, all unshielded, but I am not sure if shielded would help, Im assuming that EM interference is causing the bulk of my problems, but I have no way to test or validate this theory, because some different cabling that was running before seems to run longer distances, not much but a little, any ideas? thoughts? Im goona query some other network guys on this base tonight, but...when I cant run a 50-120 ft cable without having problems, something is up. Never seen anything like it, any thoughts and/or ideas would be appreciated.
Tek-Tips is the largest IT community on the Internet today!
Members share and learn making Tek-Tips Forums the best source of peer-reviewed technical information on the Internet!
-
Congratulations TouchToneTommy on being selected by the Tek-Tips community for having the most helpful posts in the forums last week. Way to Go!
Cat 5 cable signal loss?! cable less than 50 ft? 2
- Thread starter Rayek
- Start date
-
2
- #3
daronwilson
Vendor
I haven't had to use any shielded cable (actually it is screened) in any EM location except broadcast facilities and that was only once out of several facilities I worked in. The problems you are describing sound more like 100 Mbs connectivity over a wire that has split pairs. Are you sure you are getting a pair for pins 1&2 and a pair for pins 3&6?. Failure to do that will really shorten the length of cable that will work at 100 Mbs full duplex. Using screened (foil shielded with a drain wire) could help if there is considerable RF interference, provided that the ground drain does in fact go to ground. Use of the proper plugs, jacks and patchrack are also necessary to complete the grounding system.
Good Luck It is only my opinion, based on my experience and education...I am always willing to learn, educate me!
Daron J. Wilson, RCDD
daron.wilson@lhmorris.com
Good Luck It is only my opinion, based on my experience and education...I am always willing to learn, educate me!
Daron J. Wilson, RCDD
daron.wilson@lhmorris.com
I had to run some cabling in a small manufacturing environment and one machine was causing problems. We decide to run the cable in conduit and that fixed the problem. All situations are different and you may even want to consider running shielded cable in the conduit ....good luck.
daronwilson
Vendor
Assuming you use a screened cable, that is a 100% foil screen around the cable with a drain wire attached, what possible improvement would you expect from a second shield of conduit? Also, unless you are carefull to use compression type conduit fittings (as opposed to set screw type), the conduit system will not make an RF tight shield. I will bet that whatever device you are plugging into on each end will not have as good a shield around the components as the 100% foil. We just took care of a customer whose gas pump system was having problems. The engineers at the factory had us replace the screened wire with more screened wire. Then, they had us reinstall it inside metalic conduit with compression fittings. No change in signal problems. We replaced it last week with fiber.
It is only my opinion, based on my experience and education...I am always willing to learn, educate me!
Daron J. Wilson, RCDD
daron.wilson@lhmorris.com
It is only my opinion, based on my experience and education...I am always willing to learn, educate me!
Daron J. Wilson, RCDD
daron.wilson@lhmorris.com
- Thread starter
- #9
Just to let you all know, Im consulting the network guys on base tonight, they are supposed to show up, what is odd in this situation is that one cable can be 75ft, run fine, other can be 35, and be dead, even though it is a 100mbps network, and I am completely unfamiliar with your 1&2, 3&6 pair, unless you are talking of a crossover...Im running straight through to a 3com switch....we have some red wire that runs slightly better(simply talking cat5 cabling color), but I think the foil based shielded would probably be the best thought, though the fact that we have such a small network here, I dont really see the justification of buying expensive stuff for it at this point. After consulting their "networking experts" I will get back to you all, and needless to say, I am in a very heavy EM environment...much heavier than a typical radio station. If you have anything on the 100 mbps full duplex issue, let me know, the switch I currently have in here does not support auto sensing, so i had to run straight through, even though I am behind as far as times go, I can still learn and am willing if you have anything to add, thanks for all the help.
daronwilson
Vendor
Hmm..let me address a couple things:
>Why do they make Double paned windows?
If you are trying to tell me that having two panes of glass is somehow analogous to double shielded wire, your point is far too cunning for me to understand. FYI, double pane glass is done to provide an insulation factor, it provides a dead air space which is generally filled with an inert gas I think. I'm sure no window expert, but shielding is shielding. If you have a 100% shield then you have a 100% shield. There are double shielded and quad shielded cables on the market. Much of the cabling in high RF environments (duplexers, combiners, filters) where the alien RF source is very strong are double shielded, but generally this is two braided shields. Braided shields by nature are very hard to construct to 100%, so by doing a double shield you make it pretty close to 100%. The point is, if 100% shielding is not keeping out the stray EM noise or RF, then it is likely coupling to the shield and causing problems that way, and you most likely have a grounding issue. Yes shielding can be tricky, and yes every circumstance is different. I find it best to test the wire and see where it fails. NEXT? ATTENUATION? NOISE? It is much easier to determine a solution when you know which parameters are the problem.
>I am completely unfamiliar with your 1&2, 3&6 pair...
Well here goes....for ethernet, pins 1 and 2 MUST be on a pair of wires that are twisted together, this keeps the good signal in and the bad signals out. Pins 3 and 6 must also be the same way. If you are making cables and putting modular plugs on the end, and not putting the colors in such a manner that pins 3 and 6 are a pair, you have a major problem. From your description, it sounds to me like a split pair. That is if you started on pin one and just went white/blue pair, white/orange pair, white/green pair, white/brown pair. Hopefully this description is understandable, I can't explain too well without waving my hands and showing you the parts. Maybe you can share with us how you are terminating the ends and maybe clarify that or rule out the possibility. It is only my opinion, based on my experience and education...I am always willing to learn, educate me!
Daron J. Wilson, RCDD
daron.wilson@lhmorris.com
>Why do they make Double paned windows?
If you are trying to tell me that having two panes of glass is somehow analogous to double shielded wire, your point is far too cunning for me to understand. FYI, double pane glass is done to provide an insulation factor, it provides a dead air space which is generally filled with an inert gas I think. I'm sure no window expert, but shielding is shielding. If you have a 100% shield then you have a 100% shield. There are double shielded and quad shielded cables on the market. Much of the cabling in high RF environments (duplexers, combiners, filters) where the alien RF source is very strong are double shielded, but generally this is two braided shields. Braided shields by nature are very hard to construct to 100%, so by doing a double shield you make it pretty close to 100%. The point is, if 100% shielding is not keeping out the stray EM noise or RF, then it is likely coupling to the shield and causing problems that way, and you most likely have a grounding issue. Yes shielding can be tricky, and yes every circumstance is different. I find it best to test the wire and see where it fails. NEXT? ATTENUATION? NOISE? It is much easier to determine a solution when you know which parameters are the problem.
>I am completely unfamiliar with your 1&2, 3&6 pair...
Well here goes....for ethernet, pins 1 and 2 MUST be on a pair of wires that are twisted together, this keeps the good signal in and the bad signals out. Pins 3 and 6 must also be the same way. If you are making cables and putting modular plugs on the end, and not putting the colors in such a manner that pins 3 and 6 are a pair, you have a major problem. From your description, it sounds to me like a split pair. That is if you started on pin one and just went white/blue pair, white/orange pair, white/green pair, white/brown pair. Hopefully this description is understandable, I can't explain too well without waving my hands and showing you the parts. Maybe you can share with us how you are terminating the ends and maybe clarify that or rule out the possibility. It is only my opinion, based on my experience and education...I am always willing to learn, educate me!
Daron J. Wilson, RCDD
daron.wilson@lhmorris.com
josepablomir
IS-IT--Management
I want to put my half penny.
If you are having troubles with a 35 feet wire, and a 75 feet work fine, your situation is a bit unstable.
Let me indicate some points:
a) EM problems in a 35 feet long wire are insane. Like other previous answer states, you should be too close to a RF high output device, like a commercial radio transmitter in which the installation could have been done with not too much precaution. Other sources of interference could be electrical transformers or electronic ballast for fluorescent lights. Try not to plug the HUBs in outlets shared with ballast.
b) This problem sounds to me like a jack or plug (or both) connection problem. Maybe the tool you are using is not good. Maybe the jacks are not good, and/or maybe the plugs are not good. It is not a good idea to reuse old jacks.
c) But the possibility that is more likely to be the responsible is the electrical ground. Yes, electrical ground. You must be sure all the devices connected to the network share the same electrical ground. If your HUBs don’t have grounded inlet don’t worry about them, but all your workstations must have, in EM adverse environments, the same wired ground. If your wiring, for being old, don’t have ground, you will have to run it. If different panels power different workstations you will need to correct this; and if you have different panels you could have different transformers and different ground reference points. But to be sure, all you have to do is to buy THHN wire number 12 and run for the floor a common ground path, just to experiment and see the results. Be careful when wiring, maybe is preferable to ask for help to an electrician (120 Volts are not Ethernet signals, they can kill!!!).
d) For least, if you use shielded wire, remember to ground only one extreme; usually the one in the HUB side, and cut the shield ¼ inch of the contacts in the plug.
Regards,
_________________
Jose P. Mir
jpm@jpmir.net
If you are having troubles with a 35 feet wire, and a 75 feet work fine, your situation is a bit unstable.
Let me indicate some points:
a) EM problems in a 35 feet long wire are insane. Like other previous answer states, you should be too close to a RF high output device, like a commercial radio transmitter in which the installation could have been done with not too much precaution. Other sources of interference could be electrical transformers or electronic ballast for fluorescent lights. Try not to plug the HUBs in outlets shared with ballast.
b) This problem sounds to me like a jack or plug (or both) connection problem. Maybe the tool you are using is not good. Maybe the jacks are not good, and/or maybe the plugs are not good. It is not a good idea to reuse old jacks.
c) But the possibility that is more likely to be the responsible is the electrical ground. Yes, electrical ground. You must be sure all the devices connected to the network share the same electrical ground. If your HUBs don’t have grounded inlet don’t worry about them, but all your workstations must have, in EM adverse environments, the same wired ground. If your wiring, for being old, don’t have ground, you will have to run it. If different panels power different workstations you will need to correct this; and if you have different panels you could have different transformers and different ground reference points. But to be sure, all you have to do is to buy THHN wire number 12 and run for the floor a common ground path, just to experiment and see the results. Be careful when wiring, maybe is preferable to ask for help to an electrician (120 Volts are not Ethernet signals, they can kill!!!).
d) For least, if you use shielded wire, remember to ground only one extreme; usually the one in the HUB side, and cut the shield ¼ inch of the contacts in the plug.
Regards,
_________________
Jose P. Mir
jpm@jpmir.net
You need to test your cables before you do anything else. The problems you are describing sound more like the previously mentioned split pairs or bad connections than anything else.
If you cannot get anyone to come to your location with at least a PentaScanner make some cables in varying lengths and take the cables to them for certification. Then use these "known good" cables as substitutes (run them across the floor) for ones you are having problems with.
If wires are not defective then swap the hub/switch. Next step is to strip down the network and see if you can get even 2 PC's to communicate without error. Once you get two working then add one at a time until you get failure. It could be anything. I have seen things like this when someone plugged PCs into both port 8 and the uplink port on a 8 port hub. Check EVERYTHING, then doublecheck.
It may be the only solution in your heavy RF/EM enviroment is fiber. BUT - Eliminate the simple before working on the hard...
Do you have any idea about what frequency the RF (EM) in your area is?
Am I missing something? What do windows have to do with this discussion? The first mention I see is:
> Kwick18 (MIS) Sep 6, 2002
> Why do they make Double paned windows?
Where did this come from?
If you cannot get anyone to come to your location with at least a PentaScanner make some cables in varying lengths and take the cables to them for certification. Then use these "known good" cables as substitutes (run them across the floor) for ones you are having problems with.
If wires are not defective then swap the hub/switch. Next step is to strip down the network and see if you can get even 2 PC's to communicate without error. Once you get two working then add one at a time until you get failure. It could be anything. I have seen things like this when someone plugged PCs into both port 8 and the uplink port on a 8 port hub. Check EVERYTHING, then doublecheck.
It may be the only solution in your heavy RF/EM enviroment is fiber. BUT - Eliminate the simple before working on the hard...
Do you have any idea about what frequency the RF (EM) in your area is?
Am I missing something? What do windows have to do with this discussion? The first mention I see is:
> Kwick18 (MIS) Sep 6, 2002
> Why do they make Double paned windows?
Where did this come from?
daronwilson
Vendor
Well? What the heck was the answer? You know we offer our suggestions and ideas, it is always nice if when you determine what solved it that you return and let us know huh?
It is only my opinion, based on my experience and education...I am always willing to learn, educate me!
Daron J. Wilson, RCDD
daron.wilson@lhmorris.com
It is only my opinion, based on my experience and education...I am always willing to learn, educate me!
Daron J. Wilson, RCDD
daron.wilson@lhmorris.com
Rayek. tell us how you made out? this post was interesting. Humor
= sorry I dont do windows, But! good advice from everyone. is the RF a VHF band or a UHF band curious? do you know the wattage dispersal?. Is the cabling near the base stations ( if any?)
= sorry I dont do windows, But! good advice from everyone. is the RF a VHF band or a UHF band curious? do you know the wattage dispersal?. Is the cabling near the base stations ( if any?)
telcomwork
Technical User
Definitley make sure to follow daronwilson's advice. Totally agree and have seen issues when NOT using pins 1/2 and 3/6 together dropping signal when making long cable, 50 ft and over!
Ok we are all talking about problems. Here is a problem I have just walked into. My new contact has asked me to trouble shot his Network. So I take my scanner and everything checks out fine. I get a call about 11:30pm with the guy telling me that my scanner has to be JUNK. Sure I say it was only 1500 but has always told the truth. Seeing I was not there at 11:30 I asked what he had turned on or moved and or changed from the time I was there at 2pm. He said that he was talking on the CB when his Kids said that it kicked them off the network. First thought was Bad Ant cable. so I kept asking what he was doing and then it came out He informs me that He has a TUBE 1500 watt Power kicker that he uses from time to time. He also tells me that when he looked outside at the BEAM ant he has (Moon Raker?) it was pointing in the direction of the Network.
Seeing this is not something he is doing all the time what can I do short of installing Fibre to keep his kids happy.
Thanks Jack.
Seeing this is not something he is doing all the time what can I do short of installing Fibre to keep his kids happy.
Thanks Jack.
daronwilson
Vendor
Wow so much to solve. We would refer to this as an "I. D. Ten T. problem" (ID10T or IDIOT). This guy is an idiot. Certainly you know that his CB transmissions are limited in power to somewhere around 4 watts, and this illegal amplifier is likely interferring with other people as well. Unfortunately when you splatter that kind of RF around, it will likely be in your computer, in the TV, backfed throught the cable, telephones, etc.
First, this may be an opportunity for ScTP wire. It isn't a lot of RF you are dealing with there, we've done networks inside AM broadcast facilities where the transmitter was in the same room. Anyway, I would sure consider screened Cat5e at this point. However, that means shielded plugs, a switch that is designed for shielded plugs, and a very good grounding plan.
Second, IF the antenna system is matched, and if he is not experiencing a high VSWR condition on the feedline, most of the power should actually be going up the coax and out the antenna. Antenna height above the house is important, as well as type of antenna (antenna pattern). Obviously the CB transmitter, cabling, RF amplifier, etc., should all be shielded and well grounded. You need to make sure first that the RF is going out the coax and out the antenna not radiating inside the house.
This one should be fun for you, and you could do fiber, but that will only eliminate the cat5 issues. This installation is such that the RF may be getting into the switch, or multiple computers, or anthing else in the house. Simply replacing the copper with fiber won't eliminate all chances of interference.
I have cleaned up many Ham radio installations for folks who were interfering with their cable/phone/lan while broadcasting legally on frequencies they were licensed for at the appropriate power levels. Make sure the radio install is clean first, then try to find where the unwanted signal is entering the system and plug the leak.
Oh, did I say this guy is an idiot? Last but not least this should probably be a new thread.
Good Luck
Good Luck! It is only my opinion, based on my experience and education...I am always willing to learn, educate me!
Daron J. Wilson, RCDD
daron.wilson@lhmorris.com
First, this may be an opportunity for ScTP wire. It isn't a lot of RF you are dealing with there, we've done networks inside AM broadcast facilities where the transmitter was in the same room. Anyway, I would sure consider screened Cat5e at this point. However, that means shielded plugs, a switch that is designed for shielded plugs, and a very good grounding plan.
Second, IF the antenna system is matched, and if he is not experiencing a high VSWR condition on the feedline, most of the power should actually be going up the coax and out the antenna. Antenna height above the house is important, as well as type of antenna (antenna pattern). Obviously the CB transmitter, cabling, RF amplifier, etc., should all be shielded and well grounded. You need to make sure first that the RF is going out the coax and out the antenna not radiating inside the house.
This one should be fun for you, and you could do fiber, but that will only eliminate the cat5 issues. This installation is such that the RF may be getting into the switch, or multiple computers, or anthing else in the house. Simply replacing the copper with fiber won't eliminate all chances of interference.
I have cleaned up many Ham radio installations for folks who were interfering with their cable/phone/lan while broadcasting legally on frequencies they were licensed for at the appropriate power levels. Make sure the radio install is clean first, then try to find where the unwanted signal is entering the system and plug the leak.
Oh, did I say this guy is an idiot? Last but not least this should probably be a new thread.
Good Luck
Good Luck! It is only my opinion, based on my experience and education...I am always willing to learn, educate me!
Daron J. Wilson, RCDD
daron.wilson@lhmorris.com
Rayek, your problem is an interesting one with many different ideas on what may be causing the problem. Please do post what you find out.
I'd like to understand more about the 1-2 and 3-6 pairing...cuz I'm not sure I understand what it means. I've done alot of cabling and rj 45 connecting with CAT 5 and 5e, and always use the same wiring scheme, 'cept for when a crossover cable is needed.
Nonetheless, what fixed it; inquiring minds want to know....
I'd like to understand more about the 1-2 and 3-6 pairing...cuz I'm not sure I understand what it means. I've done alot of cabling and rj 45 connecting with CAT 5 and 5e, and always use the same wiring scheme, 'cept for when a crossover cable is needed.
Nonetheless, what fixed it; inquiring minds want to know....
- Status
