 vipermCarpe DiemPremium join:2002-07-09 Winchester, CA 3 edits | OFF TOPIC magetic inteference on cableing / switches Picked up a job to rewire a medical office. They are into imaging. The telco/ switch room is in the same room as some HUGE ASS helium cooled ( I think from what I was told) giant electro magnet thing??
When the imaging is going on the latency across the network goes from less than 1 ms to upwards of 4000-5000ms when these electro magnets are energized.
They have multiple switches around the other offices that all aggragate to this one room. We are tasked with getting rid of all switches in other locations and centralize into this one room. Any thoughts of how to prevent this latency spike?
I thought of running shielded cat 6 or cat 5 but not sure if this will fix the issue with the noise or magnitisim getting in to the switch or thru the cabeling. Cant really test with pulling cables off line or switches during the day and I sure as hell dont want to try and turn this thing on after hours and risk damaging something.
Anyone ever run into this? I was also thinking of using shielded patch panels.
Thoughts and or suggestions are much appreciated.
-- »www.accelwireless.com ComTrain Certified Tower Climber. Wireless and IT consultant. Proficient in Mikrotik |
|
 LLigetfa join:2006-05-15 Fort Frances, ON kudos:1 | Ja, I believe it. I was called in to solve an intermittent problem for the PoCo that occupied three floors of a high-rise building. Three or four times a day, the network would flake out. It was always first thing in the morning, noon, 1:00 PM and quitting time. Everyone thought it was related to just higher usage at those times but I proved that to not be the case.
Turned out they had CAT5 cable runs going through the elevator shaft and they ran the network cables too near to the power cable for the elevator. Whenever the elevator had a full load, the motor drew more power and induced a larger magnetic field. I could feel the current flow simply by grabbing onto the insulated power cable just like one can feel it grabbing the cable on an arc welder in use. -- Strange as it seems, no amount of learning can cure stupidity, and formal education positively fortifies it. -- Stephen Vizinczey |
|
 | reply to viperm This is probably completely OT as it was a different problem, but I had some wireless gear on an FM tower that I was having tons of problems with Ethernet negotiations.
When I switched the cabling out fro ToughCable Lvl1 to Lvl2 the problem went away completely.
If I was going to re-run cable, I would get shielded cable like you are thinking and even cable with the cross-talk divider as that seemed to make all the difference in my case. |
|
 vipermCarpe DiemPremium join:2002-07-09 Winchester, CA | reply to viperm But what about the actual switch itself will any switch be okay to use as long all cable runs, patch cables and patch panels are if the shielded variety. Do I need to spec a specialized switch? |
|
|
|
 TomS_Git-r-donePremium,MVM join:2002-07-19 London, UK kudos:4 | reply to viperm Faraday cage?
I guess it does depend on whether this is interference induced into the cables, or the circuitry of the routers/switches themselves.
Perhaps you need to make some temporary runs of shielded cable along side the existing runs to some users that are experiencing issues, and see if they experience the same kind of issues. If you cant re-produce it, then perhaps its just as simple as replacing all of the cabling. If it remains, you've got bigger problems.
It might be worth recommending a move away from this room due to the issues being experienced, rather than concentrating more stuff in it. |
|
 vipermCarpe DiemPremium join:2002-07-09 Winchester, CA | I agree gonna talk to the IT manager, my friend and see if we can try his first I really don't want to pull all new cables only to find out its the Switch. Then again we have to pull all new cable to most locations... They have lead blankets for the techs there, was going to suggest to him to take a lead schmock and throw it over the switch to see if it makes any difference. |
|
 WHT join:2010-03-26 Rosston, TX kudos:5 | reply to viperm said by viperm:We are tasked with getting rid of all switches in other locations and centralize into this one room. Sounds like the "retasker" decision person bit off more than they could chew. Try to talk them into a different room. |
|
 John GaltForward, MarchPremium join:2004-09-30 Happy Camp kudos:5 | reply to viperm Shielding in cable only affects the electric field, not the magnetic field.
Also, a Faraday cage would prove ineffective as well. |
|
 vipermCarpe DiemPremium join:2002-07-09 Winchester, CA 1 edit | reply to WHT Ha ha not really just helping a Friend out as well as getting paid to do it. He just got hired as the IT manager I suggested the same thing but locations are limited.... |
|
 John GaltForward, MarchPremium join:2004-09-30 Happy Camp kudos:5 | said by viperm:I suggested the same thing but locations are limited.... Distance is the only (practical) solution.
I've built this kind of facility before. All of the gear not directly related to the imager was located as far as possible from the machine as we could get it. |
|
 WHT join:2010-03-26 Rosston, TX kudos:5 | reply to viperm said by viperm:locations are limited.... Change location, put up with crappy service...Pick one. |
|
 | reply to viperm What he ^^ said.
You have NO option in this room.
Either the equipment moves, or they can stuff it and suffer.
There is no viable way to fix what you're dealing with, and most certainly not at the energy you're dealing with. (Look those things up and you'll soon see there's no messing with them).
Anything you do in there will be a cobbled band-aid solution that either won't work, or will work like shit.
You need to drill it into the client's head that the equipment being in there simply can't happen if they want the results they expect. Either the networking stuff moves, or the magnet moves. lol |
|
 vipermCarpe DiemPremium join:2002-07-09 Winchester, CA | My buddy says they are the "power supplies" for the imaging machines not the electro magnetic part of it, not sure if it makes any difference.
Several of their offices are like this. I will be looking at a local office tomorrow evening to see the layout myself and run some tests. Shielded and non shielded cable into the room and around the room to another location if possible to get the best results.
What we do tomorrow will dictate what we can expect at other offices.
I have already speced out shielded patch cables, panels and cable.
-- »www.accelwireless.com ComTrain Certified Tower Climber. Wireless and IT consultant. Proficient in Mikrotik |
|
 John GaltForward, MarchPremium join:2004-09-30 Happy Camp kudos:5 | Helpful hint: don't have any ferrous tools in your pockets around the machine when it's operating...  |
|
 | reply to viperm Viperm, it appears that you are involving yourself with Magnetic Resonance Imaging. Thoroughly educate yourself, (the basics are really not that complicated), before you jump into this project. The age and strength of the magnet determines the Gauss line at distances measured from the magnet. I assume that these days the magnets are very shielded and the strength of the field does not extend very far from the magnet. Exposing a conductor to a changing magnetic field induces a current in that conductor. Do not forget that magnets attract metallic objects. I am surprised that your were not briefed on magnet safety. Most conductors entering the magnet room usually must pass through a wave guide to eliminate frequencies detrimental to image acquisition.
In the early days I saw people wrestle with unattached pipes in the area of the magnets. I also saw a carpenter loose all of the nails in his pull behind wagon because he did not heed posted warning signs. Do not be responsible for quenching the magnet, if that is still possible! |
|
 vipermCarpe DiemPremium join:2002-07-09 Winchester, CA | I know about electro magnets... I am and was not even going to walk into the room while it was functioning....
Thanks for all the input.. |
|
 John GaltForward, MarchPremium join:2004-09-30 Happy Camp kudos:5 | reply to viperm »www.willis.com/documents/publica···9_V6.pdf |
|
 | reply to viperm Pull fiber instead? |
|
 | reply to viperm Delco Wire makes a STP wire (it has a solid aluminum shield that is about 1/16's of an inch) when coupled with a proper grounded rack with emi shielding and proper grounding techniques should minimize the effects of this as low as you can.
That said its not cheep its at least $0.80 per foot.
That said as others have told you its a better idea to just relocate this, This is a bad room for that and you will never fully stop the effects of this device. |
|
 | reply to viperm What gunther said above. Pull fiber. Thats it. |
|