Join or Manage Your Profile
Posting Boards
Machinery Condition Monitoring and Predictive Maintenance
Posts About vibration/alignment/balance
Cable Question|
Go
![]() |
New
![]() |
Find
![]() |
Notify
![]() |
Tools
![]() |
Reply
![]() |
|
I am considering using a Beldon 6 conductor shielded cable (20 AWG) for some online sensors. The cable is red and has 6 wires wrapped in plastic and a stainless braid. (PN 83606 002) The individual conductors are insulated but not shielded from each other. The maximum length would be about 90 feet. Do you think I will have any issues with signals crossing over?
In the past I have used individual twisted pairs on their own or in a Unitech cable bundle which is used for instrumentation. Thanks. ensing-dot-ron-at-irvingtissue-dot-ca |
|||
|
I'd recommend a cable with twisted pairs, with each sensor connected to its own pair. Otherwise, cross talk is very likely, especially if some of the channels have higher levels than the others.
Jon Spintelligent Labs |
||||
|
Thanks Jon. My thoughts are that direction as well. I'll play it safe.
ensing-dot-ron-at-irvingtissue-dot-ca |
||||
|
Jon,
Are you saying that each pair should be twisted and shielded? Thanks, Danny |
||||
|
Danny,
The belt and suspenders approach would be for each channel to be on a twisted, shielded pair, but this is usually overkill. I might do this if I had a tach pulse in the same cable with accelerometer signals because the tach pulse will be much stronger than the accelerometer signals and the waveform (a pulse) will generate a lot of harmonics. In the case of all accelerometer signals, the levels will all be the same order of magnitude, so crosstalk between non-shielded pairs will be insignificant. Individual twisted pairs without separate shields will be fine. A single, overall shield would be a nice touch, but isn't essential. Shields can create some problems if not connected correctly. Grounding and shielding is a science of its own, but for our purposes it's not too difficult: Shields must be connected to ground. Otherwise, they can act as antennas and actually add to the noise. But to prevent ground loops, the shields should be connected to ground at a single point, not at both ends of the cable. A common approach is to connect the shield to ground at the instrumentation end, but leave it floating (disconnected) at the sensor end. A ground loop will result in 60 Hz noise. I've seen cases where making the final connection has resulted in a spark when the ground makes connection....this is definity a sign of trouble! Jon Spintelligent Labs Oh....a random thought....if installing sensors in intrisically safe areas, there are some other considerations. Follow the IS documents to the letter to have a safe installation. |
||||
|
Danny,
Yes, and shield should be grounded at only one end. It's the best. Especially on the long distances. Regards |
||||
|
Jon and Alexander,
Thanks. Danny |
||||
|
Hey Jon;
Thanks for the heads up on the shield. I'll be sure to check that it's done right. ensing-dot-ron-at-irvingtissue-dot-ca |
||||
|
What is the acel? 10 mV/g 100 500 or 1000? As Jon pointed out the 5V TTL could ruin your day. Or what is your transducer? Is it an acel? Continually hot muxin' in for full-time monitoring or plug-in?
Cordially, Sam Pickens pdmsampickens@gmail.com |
||||
|
Sam;
100mv/g, full time monitoring, no TTL. ensing-dot-ron-at-irvingtissue-dot-ca |
||||
|
Ron,
I use CAT-5 cable for temporary monitoring up to 400-ft. The cable has 4-twisted pairs with an outer shield. This handles 4 accelerometers or speed tachometers. Walt |
||||
|
Walt,
Have you ever ran a tach signal in the same CAT-5 cable with accelerometer signals? Theory says that would give a lot of cross-talk, at 1x and harmonics. I've never tried it so I can't speak from personal experience. One note about pairs....be sure to use them as pairs! The high and low pins of each accelerometer connected to the same pair. I think this goes without saying, but people sometimes get inventive. Also on the subject of cables....routing is important. It may seem handy to tie-wrap your cable to the machine's power cable, but don't do this. Depending on the voltage and current in the power cable, a lot of noise can be induced in the signal cable. Keep signal cables well separated from power cables, don't run them parallel, and where they must come close together, cross them at right angles. I haven't seen many problems in this regard, but it's best to avoid the problem from the start rather than troubleshoot it later. Jon Spintelligent Labs |
||||
|
Jon,
I have run one TTL tach signal and 3-accelerometers on same CAT-5 cable. I made my own BNC termination boxes so that I can use standard CAT-5 cable terminations. Walt |
||||
|
I thought CAT-5 was used for the internet. Is there something else that goes by CAT-5?
|
||||
|
CAT-5 is the cable usually used for network connections. It has 4 twisted pairs with an overall jacket, but no shield. It's a great low-cost cable that you can buy at any home center store, and RJ-45 connectors are very inexpensive. With a little practise, they can be installed pretty quickly.
This cable has a PVC jacket, so it's not nearly as tough as normal accelerometer cables but it will work fine in protected locations and for temporary connections. Jon Spintelligent Labs |
||||
|
There are many Flavors of CAT-5 cable including with shielding and suitable for outdoor or burial installations.
Walt |
||||
|
So...I can't believe Cat 5 works with accels. I am not doubting anyone, I just would have never thought of that.
Right now I am installing a monitoring system for a client. I have been trying to come up with a way of a cheaper cable to use. There is conduit and cable trays. Monitoring device is located in MCC room. Cables running out into the plant to the machines. This is somewhat of a finish to a incomplete project from many years ago. So I could run a Cat-5 to a jbox at the machine and terminate my "normal" cable (CTC) and the Cat 5? |
||||
|
J.,
I have not done a permanent installation with CAT-5. I have only used it many times for testing in power plant and ship engine room environments. I don't know of anyone else who has done this. You have to decide whether it is worth the effort, technically acceptible and saves money. Walt |
||||
|
J.,
I have some ideas about a termination box system to do just that. Drop me an e-mail and I'll send you a sketch of what we're thinking of building. Jon jon@spintelligentlabs.com |
||||
|
I really like the CAT5 cable idea for temporary test connections. Finding enough extension cables is always a hassle on large projects...I did one shipboard monitoring project where I brought along all the cables we had AND bought 7000' of extension cable on the job site!
CAT5 cable has the advantage of being available everywhere these days and being cheap ....errr.... cost effective. RJ-45 connectors can be crimped on pretty quicky with a little practice and the tools to do so are cheap and available. I've whipped up a simple termination box to adapt CAT5 cable to BNC connectors and terminal strips. A typical application would be to use accelerometers with short integral cables with BNC connectors on a machine in an inaccessible location, say an air handler. Connect the accels to the termination box, plug in a length of ethernet cable to a termination box outside the air handler, shut the door and connect your data collector to each channel in turn. I know this will make my test projects a lot easier to set up. I've attached a pdf showing the system. If anybody is interested in these, we can supply them for $75 each. Thanks Jon Spintelligent Labs 206-855-9284 Cat-5_termination_Box.pdf (60 Kb, 27 downloads) |
||||
|
| Previous Topic | Next Topic | powered by eve community | Page 1 2 |
| Please Wait. Your request is being processed... |
|

