LabVIEW

cancel
Showing results for 
Search instead for 
Did you mean: 

ground loop + BNC 2120 behaving like antenna(problem)

tbob:

Yes, I too have been a Test and Automation (TnA?) Engineer for the past 15 years, but mainly at small private companies. Did a few years as a Co-Op student with GM where I got to program the big PLC's. Like hands on, especially estimate, design, spec out matl, build, program, debug, and train for systems I build in-house, and support them from cradle to grave. You picked up on my handle, actually it is a combination of titles to 2 Rush songs on the same album, The Analog Kid and Digital Man. Pretty much explains that I started tinkering with analog electronics at age 7, but now that I'm in my 40's I find myself mainly in the digital realm of computers and programming.

Cheers

-Pete

tmdar:

Any luck so far? I did not intend to sidetrack your issues, wanted to reply to tbob.
~~~~~~~~~~~~~~~~~~~~~~~~~~
"It’s the questions that drive us.”
~~~~~~~~~~~~~~~~~~~~~~~~~~
0 Kudos
Message 11 of 24
(1,513 Views)
AK2DM: lets take this off line. Send me an email: robert.a.duet@medtronic.com
- tbob

Inventor of the WORM Global
0 Kudos
Message 12 of 24
(1,508 Views)
Hi,
I have changed my NI BNC 2120 with NI ELVIS and amazingly it worked first time, before giving the statement that it is success I want to run few more tests but so far results are better than the connector I was using earlier.
I have connected the scope on the signal wire and I can see lots of ripples in the signal as I turned on my motors. although my signal wire is shielded but still it was picking up the RF and as I move the motor closer to the wire the distortion increased.
Do I need to replace all the wires with shielded wires or you think only replacing signal wires are enough.

I am making few more changes like replacing capacitors with high value(300uF) for motors, also I am placing one schottky diode in motor circuit to decrease the earlier shock that it creates due to back emf.

But NI elvis has atleast worked for me, I still dont know what was wrong with BNC 2120( I shall investigate and will let you know) meanwhile if you have any idea do let me know.
i am still working
Thanks
0 Kudos
Message 13 of 24
(1,508 Views)
Glad to hear you made some progress. I would shield all wires since it is obvious that EMF from the motor is bleeding into the wiring. Be sure to ground only one end of the shielded cable. The capacitor probably won't hurt, but if the noise is truly coming from motor EMF, you might find that the capacitor makes no difference at all. If the EMF is going through the power lines, then the capacitor will help. Experiment and use the scope.
- tbob

Inventor of the WORM Global
0 Kudos
Message 14 of 24
(1,504 Views)
Hi,

what do you mean by ground only one end of the shielded cable-- do you want to say that all cables which are acting like ground should have one common node
Thanks
0 Kudos
Message 15 of 24
(1,501 Views)
A shielded cable usually has its signal wires surrounded by tin foil or aluminum foil inside the outer jacket. A bare wire is also wrapped around the foil to make contact with the foil shield. This bare wire is the shield connection. If the cable goes from point A to point B, your signal wires inside the foil will have connections at both ends. However, the bare wire should be connected to ground only at one end, preferably the end away from the noise generating device (motor). If your shielded cable goes from the motor to the computer, ground the bare wire near the computer end and leave the bare wire at the motor end unconnected. Of course you should still have another separate ground wire connecting the motor ground to the computer ground, or connecting all grounds together. Without the shielded cable, the motor could induce noise into the signal wires. If the shielded cable were connected at both ends, the motor could still induce noise into the signal wires through the bare wire that runs along the enitre length of the shielded cable, which would render your shield useless. If the bare wire at the motor end is unconnected, the shield is at ground potential with respect to the computer, and effectively shields the inner signal wires from EMF noise.
Don't get a shielded signal cable confused with a coax cable. They are two separate things. Coax has only one center conductor, shielded wire comes in a variety of numbers of wires inside the shield. Coax is meant for RF and it has a characteristic impedence (like 50 ohms for RF transmission and 75 ohms for TV signals). Coax is usually terminated with a special connector, like BNC or Type N or SMA, etc. Shielded cable wires can be stripped and connected with crimped connectors on each wire or even soldered into place. Don't use coax for non RF signals because the impedence could upset any AC signals running through the wire (depends on the impedence of the generating and receiving devices). Another useful wire is twisted pair. Two wires are twisted along the entire length. You can have a number of twisted pairs inside one jacket. This is usually used in telephone applications, and is very good for differential signals. By being twisted, any EMF induction is cancelled by the orientation of the two wires. The induced noise in one wire is 180 degrees out of phase with the induced noise in the other wire which causes cancellation. You can even get shielded twisted pair wire which is the ultimate to use with differential signals and would probably be your best bet to use. Look into this type of wire and use it if affordable. This will go a long way to protect your signals from the motor EMF noise (again, ground the end away from the motor).
Whew, this is getting too long...
- tbob

Inventor of the WORM Global
0 Kudos
Message 16 of 24
(1,498 Views)
Hi,
once again thanks for the very comprehensive reply on shielding of cables and my fault once again I was using the shielded coax cable.
I will get other cable(twisted) and will inform you about the results.
Thankyou very much
0 Kudos
Message 17 of 24
(1,473 Views)
Hi,
today I have got shielde wire and twisted two pair wire but I was not able to fine even at frys electronics any twisted shielded wire. Now I think using shielded wire is better option as compared to twisted wire.
what do you say
Thanks
0 Kudos
Message 18 of 24
(1,462 Views)
Shielded wire will be fine. Let us know how it turns out.
- tbob

Inventor of the WORM Global
0 Kudos
Message 19 of 24
(1,459 Views)
Hi,
still disappointment, infact a day back I was able to run and now again same thing is happening. range sensor works fine if motors are not turned on, I have connnected the O scope on signal line aswell. SO when I turned my motors on I can see rapid pulses as motor speed up so is the pulses.
I have connected shileded wire. Shielded wire is going from sensor to NI elvis and is grounded there at NI elvis. Now to see behavior at Oscope without wire being connected to NI elvis means no labview the sensor is stable and so are pulses at Oscope even motors are runing at full speed.
But moment wire is connected to NI elvis then O scope also shows same rapid pulses. Now I dont know why this thing is happening even when I am using all shielded wires.
What do you say now
thanks
0 Kudos
Message 20 of 24
(1,446 Views)