AC prox noise issue

[Guest]

Happy Holidays!

I have an application where an AC capacitive prox switches to its on-state whenever a VFD is running. Here are the particulars of my problem:

1) all controls are 120vac, including PLC I/O. The proxes are 2-wire AC capacitive.

2) the OEM company I work for (in their Infinite Wisdom) has shipped this machine to our customer.

3) Looking to sense a black plastic tray, not enough refectivity for a diffuse photoeye, and no mounting area available to be able to use a through-beam or a retroreflective photo-eye.

4) a capcitive prox works great! Great, that is until the machine starts running, and the VFD noise kicks in.

5) tried running cabling direct from the control panel where the PLC is to the prox location (sometimes a 10m cabe does have advantages), but no change in operation.

The solutions I see here are:

1) a completely isolated circuit for the proxes.

2) a filtered circuit for the proxes.

3) add a resistor to the existing prox/PLC input circuit, whether in series or as a shunt, I'm not certain.

4) bang my head with a rock.

The Install for this machine is scheduled for December 27, so if anyone can point me in the right direction, I'd certainly appreciate it.

Regards,

Jordan

 

[Eric Nelson]

It may be that the prox. switch itself is susceptible to the VFD or motor 'noise'. I found THIS article from Turck. Might be worth swapping sensors...

Let me guess... You're not using quick-disconnect sensors, so swapping them out will be a pain...

beerchug

-Eric

P.S. I'm not a fan of using AC inputs, much less using 2-wire sensors, but that's just me...

 

[Johnny Rotten]

Are you using a SHIELDED Proxy? Might be worth a try.
John

 

[jimcav]

noise

try putting chokes on the motor side of the vfd, put the ones that clip arond the wire. a long shot but they mite work.

 

[panic mode]

is the cable from VFD to motor shielded?

 

[rsdoran]

I am wondering if anyone has tried adjusting the sensitivity of the prox?

 

[mordred]

I also wonder if the VFD is causing harmonics. You might want to measure from ground to neutral of your 120 volt line while running. If you have an oscilloscope hopefully you don't see a frequency greater than 60 hz. The 60 hz can be generated through emf or a phase out of sync. you should also have less than 2 volts on this test. Keep in mind this is a remote and unlikely possibility I'll post an article on harmonic testing with a true rms DMM. I would also check your grounding. make sure you have no ground loops. Often times people shield both ends on sensers etc creating ground loops. Senser sheilds should only be terminated at one end for this reason.

 

[DickDV]

The first matter to settle is that the proximity sensor is installed using all recommended noise suppression techniques.

Second, reduce radiated noise from the VFD. There are numerous threads on this issue in this board archive but you must note that there are three different sources of VFD noise. The first and most common is input harmonics and this is not likely to be your problem.

The second is radiated EMI and RFI from the motor leads. This is most likely the source of your trouble. Unfortunately, most anything (conduit, shields on motor leads) to reduce radiated noise dramatically increases problems with the third source--high frequency pulse currents in the ground system often called common-mode noise.

Ususally the solution for common mode noise is an input transformer for the VFD. If you decide to do that and the motor horsepower is under about 10hp, you may want to consider changing the drive and motor to 230V rather than 460V operation. The motor is most likely dual wound, the transformer can be 2/1 stepdown, and the only real change is the inverter which must now be 230V.

I have never had a case where dropping the voltage to 230V has not fixed the noise problems. Hold this one in reserve for the tough cases and try the other stuff first.

Good Luck

 

[panic mode]

i would keep this one as a last resort:

4) bang my head with a rock.

 

[mordred]

bang my head with a rock

that may hurt the rock