How to eliminate EMI and RFI

We have a 6 unit of primary air fan having speed of 1780 rpm, 60 hz, with installed shinakawa CBT420 accelerometer, we are facing an frequent vibration spike which cannot captured by analyzer due to not determined when it will spike again, during investigation we found out that when using radio close to accelerometer is the vibration change rapidly and it spike.

how can we eliminate this kind of problem of RFI (radio frequency interference)?

regards,

 

"Failure is part of our life, if you do not fail you will never learn and if you'll never learn you'll never succeed"

Original Post

"we are facing an frequent vibration spike which cannot captured by analyzer"

What analyzer are you using? I hope you are not trying to do waveform or spectrum analysis from that sensor! This accelerometer has 4-20 ma output, so all you could use for detecting spikes is a DCS plot history (if available) or set up a chart recorder. If there is only one sensor with "spikes", then replace it. I have found these type of sensors to not be reliable.

Walt

Walt Strong posted:

"we are facing an frequent vibration spike which cannot captured by analyzer"

This accelerometer has 4-20 ma output, so all you could use for detecting spikes is a DCS plot history (if available) or set up a chart recorder. If there is only one sensor with "spikes", then replace it. I have found these type of sensors to not be reliable.

Walt

Walt is correct, device will not yield dynamic measurement.  Specs are at https://www.shinkawaelectric.c...wnload/cbt420_cj.pdf.  The CBT-421 does have an AC output.

Walt Strong posted:

"we are facing an frequent vibration spike which cannot captured by analyzer"

What analyzer are you using? I hope you are not trying to do waveform or spectrum analysis from that sensor! This accelerometer has 4-20 ma output, so all you could use for detecting spikes is a DCS plot history (if available) or set up a chart recorder. If there is only one sensor with "spikes", then replace it. I have found these type of sensors to not be reliable.

Walt

Walt,

This spike is captured by the DCS and as usual the operator calls for vibration data gathering at that equipment by using my portable analyzer, but during FFT and TWF analysis there is no indication of abnormality with <1 mm/s rms result. but in deep of investigation during using a radio close to the accelerometer there is an interference and the vibration is begun to raise. power cable and instrument cable are in different trays. 

regards,

Many (a very long time) years ago I had this problem with B&K vibration monitor. I proved that clicking (press to talk) with hand-held radio with antennae near monitor would cause spikes from RF. I confirmed this by placing aluminum foil behind the plastic face of monitor and stopped the interference effect. Just do a simple test with radio near the sensor and then near the monitor to check if there is any response. Another source of RFI can be motors with VFD that have poor shielding of power cables. Why chase a probable electrical problem with vibration analyzer, other then proving the spikes are not real vibrations!

Walt

"is there any countermeasure"

Find the RF sources(s) (portable radios, motor VFD, capacitor switches, etc.)

Find the location in monitor system with sensitivity (sensor, cable, monitor)

If sensor, cable or connector has fault, then replace them.

Separate sensor cables from proximity to power cables

Relocate monitor or place inside shielded enclosure

I cannot make anymore guesses. You are there, so play detective!

Walt

Add Reply

×
×
×
×