Sign up Calendar Latest Topics Donate
 
 
 


Reply
  Author   Comment  
Curran919

Avatar / Picture

Sr. Member / Supporter
Registered:
Posts: 599
Reply with quote  #1 
It has always been beaten into me that you should always use the correct cable length with prox probes and that it will 'not work' if the cable length between the probes and the proximitor is wrong. I never got a good answer for my question of, "so exactly what happens when I use the wrong extension cable?"

I found myself in a situation yesterday where I had to break this rule, and everything went just dandy. The sensitivity obviously changed and I had to compensate for that and recalibrate, but the response was just as linear as it was before and I see no reason that using the wrong cable length is bad, unless this affects the frequency response (don't see how). I've been told the system must be 3m, 5m or 9m, but now I've run the same probes with 3.6m and 5.4m and the results look great.

Any further insight?
OLi

Sr. Member
Registered:
Posts: 1,917
Reply with quote  #2 
Sensor is just a coil attached by the cable to be a part of a electrical oscillator in the proximitor that change in frequency due to the change in coil inductance due to target insertion and movement. If you enter another cable you change the frequency of the system, so you were lucky. In my youth, optimistic until proven otherwise I tried and it didn't work, even trying to change the properties of the extension I could not make it work with a (substantially) longer cable. Hope it helps and buy a lottery ticket while at it.
__________________
Good Vibrations since early 1950's, first patented vibrometer 1956 in the US.
http://www.vtab.se
John from PA

Sr. Member
Registered:
Posts: 963
Reply with quote  #3 
I will give you the same answer I did back on January 24th in the thread at http://www.machineryanalysis.org/post/prox-probe-calibration-8340847?highlight=proximity+probe+cable&pid=1294615073.  That thread by the way is well worth a review by anyone wondering about cable length in proximity probe systems.

Quote:
Originally Posted by John from PA

As far as the length issue, what you suggest (calibrate as a system) and ignore cable length mismatch [text in bold inserted by JGW] is not recommended. At issue with what you say is the system is generally clearly identified as to cable length, scale factor and target material.  So lets say you take a 9m system observing 4140 steel, swap out the cable to make it a 5 m system and run the gap vs. voltage curves.  The system will have higher than normal output, something like about 9 V/mm (220 mv/mil).  You become aware of this because you run the gap vs. voltage curves and establish the "modified" scale factor.  You can set the modified scale factor in most diagnostic instrumentation, but what do you do with the monitor?  Most monitoring systems meeting API 670 are calibrated for 7.87 V/mm (200 mV/mil) with very little ability to change it except by small amounts.  There are some specialized probes (nuclear service for instance) that use 3.94 V/mm (100 mV/mil) which is also user selectable in most monitors.

OK, so perhaps you aren't using a monitoring system but some other system that has the ability to fine tune to the modified scale factor.  You change jobs, get sick, leave the company, a cable is replaced, whatever and someone "fills your shoes" in the future. New personnel aren't aware of the non-standard system but encounter the 5 m cable being used with a clearly identified 9 m proximitor.  They do what they think is the correct thing and change the cable to 9 m.  Now the system puts out the wrong voltage.  In fact it could be 10% in error.  

Hopefully you can see using the system as intended (or "correctly") is the better scenario.  In a matched system, you can change out any single component and result in no more than 4% error.


There is a humorous side to all this.  A gearbox equipped with proximity probes was being prepared for shipment and the normal electrician was on vacation so an alternative individual was doing the work.  The probes were 1 meter with the needed additional 4 meters of extension cable to form a 5 meter system.  The alternative electrician could not understand why the 4 meter cable was needed on several probes that could reach the proximitors without the 4 meter extension cable.  Consequently he chose to leave them out.  When commissioned the gearbox had some really unusual vibrations (very flat orbits as I recall).  The missing extension cables were the cause.
Curran919

Avatar / Picture

Sr. Member / Supporter
Registered:
Posts: 599
Reply with quote  #4 
Good points John. This part of my job is in testing of prototype pumps, so instrumentation is only installed a few weeks at a time, and therefore not a consideration to be monitored. With this, I think we do a lot of throwing instrumentation at the pumps that work in the tight/unique constraints that we get and is by no means a typical installation. I actually just got a copy of API 670 last week and see now where it defines this 7.87 V/mm sensitivity. This is definitely at the top of my reading list.

I dug up an old sensitivity report from when we bought these sensors (Vibrax TQ101) where an old colleague determined the sensitivity of the linearity and range to common mistakes (in a much more controlled manner), such as using the wrong extension cable, multiple or no extensions, probe/prox mismatch, etc. He even showed that changing the power supply from 24V to 15V resulted in a quadratic rather than a linear response with the same setup. I don't know how that works theoretically, but its incredibly interesting.

The effect of using two intermediate cables was shown to indeed have a small effect in the system that I am using, but definitely some strangeness on a legacy system. In German, but I think you understand the context:

doubleintermediatecables.png 

Previous Topic | Next Topic
Print
Reply

Quick Navigation:

Easily create a Forum Website with Website Toolbox.