Vibration Sensor into Analog Input

[Dood]

I have a Bentley Nevada Velocity sensor that claims to spit out a 100 mv/IPS signal. I'm being told that it will work with an analog input card (Siemens S7-1200 13-bit AI), but I'm not sure where to start with this one.


Datasheet for the velomitor is here:
https://www.bakerhughesds.com/sites... Piezo-velocity Sensor Datasheet - 141633.pdf


The AI Card is capable of 0 - 2.5/5/10 Volt signal, but how would I even calibrate it if there is no upper limit. Looking for a little guidance if anyone has experience trying to bring a 100 mv/IPS signal into a PLC.


Thanks!

 

[JRW]

Dood
I don’t think I would do it that way.
Have you looked at the cms1281 module for the s71200?
Your sensor would work with it and you can do complete vibration analysis

https://cache.industry.siemens.com/...perating_manual_en-US_en-US.pdf?download=true

 

[cardosocea]

I use Sensonics amplifiers for this application. It's pretty much standard across the site now.

I'm not sure how similar it is to the solution mk42 put here, but I believe Rockwell also makes something similar.

 

[Tom Jenkins]

I'd get the Ok from Bentley and maybe a reference to an existing application before committing to that. Bently Nevada is now part of GE, and Siemens and GE don't play well with anybody, much less with each other.

With proximity type probes, I used to use an adapter/signal conditioner like the one shown here: https://www.bakerhughesds.com/sites.../2020-03/BHCS32818A Proximitor Trifold_R4.pdf

My recollection is that they had 4-20 mA outputs and a BNC connector on the signal conditioner that would allow pugging in to get the prox probe signal directly for advanced diagnostics. This type of monitor is used with journal bearings and picks up shaft movement.

The probe you are showing is a piezo accelerometer. It should be mounted on the machine case or bearing housing. This type is used with ball bearings, where the shaft won't move radially. I wouldn't screw around with mV output on that kind of application. Metrix (and probably Bently) have piezo sensors that are loop powered and 4-20 mA output you can plug into an analog input card without screwing around.

 

[JRW]

Dood
What are you trying to accomplish exactly?

 

[Dood]

Tom,


You are correct - this is to be mounted directly on a gearbox. The problem is that the people who provided it are only required to provide a sensor with an "analog output" and their claim is that the 100mV/in/s sensitivity is an analog output. So I'm essentially trying to prove that it can't work so they will have to provide one with the 4-20mA that will make things simple.


JRW - The goal is to get a vibration reading to display in IPS (inches per second). So theoretically, if I'm reading .5V, my gearbox is vibrating at 5 IPS.


Thanks

 

[mk42]

I'm not sure how similar it is to the solution mk42 put here

wait, what'd I do now?

 

[JRW]

I saw that too.
Look how popular you are

Or he is at the pub

 

[Bit_Bucket_07]

Tom,


You are correct - this is to be mounted directly on a gearbox. The problem is that the people who provided it are only required to provide a sensor with an "analog output" and their claim is that the 100mV/in/s sensitivity is an analog output. So I'm essentially trying to prove that it can't work so they will have to provide one with the 4-20mA that will make things simple.


JRW - The goal is to get a vibration reading to display in IPS (inches per second). So theoretically, if I'm reading .5V, my gearbox is vibrating at 5 IPS.


Thanks

I don't know the brand name, but I know that such vibration sensors are available with loop powered 4-20 mA output. One of our clients uses them extensively in their plant to monitor fan and bearing vibration.

 

[mister x]

I'd get the Ok from Bentley and maybe a reference to an existing application before committing to that. Bently Nevada is now part of GE, and Siemens and GE don't play well with anybody, much less with each other.

With proximity type probes, I used to use an adapter/signal conditioner like the one shown here: https://www.bakerhughesds.com/sites.../2020-03/BHCS32818A Proximitor Trifold_R4.pdf

My recollection is that they had 4-20 mA outputs and a BNC connector on the signal conditioner that would allow pugging in to get the prox probe signal directly for advanced diagnostics. This type of monitor is used with journal bearings and picks up shaft movement.

The probe you are showing is a piezo accelerometer. It should be mounted on the machine case or bearing housing. This type is used with ball bearings, where the shaft won't move radially. I wouldn't screw around with mV output on that kind of application. Metrix (and probably Bently) have piezo sensors that are loop powered and 4-20 mA output you can plug into an analog input card without screwing around.

Good advice in keeping with the B_N conditioner. I think you'd be crazy to do this any other way.

 

[Tom Jenkins]

I think the supplier is nuts.

I would be amazed if a Metrix sensor like this http://www.metrixvibration.com/products/seismic
isn't significantly less expensive than that Bently unit. GE and Bently both have a long history of being the highest priced option.

 

[JRW]

I think the supplier is nuts.

I would be amazed if a Metrix sensor like this http://www.metrixvibration.com/products/seismic
isn't significantly less expensive than that Bently unit. GE and Bently both have a long history of being the highest priced option.

But they picked out the best there is
They didn’t cut any corners.

This must be an expensive gearbox