Sign up Calendar Latest Topics Donate
 
 
 


Reply
  Author   Comment  
Curran919

Avatar / Picture

Sr. Member / Supporter
Registered:
Posts: 599
Reply with quote  #1 
We are working more with submerged accelerometers recently. Obviously we just use welded hermetic sensors and we generally use sensors with integral cables, such that the 'connection' should be IP68. Vendors usually only guarantee them to IP65 though. Whatever. I've also ad hoc waterproofed some other sensors with Silicone, shrink tube and even Teflon tape.

However, is there a risk to the sensor if the sealing is lost, or is it just a risk to the measurement?

From my understanding, if you get humidity in the connector for a charge-output accel, you will decrease the leakage resistance, and decrease the DTC, and therefore low frequency cutoff. If you get water in there, then you essentially just short the sensor and you can't read the charge signal, but there should be no harm to the sensor.

For an ICP sensor, I'm not so sure. I imagine, as you decrease the leakage resistance over the connector, you will leech current from you constant 2 or 4 mA and the bias voltage will drop until it is out of spec of the sensor, and then the sensor just turns off.

Is that correct? Should soaking the accel connector essentially not damage the sensor?
Walt Strong

Sr. Member
Registered:
Posts: 889
Reply with quote  #2 
"Does soaking an accelerometer kill it?"
Yes, if water/moisture gets into accelerometer body, cable or cable connector at opposite end. I have installed a few ICP accelerometers that were rated for underwater application on vertical pumps. I also installed accelerometers on ship rudder. Output ranged from good to intermittent to failure. Make sure you have a solid and protected mounting, rugged cable mounting and protection, and keep moisture out of connector at terminal box. Make sure you have a lifetime (yours) guarantee with exchange if accelerometer goes bad. Think twice about considering for long-term monitoring.
 
"For an ICP sensor, I'm not so sure. I imagine, as you decrease the leakage resistance over the connector, you will leech current from you constant 2 or 4 mA and the bias voltage will drop until it is out of spec of the sensor, and then the sensor just turns off."

Basically water acts as a short circuit, and it may be intermittent or continuous. Sometimes it is possible to get some data with intermittent (think rapidly plug/unplug connector), but the waveform suffers and there is a lot of low frequency "ski slope". I don't know if accelerometer would work again, if you baked the water out. Dewatering a vertical pump sump or sending a diver to work in poor water conditions can be a real reality check for conducting vibration tests and monitoring!

Walt
OLi

Sr. Member
Registered:
Posts: 1,914
Reply with quote  #3 
Many years of use of sensors in paper machine wet end environment indicate that the constant powering of the accel is a problem. You get electrolysis from the DC voltage in the connector if it get wet so that the internals get corroded pretty quick and eaten up. It also appear from condensing water so fill the space with something not corrosive.
__________________
Good Vibrations since early 1950's, first patented vibrometer 1956 in the US.
http://www.vtab.se
Curran919

Avatar / Picture

Sr. Member / Supporter
Registered:
Posts: 599
Reply with quote  #4 
Quote:
Originally Posted by Walt Strong

Yes, if water/moisture gets into accelerometer body, cable or cable connector at opposite end. I have installed a few ICP accelerometers that were rated for underwater application on vertical pumps. I also installed accelerometers on ship rudder. Output ranged from good to intermittent to failure. Make sure you have a solid and protected mounting, rugged cable mounting and protection, and keep moisture out of connector at terminal box. Make sure you have a lifetime (yours) guarantee with exchange if accelerometer goes bad. Think twice about considering for long-term monitoring.


I agree that long-term monitoring would not be good. I've only used submerged sensors for short term tests (like monitoring buried pipe or testing sewage pumps). In that case, I am validating the data every time I take a measurement and there is no possibility of lost data.

Quote:
Originally Posted by Walt Strong

Basically water acts as a short circuit, and it may be intermittent or continuous. Sometimes it is possible to get some data with intermittent (think rapidly plug/unplug connector), but the waveform suffers and there is a lot of low frequency "ski slope". I don't know if accelerometer would work again, if you baked the water out. Dewatering a vertical pump sump or sending a diver to work in poor water conditions can be a real reality check for conducting vibration tests and monitoring!


I imagine there is no way you will get water ingress into a hermetic sensor. If the sensor has an integral cable and the connector gets soaked, then yeah, that would be a real pain to bake the water out of, and then you wouldn't want to use the sensor underwater again anyway. If it is a 'temporary sealed' removable cable, then I can just cut the seal, remove the cable and let it dry out in ambient. Then reseal and it should be fine.

We were working on some vertical pumps and taking measurements on the column (below grade). It was easy for them to close the sump and use the main verticals to drop the water level to below the sensors, but not lower, so we bought an small inflatable raft and worked from that to mount and dismount our sensors. The safety guys on site weren't really sure how to react. It was in Bermuda though, so it was fine, but it looks like I am going to have to do that in the USA pretty soon. Lets see what they think of my inner tube!
Curran919

Avatar / Picture

Sr. Member / Supporter
Registered:
Posts: 599
Reply with quote  #5 
Quote:
Originally Posted by OLi
Many years of use of sensors in paper machine wet end environment indicate that the constant powering of the accel is a problem. You get electrolysis from the DC voltage in the connector if it get wet so that the internals get corroded pretty quick and eaten up. It also appear from condensing water so fill the space with something not corrosive.


Agreed. If you have water shunting your sensor, corrosion will knock it out quickly. If I can monitor the bias voltage though, and shutdown the sensor within a few minutes if it ever falls below nominal, then I should be able to save the sensor without any significant corrosion damage.

Essentially, I don't want to risk causing an overvoltage or overcurrent to the sensor as that's not something I can monitor and prevent.
OLi

Sr. Member
Registered:
Posts: 1,914
Reply with quote  #6 
I think all scenarios I can think of result in voltages get lower and current is limited to normally 2 or 4 mA so there will be no spectacular events that way, electrolysis normally take a while to make any significant damage, it is not instant and yes you will see on the bias when it goes bad. So unless you let any electrical eel inside :-) it is not so big problem. We have had sensors but not accels in hydroplants (yes, naturally MCV's, no supply voltage on them) on the wet side btw. 6 and 10 years on 6-10+m depth but then we had double sealed housing and it was difficult for the physical protection as trees, ice and maybe a moose and definitely eels but not electrical pass by from time to time so some were lost as they got knocked off. Inside containment is also interesting, in this case the MCV's survived longer than the hi spec accels :-). I have on the other hand worked with vibration and loose parts monitoring systems on reactor vessels and they worked many years so with the correct spec it is possible. Note, on, not inside. 
__________________
Good Vibrations since early 1950's, first patented vibrometer 1956 in the US.
http://www.vtab.se
Previous Topic | Next Topic
Print
Reply

Quick Navigation:

Easily create a Forum Website with Website Toolbox.