I would like to find a standalone switch that will respond to 2 mils displacement (p-p) at 5-7Hz.

So far all I can find are accelerometer based switches that lack the low-frequency sensitivity.

Thanks, David

Original Post

2 mils (I assume peak to peak) at 6 Hz is 0.004 g peak.  In mechanical switches travel of a spring against a force is the measure of vibration amplitude.  The threshold value is adjustable by changing the proximity of the magnet to the spring.  This can't reliably be done at 0.004 g.  

There are some electronic switches, PCB being a key manufacturer.  I suggest you talk to them about your needs.  Unfortunately the electronic switches are expensive when compared to the mechanical switches.  See http://www.pcb.com/Vibration_S.../Electronic_Switches.


Last edited by John from PA

If you have a lot of money take a look at the GE-Bently 330505 Velomitor with F-min at 0.5 Hz that is recommended for low speed hydro turbines. Be aware that trying to monitor low frequency displacement with fixed alarm/trip limits can be problematic. Good luck with that challenge!




I appreciate the leads.  Right now the only thing that seems to correlate with risk of failure is displacement vibration (overall is fine) but the only instruments that measure it well are the premium spectrum analyzers with appropriate probes.

So first we will test that a given switch can be configured to trip in this range.  I will be happy if I can find a switch that at least does that.

The challenge of what alarm settings to use and the overall effectiveness of detecting failures - may be much more difficult.  But first we need the hardware.

At the low frequency that you are looking at, have you considered using an "LVDT" type sensor. I have used these in the past on bottle capping machines, and on a relatively slow, I think 80cpm pick and place robot with good success. However, you do need a solid mount, most likely remote from what you are trying to monitor. 

The units that I used had a small (5mm dia.) tip that touched the machine with a 17mm diameter body that we mounted in a block with a fine screw adjustment. It provided an indication of a part not setting correctly and would flag the station. We could see 2-3 mils deflection with the unit. 

I neglected to mention that the output from the LVDT can then be input to a control device to alarm, or shutdown the machine in question. As I can recall the output is a voltage and we input that into the controller, adjusted for the nominal position at a zero point and then an alarm input at a value above the nominal. 

To be clear, I work for a machinery OEM and low cost vibration switches (Murphy, i-Alert, etc) are a common customer request for one of our machines.  The trouble is, they didn't work in our own tests because of the frequency issue.  Customers who use these probably don't realize the detection issues.

The PCB device is around 2x the cost of the ordinary switches and looks good on paper.  Assuming it works in our tests, we might get some traction with this over the cheaper switches if we can also endorse them for being "special" switch that is most effective.

It's also good to know of sensors that are out there, however very few customers are investing at that level for this type of machinery.

I-Alert is not in the same category as Murphy switch, and probably does not meet your requirements, and probably costs a lot more. Murphy switch (acceleration device) is useless on low speed pumps, low head hydro turbines, cooling tower fans and air cooled condenser fans that I have experience with. I hope as OEM, that useless vibration sensors would not be sold! I have seen a cooling tower fan throw 3 of 6 blades to the ground and the broken drive shaft onto the fan/motor deck without vibration detection by Murphy switch on motor support! I do not recommend these devices to anyone for any reason. A smoke alarm or a zero-speed detector would be a better choice!


Hi Walt,


I brought up iAlert because it monitors vibration and also cannot detect <10Hz frequencies.  You're right, it's not actually a switch.  But they are affordable.

My competitors and customers have been bolting on "reliability" modifications that don't hold up to closer scrutiny.  Sometimes a site manager says "we put these on all rotating equipment" regardless of its effectiveness.  I can't stop them.  I can only find a functional alternative, present the option, and describe why it stands out from the rest.

Last edited by dgeesaman

Add Reply

Link copied to your clipboard.