Vibration Pens, Stingers, and Sensor Mounting

There are several issues that limit the use of handheld vibration pens or “stinger” attachments on an accelerometer. One is the variability of pressure, angle, and location from measurement to measurement and from operator to operator, which makes it difficult to get repeatable readings. A second issue is the fact that the probe acts like a long, stiff spring, which has the effect of attenuating high frequencies. The third issue is the mass of the hand and arm holding the accelerometer is a big damper, which also absorbs high frequency energy.

Accelerometer manufacturers go to a great deal of effort in design and manufacture to make the accelerometer stiff and rigid so as to have a high resonant frequency that won’t interfere with machine vibration readings. This care in design and manufacture is largely negated by vibration pens or stingers and therefore limits the viability of this approach in vibration analysis and in building machine histories for long-term trending of machine health.

Conversely, the handheld vibration pen or stinger is an acceptable quick-and-easy method to get a single relatively low-frequency measurement. If you are looking for unbalance or misalignment on a 1,800 rpm or even 3,600 rpm motor (30 to 60 Hz), these methods will give a reasonable reading. If, on the other hand, you are looking for bearing tones in the range of 3 to 10 times rpm, or high-frequency bearing noise or gear-mesh frequencies, you will not be seeing them with this method — and that can give you a false sense of confidence and possibly an unwelcome surprise later on when the machine you thought was healthy takes an unexpected turn for the worse.

Quickly moving on, magnet mounts are either two-pole for mounting on round surfaces or flat for mounting on flat surfaces. The two-pole magnet will normally give you an maximum reliable frequency in the 1 kHz to 2 kHz range. Since the standard Datastick accelerometer is very small, and we use a fairly large magnet, we can expect to be at the upper end of that range if the magnet is firmly attached and not rocking.

A flat magnet mounted to a machined flat steel plate with a very thin layer of grease (just to fill the voids between the high spots) can give performance up to 5 kHz to 10 kHz but it is easy to get dirt in the joint and reduce this back to 1 kHz if you are not careful.

Using a heavy steel washer as a mounting point is a good compromise. Epoxy it in place with any quick-set epoxy, being careful to apply the very minimum thickness of the adhesive. You really want metal-to-metal contact at the high spots and the epoxy is to fill in the gaps. Loctite 330 with spray hardener 7387 is a good choice.

To avoid walking around with a magnet, my personal choice is the quick release connector with a “double D” threaded plug on the machine and the nut attached to the probe. This attaches with a 90-degree twist and is about the best compromise for effectiveness and ease of use but does require a little more investment in hardware.

The magnetic base is good for normal checking, but for detailed analysis nothing beats a bolted connection.

Derek Norfield
Director Of Applications