Hi guys

Some DLI reps. are trying to sell to me triaxial sensors. Supposedly they reduce the data acquisition time at least by half as they take measurements in the three directions at the same time.

Does anydoby has experience with this sensors?. Do they response the same way as the regular ones?

They can't take truly radial readings in the horizontal and vertical planes without repositioning. One has to be tangential.

It´s not the sensor per se that makes the trick, you need a box that sample 3 channels at the same time but I guess they were selling you one of these to. Barely maybe in the ISO standard nowadays, certainly not before. So you collect fast, but what do you get? It´s also very sensitive that you make sure the direction is always the same or the trend will be whatever you like so you need fastening points in all places at a cost, they may be tricky to get near a bearing. Worst case I have seen was on the foot of the motor, when that shakes if it´s an old motor it´s real bad. Olov

THe idea of using the triax for route collection is not a bad idea, so-to-speak, but.... There are considerations that need to be taken. Recognize that the data on the off axis will be attenuated somewhat. The alarm levels will need to be adjusted for the attenuated signal, as well.
It is very important to use a mounting method that "Keys" the sensor in the same orientation every collection date. I would suggest a quick connect method, and stay away from a magnet mount.
I personally do not see the advantage to "screwing" a triax down to a mounting target, every time you take a reading. An experienced hand with a single axis sensor and a high quality magnet is just as fast (in my opinion) as trying to screw a triax down on a target. Plus, there are more opportunities to place a single axis sensor directly over the "radial position" of the bearing point.
I have used both, and I would stay with the single axis sensor, and a smaller box for Route collection.

Just my two cents :>

Plus the expense of the tri-ax compared to the uni-ax.

I would steer away from triax.

As mentioned in earlier replies, triax sensors are very expensive. The one we used had integral cable (supplied with the hardware)so if that is damaged you can buy another one. If you really want to you can make a block to which you can fit 3 small general purpose accels for a fraction of $$ to do your own trial - then there's the 3 channel box requirement Olov mentioned.

Went through trying all this on a paper mill some years ago where sensor had to be attached to an extension pole (too heavy at that length) to safely reach difficult places and stay away from heat. Were forced to go magnetic mount. Signals for the 2 tangential directions were then doubtful.

Keeping track of accel placement was also really confusing. Quick disconnect adds a lot of time and expense (you need to glue or fit the base to every point on the route - which in most cases is not practical). Take it from me - does not save time compared to a single sensor (and others as I can see above). If you know what you're doing you don't need 3 directions. Understanding the best direction a fault will show is a much more efficient way to go. Some push this method since auto analysis relies so heavily on the 10x & 100x spectra for each of the mutually perpendicular directions. And if you leave one of these out the auto analysis starts to complain.

You'll have so much data leading to information overload if you ever try to attempt analysing anything manually. Beware of analysers that run windows and have the software installed on top of it, it is not an advantage. You really will not be doing anything on Microsoft Office whilst collecting data on slurry pumps. There is another thread re VSD influences, we had continual windows blue screens (Win2000)which meant back to the office. This was our experience 6 yrs ago and I do not know whether any improvements have been made to that system which shall remain nameless.