To All
I am setting up Peak Vue parameters in order to trend the bearing condition. Most of our equipment are in the range of 1500 - 3000 rpm, as per the recommendation I have set up collection parametrs as follows :
F max - 120000
HP Filter - 2000 hz,
Avg 1
lor - 1600
window - Hanning
normal mode
vel << acceleration
When I collected data with the above set up most of the readings are marked as bad reading and not stored in the route.
what I should enter in the units field ?
sensor units/acceleration/velocity/displcement
in the time waveform tab do i have to check collect special twf?
what should be the fmax in the twf tab - 120000 or less?
and the samples 1024
Am I missing something which is resulting in this problem?

Ramesh,

It looks like you are mixing your velocity and Peak-vue collection parameters. You definitely can NOT collect a valid Peak-vue reading with those settings. The setups you had will collect rather low resolution data -but- with the units set to: vel<<acceleration you will have an incorrect display and your data wouldn't be very valuable. Following is a suggested setup for Peak Vue for shaft speeds >200 rpm.

Fmax - 1000hz/60000cpm
HP Filter - 1kHz
Avg - 1
LOR - 3200
Window - Hanning
Normal Mode
Acc <-- Acc [!!!!]for the Units Type Code

Try that, I think you'll see better results. At least it will display g's rather than in velocity/ips.

This message has been edited. Last edited by: Dana,

You want your collection to be in sensor units (accelerometer preferred). For machines running at or below 1800 RPM. I would recommend an Fmax and a High Pass Filter of 1000 Hz. The 2000 Hz is OK for the higher speed machines. As the high pass filter increases, the results become more sensitive to sensor mounting issues.

One average is what is recommended. The 'special twf' is NOT recommended.
The 1600 lor is OK.

Ramesh,

Good to hear from Abu Dhabi. Finally a topic that I feel comfortable weighing in on. My partner and I have been immersed in peakvue for six months now. I do not have my notes with me, but it can make a difference if you are using a 2120 or 2130. I spent days on the phone with Knoxville, and unless you are using a velometer, my 2120 was collecting data in a way similar to what you are describing, but the 2130 will come up as bad sensor ( I had a 2120 & he had a 2130 ). Regardless, if you are attempting this project without tech support, be prepared for some frustrating days.

What I can offer is this : on equipment from 1200 rpm to 1800 rpm, we have had success using these basic parameters. For the motor, 1000 hz f-max with 1000 hz hp filter has been very successful. Radial or axial mounting is not as important as getting as close to the bearing as possible. For what the motor is driving, 500 hz f-max with a 1000 hz hp filter works good. Agitator and fan bearings are simple enough, but pumps are ornery. A clear defect does not last long, so we are trying to increase our route monitoring period. Also, getting as close to six o'clock helps.

I have caught a couple of inner races that would have been sure failures. Outer races are not always as important to me, for I can usually make it to shutdown with one. But peakvue has been useful in confirming what I am up against. Here is the snag, as soon as we find the answer to one peakvue question, another one pokes it's head out. I work in a paper mill, and dryer felt roll peakvue spectrums stick out like a sore thumb. But press & wire felt rolls are baffleing us. Maybe it is the f-max/hp filter settings, but we have tried them six ways from Sunday with no luck. Our latest theory is the mass of the housing or maybe the metalurgy.

Take what I say with a grain of salt, as it was stated in one of these postings, "Stick five vibe guys in a room, and you'll get 10 opinions".

One thing I can state for certain, "Robinson, you da man".

Thanks for the replies.
Dana, your suggestion of reducing the min value helped. The bad reading message stopped flashing in my 2120 screen and the point definition unit type was changed to acc <-- acc an error i made due to cutting and pasting. Let me in the peak vue time waveform how many rotations should i view? 10 or 20?

rnl: our machines are either 1500 or 3000 rpm so are you suggesting i have different parameters for peak vue collection for each of the speeds?

Thanks Ralph,Robinson

Ramesh

Hi Ramesh,

For 1500-3000RPM machines the following setup is recommended.
Upper Freq:- 40 orders
Lower Freq:- 0
No: of lines:- 1600
No: of averages:- 1
Averaging Mode:- Normal
Window type :- Hanning
2000Hz HP
Obtain Special Waveform
Waveform frequency:-40 orders
Data unit:-acceleration
Type of parameter: P-P wave

Maybe this will help you!!

Have a nice day!!

Hi,

The settings that I suggested, 1000 f-max/1000 filter & 500 f-max/1000 filter are good for your 1500 rpm equipment. I also have a couple of 3600 rpm pumps, & these settings basically give me a flat line at 3000 rpm. Coincidentally, I just set up one of my 3600 rpm pumps today with 2000 f-max/2000 filter. I did not get a chance to look at it yet, but if I am correctly interpreting what Jim Robinson told us at the seminar, as the rpm increases, so must the f-max & filter. I may need to go higher?

So to answer your question, yes you will need to set up these speeds differently. I have also found that I need to set up certain pieces of equipment differently. The best way to learn is to experiment.