RCM-2009 Reliability Centered Maintenance Managers' Forum - March 23-26 - Daytona Hilton Ocean Walk Village - Daytona Beach Florida Join or Manage Your Profile Posting Boards Machinery Condition Monitoring and Predictive Maintenance Posts About vibration/alignment/balance Feeding RPM of machine through tach or manually.
Join or Manage Your Profile Posting Boards Machinery Condition Monitoring and Predictive Maintenance Posts About vibration/alignment/balance Feeding RPM of machine through tach or manually.|
Go
|
New
|
Find
|
Notify
|
Tools
|
Reply
|
|
Normally I am feeding machine RPM into the analyser manually based on nameplate details but due to grid frequency fluctuation, it does not match with actual speed of the machine. During analysis, it is found that either bearing defect frequency orders or other orders 1X, 2X, 3X don't match exactly.That creates a bit confusion regarding defect frequency. For taking tachometer or trigger sensor readings to be fed to the anlyser, the machine will have to be stopped to put reflective tape? What is best to be done in such cases? Can we approximatye the defect frequencies?
|
|||
|
Akhtar,
Your could take a small strobe tachometer on your survey to get accurate shaft speeds. CSI makes a fancy laser tach for their analyzers that doesn't require a target. Another method that I often use is to get machine speed from the vibrations. I use the Analyze-Mode to measure a low frequency high resolution spectrum and then use the cursor interpolation to get an accurate frequency and harmonics to be sure I have 1xSS. Some judgement is involved in selecting a good measurement point and making sure that the 1xSS vibration is actually 1xSS. After measuring the speed once, then manually enter this value into your Route measurement points. Walt |
||||
|
Walt has good points.
How much grid fluctuation are you talking about? It is my opinion that you should never rely on the predicted bearing fault frequencies to line up dead on. These frequencies are calculated from a brand new bearing with no wear. They are also calculated from the number of balls in the bearing and this can vary slightly from one manufaturer to another so if you are only using bearing numbers and do not know the brand, your bearing faults may not line up. |
||||
|
I once had to collect data from the drives on a a variable speed steel coatings line, so the speed could be different at every collection. I found the best method was to wait for a maintenance shut and fit reflective tape to every motor shaft, and then use a tacho to get the speeds during collection. Still manual entry, but at least the speeds were correct. Obviously it all depends on what type of guarding you have and if you can actually see the rotating shaft. It's important to accurately identify your shaft rpm, but as Vibeguy says, don't worry too much if your bearing frequencies don't match exactly.
|
||||
|
If the speed doesn't change significantly during route data collection, set up each machine for vairiable speed and check the speed with a strobe or tach of some kind. I carry a Monarch tunable strobe because it is faster for me than the CSI SpeedVue Speed Sensor. The SpeedVue is nice because it doesn't require marking the shaft but it compares it's results to the spectrum and picks out running speed and as near as I can tell won't work very well as a tachometer.
If the speed changes significantly during operation a tachometer input and order tracking will help. And, no, your fault frequencies will not always match up exactly but they never will if your speed isn't correct.  Danny |
||||
|
If your machines are mostly "constant speed" (not DC or Variable Frequency Drive) and you know the approximate running speed, then an inexpensive strobe light is your best bet. You "freeze" the shaft, and then manually input the speed into your analyzer before you start your data collection. You should be able to set the machine up in your database as "variable speed" and it should ask for the speed when you start the data collection on the first machine point. Most strobe lights have an optional TTL output that you can feed into the tach input of your meter. When you freeze the rotating shaft, the correct speed will be feed into your meter, and if you are at the running speed input screen (of your meter), it will show the same speed as the strobe light.
The strobe that I use every day is a PK2X Pocket Strobe. It is a good basic strobe, is light enough to carry on a tool belt, has very good battery life, and the simple thumb-wheel makes it very easy to use. But don't drop it too often.... it is not the most rugged strobe you can buy, but is the best for the money, in my opinion. What kind of analyzer/software combination are you using? Regards, Rusty |
||||
|
I am using VIB 1000 of Comtest with Ascent level 3 software. Can I approximate the frequencies if they do not exactly line up with the spectrum . |
||||
|
You should also be able to correct the speed based on 1 x running speed in the analysis portion of the software.
Sorry, but I am not familiar with the Comtest software. Is the VB1000 compatible with their laser tachometer? It comes standard with the VB3000 in the US but I'm not sure about the 1000. Danny |
||||
|
I use Odyssey software. I take a high resolution data point on one point of the motor. As long as the speed is fairly constant, small changes due to load. I have the software extract the speed from the spectrum.
It has proven to be very accurate. The software lets me input an approximate speed and a bandwidth around that. It will pick the highest peak it finds and set that as running speed. If there are other frequencies close to running speed you can narrow the bandwidth. |
||||
|
In RBMWare you hit R and tell it to save it.
Danny |
||||
|
In Ascent you put the cursor on the suspected 1x peak. Hit control P to locate the peak. Then hit control 1. This will open up a box with the value of the cursor in it. You can apply the rpm to the entire machine there.
If you have fault frequencies set up you can use the shift key along with F5 and F6 to adjust the frequencies by small percentages. The strobelight is a good way to go as others have said. You can also use it to find bad coupling inserts, broken fan blades, loose bolts, and cracked frames without shutting the machine down. Just make sure that some curiosity seeker does not come by while you are checking it and try to touch the "stopped" shaft. Believe it or not it has happened before. |
||||
|
I will take 1600 LOR with Fmax around 10,000cpm in the monitoring mode. Then mark the 1x peak. Switch back to route mode and enter the RPM. This works fine but not in all cases.
|
||||
|
I collect a 12000 cpm fmax, 1600 lor, 3 average velocity spectrum on every motor, too.
Danny |
||||
|
| Previous Topic | Next Topic | powered by eve community |
 | Please Wait. Your request is being processed... |
|
|||
| Site | Join or Manage Your Profile | ||
| Servlet | eve1da019 | ||
| Version | 1.3.0 build 9383 | ||
| Module | Forums 4.0.3 | ||
| Stylesheet | "Eve" | ||
| Wordlet Set | "Default Wordlet Set" | ||
|
RCM-2009 Reliability Centered Maintenance Managers' Forum - March 23-26 - Daytona Hilton Ocean Walk Village - Daytona Beach Florida Join or Manage Your Profile Posting Boards Machinery Condition Monitoring and Predictive Maintenance Posts About vibration/alignment/balance Feeding RPM of machine through tach or manually.
Join or Manage Your Profile Posting Boards Machinery Condition Monitoring and Predictive Maintenance Posts About vibration/alignment/balance Feeding RPM of machine through tach or manually.
Contact Us | RCM-2009 Reliability Centered Maintenance Managers' Forum - March 23-26 - Daytona Hilton Ocean Walk Village - Daytona Beach Florida | Privacy Statement | Terms of Service
Copyright © 2004-2008 NetexpressUSA Inc. All rights reserved.