We are creating a new vibration analysis system with single and dual plane balancing for our production testing of fans. We are using NI hardware and working on developing the program. The analysis is working fine but I have questions on the balancing portion.
When single plane balancing we achieve accurate correction weight and position based on the original amplitutde/phase (O) and the original + trial (O + T). When I try to go to the next step (trim)is where I am having problems. Can anyone provide me with the trim math? Currently I am using the new amplitude/phase from the first correction weight with the (O) to calculate my trim.
Weight placement is against rotation. Phase is measured with a monarch remote optical sensor (ROS) and reflective tape. My amplitude is in velocity and we use CPM for freq.
Any comments or direction would be greatly appreciated.
Use the calculated T vector with the current vibration, assuming you keep the same instrumentation setup.
No offense intended, but why? Did you go out and buy a NI setup for this or did you already have it and decide to use it for this?
There is plenty of good balancing equipment and balance programs already on the market..
Let's see. Two plane trim,
1) leave first and second plane trial weights
2) leave first plane trial weight and remove second plane trial
3) remove first plane trial weight and leave second plane trial weight
as well as the usual
4) remove both trial weights
5) do trim after balance.
You need to get a book, read some papers, or hire a consultant. Is this just for the production use, or will you try to sell it?
It would probably be less expensive to buy an existing product than to develop and maintain a custom solution.
On the other hand, it can all be done with graph paper and/or a calculator.
Thanks for the trim info. Mr. Foiles. Single plane is working very nicely. Today the 2 plane was put into the program, I will be debugging it Tues. and Weds.
As far as why NI vs. buying a off-the-shelve analyzer/balancer? Very good question. The following is a little history, I will try to make it short.
We have been production testing assembled fans for over 20 years. We started out with IRD 880's for the analysis and IRD 245's for balancing. I would imagine only a few people on this board would remember these. Advantages = wonderful instruments, rock solid, reliable and accurate. Disadvantages = single channel, slow, analysis is done once you print out the strip chart (important factor).
Now vibration test a belt drive fan (20 years ago) and lets say we take a total of 12 velocity readings. You printed out the 11th reading (things are great) then after you print your last measurment look at the strip chart (#@%$^&&*#&) vibration is out of spec. Make changes to the fan/motor what ever and start over.
Now you have finished the vibration test (20 years ago) but your strip chart is magnetic style paper. You cannot just file this away because it disappears with age. We used to cut the strip charts down and make a photo copy so it can be micro-ed and kept in a file. Pretty high tech 20 years ago but it worked.
With the increase in production (15 years ago) we just spent too much time waiting to print out the strip charts, cutting the charts/making copies, etc. all non-valued added steps. We looked into all the off-the-self analyzer/balancers but they where geared towards field work and not a production setting.
We ended up (15 years ago) working with a software company and developed our own analyzer/balance system using NI/Labview. This new system (from 15 years ago)uses two tri-axial accels. so we view a 6 vibration measurements on acomputer screen, which is continually updating. If things are out of spec. we can make adjustments live and see the reaction on both accels. If vibes. are good we can freeze the computer screen, type in some standard fan information, test setup etc. and save an image electronically and also the data that generated the FFT.
This new system from 15 years ago, made improvments in test times of over 193%, compared to the IRD equipment. Again not that IRD was bad it just did not fit our production needs. The only disadvantage of this NI system was the vendor had put some custom circuitry on the NI boards. That meant if something went bad on a board we could not buy a off-the-shelve NI boards to fix the system.
2008 Time to upgrade analysis/balancing. Use off-the-shelve NI hardware, faster computers, faster NI boards, larger monitors (I am getting older and my eyes are not as good) up-to-date operating systems, SAP with bar coding with the ability to scan fan info. into analyzer database(important factor), I could list a lot more.
If I had this system in place last year with the bar coding alone I could have saved our vib. techs. from typing in over 80,000 fields of information.
We also have plants in other countries that require vib. systems so once it is debugged in the states these other plants will get this same system.
This project has been very interesting and fun to develope and is now in the testing stage prior to putting into our plants. If time permits in the next couple of days I will post some pictures of my testing/debugging.
(Side note: we also used a IRD 885, 15-20 years ago for field work. Has anyone ever carried an old IRD on an airplane? These old analyzers weighed somewhere around 50 pounds. That was portable in its day)
I too have opted for NI and Labview for my vibration test stand. Our application is not unlike your own. I also chose tri ax accels and a six channel front end. I wrote my own vi to perform the vibration analysis and now have users within my own company interested in balancing with it. I am also the field service manager for my firm, so I haven't had the time to dedicate to that end. Please contact me about your system. I would also be more than happy to share with you what we have done.
Those old IRD suitcases were great. I still have an 880, and it still works! Sometimes, the "flashing phase mark" is more beneficial than opticals or lasers, at least for me.
It's also the reason one of my shoulders is a couple inches lower than the other one!
I have an old 880 and its little brother the 820. I used to carry the 880 down tight vertical hatch spaces on Nuke sub's, tell me about the weight of them things, nevertheless it's still a great bit of kit for diagnostic's and insitue balancing.
|Powered by Social Strata|