Announcement

Collapse
No announcement yet.

Vibration analyzer

Collapse
X
 
  • Filter
  • Time
  • Show
Clear All
new posts

  • Vibration analyzer

    Has anyone made a vibration analyzer? I have been playing around a little bit with dynamic balancing. http://bbs.homeshopmachinist.net/showthread.php?t=51162 I would like to build a vibration analyzer to activate a strobe light. I am thinking about using a piezo element and adruino to activate a strobe light. http://www.arduino.cc/en/Tutorial/Knock. I have never played around with adruino. Does anyone think this might work? In my thread on dynamic balancing alsinaj said to use a zero-crossing detector to fire the strobe. I think that sounds like a good idea. I have not found a zero-crossing detector that I think will work. Can the arduino be programed to act like a zero-crossing detector? This is a link to the strobe light I want to use. http://www.amazon.com/Velleman-MK147.../ref=de_a_smtd

  • #2
    Arduino is overused for simple things. It will add all sorts of timing issues when you could just use analog electronics.

    Comment


    • #3
      If you don't want to use a specialized device like the one indicated in my last message, you can make a zero-crossing detector from any op-amp or comparator. Just use a circuit that places one of the comparator's inputs at the midpoint between the positive peak voltage and the negative peak voltage, and use a bit of hysteresis to prevent nuisance triggerings due to noise on the signal. This is not difficult. Google "zero-crossing detector circuit" and you will get lots of information.

      Comment


      • #4
        Getting it right

        There are some serious maths in "Vibration Analysis" - as there are in "dynamic balancing".

        http://en.wikipedia.org/wiki/Vibration

        http://en.wikipedia.org/wiki/Balancing_machine

        http://www.google.com.au/#hl=en&cp=1...w=1280&bih=542

        There is considerable training required as interpretationpretaion can both be objective and subjective (ie "gut feeling").

        Any predictions can be "embarrassing" if got wrong.

        You may as well go the whole way and get into oil/sump analysis as well.

        Comment


        • #5
          Macona,
          can you explain how it can be done with analog electronics?

          alsinaj,
          could you tell me which one of the recommended componets in the zero-crossing link you provided would work best? I would think the ALD2312 would be the best. I found these http://www.findchips.com/avail/?part...=Link#A2187335. Would I use an accelerometer or a piezo element to drive it?

          oldtiffie,
          This is something that I am curios about and want to learn something about. If I do something wrong hopefully I can learn something from it. What would oil/sump analysis do for me?

          Comment


          • #6
            In suspense

            ces224

            I have no objection to learning but is my opinion is that its best to understand the principles and required outcomes so that a machine can be designed and built and to operate correctly.

            Vibration analysis as well as dynamic balancing seek out any defects there as well as optimising the machine/s in those area.

            Oil/sump analysis is the testing and identification of lubricating oils and sump residues for impurities, foreign matter and metal particls (precipitated and suspended).

            All three are tools for establishing the current state of a machine to be compared to its correct limits/parameters as a tool to assist in determining wear and to decide or predict urgent or planned periodic maintenance.

            http://en.wikipedia.org/wiki/Oil_analysis

            http://www.google.com.au/#hl=en&cp=2...w=1280&bih=542

            Comment


            • #7
              Pickup and software

              The typical pickup used is piezoelectric. I used an accelerometer on a Rion vibration meter and put the analog output into the microphone jack on my laptop computer. I downloaded and used a shareware program called "Spectrogram" which diplayed amplitude and would also provide a cascade display that showed the frequencies and amplitudes present in the signal.

              I used Spectrogram to check amplitudes and freqencies to find resonances and peak vibrations in gas powered string trimmers. You should be able to use a piezoelectric contact microphone and plug it directly into the computer to do the vibration studies. The nice thing is that you can see shaft rotation frequencies, but also the bal rotation frequency in ball bearings, ball carrier rotation and in fans you can even see the blade tip frequency.

              Comment


              • #8
                First, use the right sensor for the job. Something along these lines:

                http://www.ebay.com/itm/BRUEL-KJAER-...#ht_500wt_1311

                Then you will need an amplifier, I have one specifically intended for accelerometers, but you can make one with an op-amp and a few extra components. Then you can use the software that Jim mentioned.

                There is one here:

                http://www.ebay.com/itm/Endevco-Char...#ht_500wt_1078

                Most commercial accelerometers use a connector called a microdot connector. It is similar to a SMA connector except about half the size with 10-32 threads on the barrel. You can buy cables on line as well. I don't see any microdot to microdot cables right now, but there are some cables with single ends:

                http://www.ebay.com/itm/Connector-Mi...ht_1145wt_1063

                The mems accelerometers you see in phones and game controllers are just not going to cut it.

                For the rest of this you will have to get someone else to do it. I am terrible with analog. I buy finished products for things like that.

                You could just dump the output to a scope and look at that. If it is rotational vibration you could use a hall effect or photo reflective sensor to create an index pulse on a second trace or use it to trigger the sweep.

                Comment


                • #9
                  Thanks.

                  If I could just hook up already made components and not break the bank is what I would really like to do. I was not sure which components that I needed to make anything work.

                  I had no thought of hooking something up to a computer. That sounds like a good idea.

                  Comment


                  • #10
                    In Nick's post on the Balancing thread, there is a link to a home built balancer. The builder used inexpensive piezo buzzers as transducers.

                    bob

                    Comment


                    • #11
                      Link to "Balancing thread"

                      Thanks Bob.

                      Do you have a link to "Nick's post on the Balancing thread" that you mentioned in your previous post please?

                      I'd appreciate having a look at it.

                      Comment


                      • #12
                        Stethescope

                        Originally posted by ces224
                        Has anyone made a vibration analyzer? I have been playing around a little bit with dynamic balancing. http://bbs.homeshopmachinist.net/showthread.php?t=51162 I would like to build a vibration analyzer to activate a strobe light. I am thinking about using a piezo element and adruino to activate a strobe light. http://www.arduino.cc/en/Tutorial/Knock. I have never played around with adruino. Does anyone think this might work? In my thread on dynamic balancing alsinaj said to use a zero-crossing detector to fire the strobe. I think that sounds like a good idea. I have not found a zero-crossing detector that I think will work. Can the arduino be programed to act like a zero-crossing detector? This is a link to the strobe light I want to use. http://www.amazon.com/Velleman-MK147.../ref=de_a_smtd
                        Ces224.

                        As an interim measure pending your vibration analyser, have you got or have you considered a "(car) engine stethescope"? With a good operator (and good hearing) and knowing what each noise is or should sound like you can do quite well.

                        http://www.google.com.au/#sclient=ps...w=1920&bih=813

                        (Edit open)
                        Added:
                        http://www.amazon.com/Mechanics-Stet.../dp/B000RB5WQY
                        (Edit closed)

                        A screwdriver doesn't do a bad job at a pinch either.

                        I hope it helps.

                        Best of luck.
                        Last edited by oldtiffie; 12-02-2011, 05:44 PM.

                        Comment


                        • #13
                          Oldtiffie,

                          I have a stethiscope like the one in your second post. I have used it on bearings to see what noise they were making. I never thought of using it for vibrations. Maybe I will have to try that.
                          http://www.turbinenmuseum.de/Gasturb...cing_tool.html
                          This is a link to Nicks balancer.

                          Comment


                          • #14
                            Resonance

                            Thanks Cess224.

                            Not all vibration is due "rough" bearings and gears etc. Some may be caused by various "outs of balance" of various components but it will be at the rotation frequency.

                            It is also possible that having the speed too close to the composite or component/resonant frequency - or its harmonics - will add to a problem just as will excessive damping (oil viscosity) and within the components, bearings and other stuff etc.

                            http://en.wikipedia.org/wiki/File:Resonance.PNG

                            http://en.wikipedia.org/wiki/Resonant_frequency

                            http://www.google.com.au/#hl=en&cp=5...w=1920&bih=813

                            http://en.wikipedia.org/wiki/Damping

                            http://www.google.com.au/#hl=en&cp=6...w=1920&bih=813

                            This is not normally a problem unless a rotor is over-sped either under test of perhaps during a "(wheel/rotor) balancing" test. From previous posts the "balancing" is an acceptable compromise of performance and "vibration" at a particualr frequency.

                            In some high-speed/revving bodies/rotors such as steam and gas turbines and naval/navigation gyroscopes it is essential to accelerate the rotor at a high acceleration so that it passes quickly through the lower dynamic frequencies as the rotor approaches and settles at the required operating speed. You can see, hear and feel it passing through - both starting up and slowing down.

                            Accelerating a ball race with a high pressure air supply can show similar symptoms.

                            So vibration and "balancing" are two components of mechanical investigation, fault-finding and setting-to-work.

                            Most of this stuff is very math intensive but in reality experience and intuition, common sense and a fair amount of logic and luck will see you right.

                            I've long forgotten my higher maths but I've seen very good results from operators who were patient and had very little to fair math skills and who developed their skills over time.

                            Just about all of this is obvious in the "real" and everyday working world.

                            Thanks for posting the link to Nick's balancer - its a good read. Read it closely as he makes some very good pertinent points that should not be over-looked, disregarded or unappreciated.

                            I hope it helps.
                            Last edited by oldtiffie; 12-03-2011, 03:11 AM.

                            Comment


                            • #15
                              Thanks for the replay. I am learning a lot.

                              I work on combines for a living and deal with all kinds of moving parts. Most of the moving parts have been balanced. I do not think they have been balanced to a high degree. When the rotors are worn enough to have the rasp bars (the parts that rub on the grain) replaced I will remove the rotors from the combine. The rotor is placed on two stands and then the worn rasp bars are removed and new ones installed. The rotor is then statically balanced. The rotors are 17" in diameter and 7' long. The rotor is supported by a bearing on each end. The bearings have the seals and all the grease removed from them. A light oil is applied to the bearing so that it turns as free as possible. Weights are then welded to the light side of the rotor until it is in a static balance. I built the setep up I described in the dynamic balance thread http://bbs.homeshopmachinist.net/sho...ynamic+balance so that I can spin the rotor to see which end I need to add the weights to. The end that moves up and down the most is where I will add weights to get it to static balance.

                              Once in a while I will get a combine with a vibration that is hard to locate. My normal process is to change the speed of one item at a time threw its speed range to see if the vibration gets better or worse. If I can not locate the problem this way I will remove a belt from the chopper and see if the vibration goes away. The chopper is a drum with 56 swinging knives attached to it. I can usually get the problem resolved and back to the condition it was when it came from the factory.

                              Sometimes I will get a problem where I do not think the factory balanced correctly. This is where the vibration analysis and dynamic balance come in. I can remove parts and static balance them but it would be easier if I could just balance them on the combine.

                              Comment

                              Working...
                              X