I have recently taken readings for the first time on printing presses at a local newspaper and on two measurement points out of 104 I am seeing multiple harmonics of turning speed in my Peakvue spectrum and I am seeing impacting spaced at turning speed in the waveform of the Peakvue around 3. Gs. I all the other measurement points, I am only seeing .5 Gs and less in the waveform and a tenth of the magnitude of the harmonics in my Peakvue spectrum. These are fairly new presses. Maybe a year old and all my overall readings, even on these two points, are .02 in/sec and less in my regular measurement points but I am getting .3 in/sec in my peakvue spectrum on these two points with all the energy being harmonics of turning speed. I am on my home computer so I can't post a spectrum or waveform but if anyone wants to see any, I will log on at work and post them. Any feed back on this would be greatly appreciated.
Thanks,
Ronnie Lynn

Ronnie,
I have seen harmonics of runspeed prominent in peakvue readings on AC variable speed motors and equipment powered by them. Is this true of your printing press? The peaks usually show up at or near 4x runspeed or sometimes 2X. I think this is a function of the variable speed drive and is normal. I would like to hear from someone who is knowledgeable of both variable speed drives and vibration to shed more light on this. I have some Siemens motors and drives that show very high peakvue readings with the peaks always at 4X runspeed and it's harmonics. These same points show low levels of vibration on conventional readings. I will attach a spectrum to illustrate and if you could do the same it would be very helpful to all who might assist you.

Brake_roll_motor_spectrum.doc (27 Kb, 105 downloads) Brake roll motor spectrum.doc

Yes, these are variable speed drives. I talked to the maintenance supervisor today and he said his motor on that particular unit was making a noise prior to this vibtation problem. One problem I have here, is they sent me a list of measurement points and the motors were not included. I questioned him about that but the unit was not running today. I will probably go back at some point and take readings on his motor because I believe the vibration is transmitting from the motor to the rolls. I am sorry, I can't post any spectrums right now because my laptop is at work and I didn't slow down long enough today at work to get on the internet. Some days are like that. If I can figure out how to post the spectrum and waveforms, I will and I will also send any follow up info I can on this. Thanks for the reply,
Ronnie Lynn

One more point on this, my spectrum is harmonics of turning speed, not the forth order.

I have attached a spectrum and waveform from my peakvue reading on the press roll side 1 which is the drive side of the roll. I still have not been able to take any readings on motor but I talked to the maintenance supervisor again today and he is going to let me know when I can come back out and take readings on the motor diriving this roll and on a couple of other motors driving rolls that do not show these harmonics in the spectrum or impacting in the waveform. If anyone recognizes this pattern, I would appreciate any feed back I can get.

Thanks,
Ronnie

I'm sorry, my attachment was rejected. I sent an email to maintenanceforums.com and asked how to attach them. As soon as I get a reply, I will attach the specrum and waveform.
Ronnie

I am trying to attach the spectrum and waveform as a pdf file. I hope it works this time. If not, I will wait for the reply from maintenanceforum.com.
Ronnie

Ronnie

Click on reply - then wait as the window opens and the toolbar appears across the top. Type your message and click the paper clip icon (some computer geeks idea of attachment). I will attach a messsage so you can see.

Terry O

Very strange indeed. It would not accept my DOC MS Word file.

I see it did accept one above and I checked the board settings and we allow MS Word Doc files.

I am reporting the problem now and hope to see it fixed right away. Sorry Ronnie.

Terry,
I did all the above on both the word document and the pdf file. Neither one of the attachments would work. Both times I got the message, "cannot accept file, file must be image or text." Does the page accept pdf files?
Ronnie

I am attempting to send the spectrum and waveform again as a word document. Maybe it will work this time.
Ronnie

test

One more try posting spectrum and waveform from peakvue reading in word document.
Ronnie

PressRegister.doc (32 Kb, 161 downloads)

Ronnie,
I have not seen a spectrum, in my limited experience, that looks quite like the one you are showing. I am completely unfamiliar with printing presses so I don't know if any of the patterns are normal for that particular equipment or not. The 1X harmonics generally mean looseness as I am sure you are aware. The mounds of energy may be centered around a bearing fault frequency. Try to identify all the relevant fault frequencies and try to match them up with those mounds of energy. This may be complicated greatly if there are gears involved. I have seen something recently that looked similar to your waveform. It was a motor where the cooling fan was rubbing the tubes for the grease fitting and the grease vent. When it was down I took the fan shroud off the motor and the fan had almost rubbed the tubes through. It was making a very unusual sound as well as giving strange looking readings.

Randle,
Thanks for the reply. There is a gear between the Plate and Blanket roll. In this set, the blanket is being driven by the motor and the plate roll is driven by the gear. Both rolls have 88 teeth so they are turning synchronous. I didn't show it in that spectrum I posted because the gears are on the opposite end from the motor. But I do have the first, second, and third harmonics of the 88th order through out all the rolls on the gear end. But the levels are very low and I expected to see them so they don't concern me. I did physically count the number of teeth on both rolls so I know there are 88 on each. My first thoughts were looseness because of all of the harmonics but the mounds of energy didn't fit any looseness pattern I have seen before. I am still waiting to hear back from the customer as to when I can go back and check motors on that particular set and do some comparison checks on similar motors in the same positions. Any further info I get on this, I will post as a follow up.
Thanks,
Ronnie

I’m not too familiar with the machine you are testing, but I am familiar with PeakVue data. The “mounds” of energy, could they possibly be harmonics of a non-synchronous impacting? Are the peaks around the center of the mound, sidebands? Are any of these peaks non-synchronous? Is the RPM set correct? Can you auto correlate the waveform?

I have seen this before in a ball fault which is being modulated by the cage. The amplitude of the waveform is not too high. However, for a machine running at 450CPM, I would want to trend it.

Red

Red,
The RPM is set right. I strobed the unit before taking readings. I do believe all the peaks are synchronous. I am not familiar with correlation of the wave form. If you could explain this to me, I would gladly do it. I have asked the customer for bearing information and advised him to take an oil sample. I do want to take more readings but am still waiting to hear back from them. At this point they only want to do the route every 3 months but my advise was that this unit warranted more frequent monitoring. Thanks for the reply,
Ronnie

If all the peaks are synchronous, that rules out the bearing fault.

The next time you get an opportunity to collect data on this unit, take several reading of the same type, at different points, closest to this “exception”. This will help you find the point with the highest amplitude (Remember you are interested in the waveform amplitude, PeakVue). Once you find the position with the highest amplitude you’ll know you are close.

If those mounds are synchronous, they appear to be around 20X running speed. So, there appears to be an impacting at 20X which is modulated at running speed. I also would say there is some impacting “Looseness” at 1X.

I think you have to have RBMware 4.5 and higher in order to use the “Auto Correlate” option. This is a tool which takes the waveform, divides the “time waveform” in half, lays one side on top of the other, dropping all random noise out, leaving you with the repeatable impacting, only. It’s a great toll and gives you a great picture of what is going on. Note: you should always collect at least 20 RPMs of data.

Red

Red,
I am still using mastertrend so I guess that rules out correlation. Thanks for the advise.
Ronnie

Ronny,
If I was to see a spectrum looking like this I would have the bearing removed. There is a possibility that the inner race is cracked.I would however like to see the inner race defect freq. matching there harmonics.
Printing presses are more complex than a paper machine dryer roll so care should be taken that the press timing is acurate? Is this a metal decorating press? I have worked on a marques printer years ago and recall cams, split gears clamps on the bottom roller.