Good Morning!

I have just started monitoring in a factory and my first readings of an air handling unit (supply) have turned out very high. Can you agree or disagree that it is belt frequency harmonics causing the high readings?

Here is the velocity spectrum. motor speed is around 2890rpm and the fan is about the same. The overall readings is 22.5 mm/s or 0.88 ips from the Non-drive side of the fan. This seems to high but I dont have anything to trend it on!



Regards,

Below is a higher resolution look at 0 - 100Hz. I think the 2 peaks highlighted are the motor speed and the fan speed. Does this mean the problem is just unbalance between the motor and the fan??

Regards,

Steve to make the right call, you really, really need to know what the correct shaft speeds are. "About" and "around" need to be narrowed down. If you knew that the motor speed is at 46.625 Hz, you can start right away at narrowing it down to a motor speed related vibration problem. Can you strobe the shafts and verify the speeds?

Hi Don,

Thanks for your reply. I cant get near the machine as I am only taking readings off mounted sensors. I could ask the maintenance guy the next time im up there to see if he knows. Someone took the motor rated speed off the tag the last time this machine wasnt running and it was 2890 or 48.1 Hz.

The only other thing I have is a picture in which the pulleys look the same size, but this is all I have.

At the moment I have a laser/infra red tach, but I need to get near the motor and attach reflective tape to the shaft for it to reflect the laser. This isnt feasable in this case.

Thanks again for your help,

Steve,
Can you put your data collector in monitoring mode and check if you have beat frequency? The motor and fan speeds are very close and structure is very flimsy.

Alex,

Im sorry but I dont know how to check, could you walk me through it? Are my settings wrong on the VB??

Thanks for your reply,

Steve,
I’m not sure what brand you have but most collectors allow switching to Manual Analyze mode. In this mode change from Acquiring to Monitoring setting and this will allow seeing vibration in real time. If you’ll observe these peaks fluctuating up and down most possibly you have beating. The fan speed would need to be adjusted by changing sheave. I don’t line this configuration shown on the picture and such a small operating speed separation.

Yes Alex, I can monitor it in real time and take a number of averages or let it run. Wont be until the end of next week or later, but I can do it.

As a general rule is this true for a belt driven system? with the peaks moving up and down as the belts fall in and out of phase with each other?

What is the problem with beat frequency? is it just that it makes the system vibrate more as a whole?

Most of the other air handling units i have seen have the motor separated from the fan a bit, probably for good reason.

Thanks for the info,

The belt appears to be deep in the sheave. I recommend inspecting sheave condition and alignment.

With such a flimsy structure, it seems likely you will always have more vibration than you would like to have.

Yea Cruiser, Do you notice the dust on the drive end of the motor there? they have another handling unit with a lot of black dust over it too. Could be from the belt.

Ill tell them to check the belts anyway because its less than 1x.

Thanks for your input,

Steve,
Some more on this phenomenon:
http://en.wikipedia.org/wiki/Beat_frequency

It does not have anything to do with the belts going in and out of phase but with two pieces of equipment operating with a small speed separation margin and generating vibration. In the time domain you'll see your signal going up and down. When these two vectors are in phase it achieves the max. amplitude and this vibration can be damaging for the components.
I hope I did not confuse you:-)

No Alex, you are shining a light on something. Below is another air handling unit in the same factory exhibiting the same behavior on the fan, but the motor is fine and it is mounted a suitable distance from the fan this time.

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

Velocity graph up to 100Hz, I will have to check for beat frequency next week and possibly just adjust alarm levels. Motor is all in green for this one, but obviously it is mounted differently.

Steve,
On this chart I don't see previously shown high peak at ~50Hz. If you take a set of reading while the beat is low it’ll be very small. These lower sub synchronoues vibrations could be from the belts. Why don't you recommend replacing existing V belts with the timing belts like Gates?

In AHU 4 , definitely, there is beat at the frequency of 1.625 HZ. You can listen to this beat while standing by the machine but other machines in the vicinity should be off. Vibration is too high. Look for belt resonance. For this check the belt natural frequency and see the separation margin with respect to the pulley RPM or belt speed.
The last spectrum is from another AHU -8 and can not be compared with AHU 4 . Both have different operating speeds or ratios.
Regards.
Irshad Akhtar