Rotary Balancing

[Midlothian]

Thanks I will try that now.

The phase and magnitude extraction from order tracking could be a better place to start from for single-plane balancing.

Converting the vector to complex number format gives a way of calculating the influence coefficient in only a few lines.

I would also like to use phase and magnitude data relative to the trigger for pump and motor pressure traces in the near future.

This was the main reason for investing in the full Suite and also the Order Tracking tools to work with DAQmx hardware. 

Thanks again.

[MickeyD]

Hi Midlothian,

I don't think a low pass filter would solve the issue as the order track is really only looking at the orders you define so there could be lots of high frequency content and it would not add to that phase angle.

I was curious if you had tried using the order tracking with a wrapped phase output, or if it was set to unwrapped. I think it will be difficult to help you without seeing a little data and knowing what VIs you are using. There should be a choice to wrap the phase in an example.

I think we can help you solve this using that order tracking example. Can you tell me explicitly which example you are looking at and what you have configured on its front panel.

Thanks

[MickeyD]

Hey Midlothian,

For this application you are trying to make and the example program you were working with the 9233 will not present any peculiarities. Some examples that use specific rate filtering algorithms are designed now with the rates of the 9234 module, which is based on 51.2 kHz not 50 kHz. Going forward the 9234 provides improved functionality(AC or DC coupling) and performance (lower noise floor) so I would recommend you get that module in any future purchases. For this problem though the 9233 is not causing the issues with DAQmx or DAQmxbase.

Let me know about your VI settings like I mentioned earlier in the thread and we will get this workign for you to do single plane balancing.

[Midlothian]

The VI I was using with best results was Order Tracking Magnitude and Phase (Analog Tacho, DAQmx)

I used a 5V rising slope tacho in the channel 0 of the 9233 and I got the rpm to report a smooth acceleration to a steady rotational speed.

I used an accelerometer in channel 1 of the 9233 with IEPE on and AC coupling (these options are hardwired in any case).

There is no selectable option but I dont see that the 50 kHz sample rate was an issue this time.

I have built a desktop model motor for development of the techniques and I am now getting some better results for phase angle.

I will need to read more about the unwrap phase option - it does not seem to change the phase result that gets plotted in this example.

Thank you for the help.

[mlr2@njit.edu]

I do have same setup but I use USB 4431 but does not working well at all, The USB 4431 read the the accelerometers without any problem but the tachometer does not gave the right reading 

[MickeyD]

To understand is the tachometer not giving the right readings? Are you not seeing the actual pulses of the tachometer. Or is the speed profile incorrect. Are you seeing an incorrect number of RPMs.

It would be good to know how many pulses per revolution your tachometer is and what is the nature of the tachometer signal. Is it a 5 V pulse, a 1 V pulse? Anything you can tell us about the signal would be helpful.

Thanks,

Michael

[RL]

I hope that the resurrection of this thread will produce options for us.  Currently I have resorted to a balancing program that supplies the position of the out of balance and the magnitude.  I have been ignoring the recommended corrections mostly because I remove material.  I do not have a Dremel that removes calibrated amounts of steel.  There are limits to the system though.  Lower than .008 inches per second there is a lot of wandering around the centre but since the Impeller only turns at 14,000 rpm it has worked out ok.

Raymond

[MickeyD]

Hi Raymond,

I would like to make sure you are not left waiting for a response. Looking back through the thread it seems like you want to do balancing with a DAQmx device.

You mention that you have resorted to a blancing program. But that it has limitations. And that you don't use its recommendations, because you are not removing a calibrated amount material. 

Can you tell me more about the balancing program, Is is LabVIEW based?

Regarding the  0.008 inches per second. What displacement do you want to measure down to and what hardware are you currently using to take this measurement. (Or what hardware do you want to use)

What rpm do you expect to get up to in future applications?

Regarding your comment about producing options. The forum is not necessarily the place to produce anything, but if you describe your current road block we can at least make sure its either worked around, or documented as a shortcoming.

Michael

[RL]

Originally I had an uptodate Control Suite but this week I found out that the cost of re-instating that is prohibitive therefore I have expanded the search field.  A company, in Texas, did contact me, years ago,  regarding their Labview based balancing software but they withdrew the offer when the engineer supporting it quit.  I then purchased Microbalance measurement system which is working to under .012 "/sec which I think is pretty good.  The cursor does wander quite a bit at less than .010"/sec which does slow things down in the final stages.  The thread was started because Labview would have allowed me to include more data,  better establish patterns, and improve the filtering. 

Raymond

[MickeyD]

Hi Raymond,

I am glad your system works pretty well. If you have a question about something specific in LabVIEW, or a DAQmx device I will be glad to help as best I can with that. There are several alliance partners that offer balancing applications if you are looking for an out of the box solution as well. I agree LabVIEW would have allowed you to include more data, and build your own custom filtering and pattern collection. Is there something in LabVIEW that you have not been able to get working?

Michael