08-24-2014 04:39 AM
With this procedure you do not measure a phase. You still don't have a reference. The simulated signal is not a reference.
08-24-2014 08:58 AM
In general you cannot make it work the way you describe.
1. The actual power line frequency differs from the nominal 50 Hz by some small amount at any instant and the amount of the difference varies with time.
2. Any measurement you make of the frequency is limited in its resolution by the time you can take for the measurement, the sampling rate or timebase clock in the measurement system, and noise.
3. Any simulation of the frequency is also limited in its resolution.
4. Due to the differences between the frequency of the current signal and the simulated signal, you cannot really define a phase difference. It will change for each cycle of the signals.
As several people have mentioned, the usual technique is the measure both the voltage and current waveforms at the same sampling rate and determine the phase of the current with respect to the voltage. The fundamental frequencies are the same in both cases so the frequency difference and resolution issues are much reduced.
Lynn
08-24-2014 08:39 PM
Dear Mr Lynn
My DAQ does not support simultaneous sampling. So is it necessary to measure current and voltage simultaneously? Mine DAQ is NI PCI 6281
Kind Regards
Irfan
08-24-2014 08:48 PM
Dear Mr ejkaiser
Reagrding your querry about what i want to achieve through phase measurement, I have one research paper in which author has done this. Please see the attachment. Section II,III and Figure 1.
Kind Regards
Irfan
08-24-2014 11:56 PM
Hi Irfan,
I think, already people in the forum explained everything you needed.
Phase measurement is always with respect to some reference it is not an absolute measurement parameter and it is a relative parameter.
Even in that paper the phase measurment is with respect to the Fundametal wavefrom. Fourier Series or FFT talks about the phase difference of the hormonics with respect to the fundamental wave form and it is not the absolute measurement.
Any rotating objects/machine will have fundamental (Motor RPM) and its associated honmonics due to is sub systems (Ex grars, bearings, balls in bearing, rings in bearing etc...).
Hope this helps...
08-25-2014 12:38 AM
Dear Mr Vijay_J
Phase is relative measurement. So my objective is to measure the phase differnce between two signals, both of 50 Hz frequency. Now one signal is being measured in real time from motor line current and second signal is simmulated. The problem occuring is that in every run I am getting different phase differnce value. As explained by others (GerdW,Edgar,SteveD123) I need ti trigger my real time measurement at zero crossing. But issue is how I should trigger measured waveform at zero crossing?
Kind Regards
Irfan
08-25-2014 12:47 AM
Dear Irfan,
These two signals can not be compared and the comparison will result in the observations you have listed at the begining.
What is the need of comparing the 50Hz simulated signal withrespect to the measured signal. If the intension is to measure the phase then the method itslef is not correct.
Please go back to basics.
Regards,
Vijay.
08-25-2014 04:04 AM
Dear Mr Vijay
Actually I am working on fault analysis of motor and I need to measure the fault frequency (amplitude and phase) which appears in motor current due to defect. I read in one research paper where the author use phase measurement method. He generate one reference signal with frequency equal to fault frequency and then he calculate the cross correlation of measured current signal with refernce signal. I was trying to achieve same. I have attach that paper for your reference (page 2544, second colum, heading Lock in for systems with slow dynamics, jus read the first paragraph of this heading)
Kind Regards
Irfan
08-25-2014 04:17 AM
Dear Mr Ejkaiser
Actually I am working on fault analysis of motor and I need to measure the fault frequency (amplitude and phase) which appears in motor current due to defect. I read in one research paper where the author use phase measurement method. He generate one reference signal with frequency equal to fault frequency and then he calculate the cross correlation of measured current signal with refernce signal. I was trying to achieve same. I have attach that paper for your reference (page 2544, second colum, heading Lock in for systems with slow dynamics, jus read the first paragraph of this heading)
Kind Regards
Irfan
08-25-2014 05:11 AM
Hi Irfan,
I had glanced through the page you have referenced.
What was my understanding is that, one way of phase locking the DSP generated reference signal with respect to the Motor Current Fault Signature and that’s all.
By and large - Motor current signal needs to be analyzed in frequency domain (Nominal Amplitude and Phase Response) for its sub system or components fault. Under normal condition measure the reference response and keep it as a reference. Later, compare reference signal with respect to the fault signal. Fault subsystem will reflect in Amplitude as well as Phase change response of the motor signal. Compare the response with respect to the reference to diagnose the fault and sub system.