measuring Phase matching between channel

[sutheesh]

How to measure phase matching between two channels in multichannel receiver systems??
can i measure it with oscilloscope?

how can i measure phase matching with signal processing(ADC+processor)

 

[BradtheRad]

Are you talking about stereo radio broadcasts? The left and right channel are combined and transmitted simultaneously.

The place where phase differences might occur is after the stereo detection stage, in unmatched lengths of wiring, or capacitive delay, etc.

If you want to observe phase differences with an oscilloscope, use the X-Y method. One signal for vertical, one signal for horizontal. Two matching signals, in phase, create a single diagonal line with the high end at top right onscreen.

This method also makes mismatched signals stand out.

You can tell when a broadcast is stereo if the image looks like rapidly changing drawings of chaotic steel wool pads.

Furthermore if the channels are not in phase, the diagonal line (or chaotic steel wool pads) will have its high end at upper left onscreen.

 

[biff44]

I once used a tektronix MSO4104 scope to measure a phase differencing microwave system prototype. I used the multichannel inputs and downconverted (sharing one LO generator) each microwave channel to lower frequency to two of the inputs. I used the labview software that came with the scope to capture the two IF waveforms, and used a simple FFT process to figure out the phase angle. It worked very well, and was pretty accurate (much better than 1 degree) for phase.

 

[sutheesh]

thanks biff44
i am trying to implement the same with my DSP system? But i have to compare two signals of 150MHz freq, but my ADC clock is only 220MHz. ie can i preserve phase information in 2 channels when i am doing subnyquist sampling?

I once used a tektronix MSO4104 scope to measure a phase differencing microwave system prototype. I used the multichannel inputs and downconverted (sharing one LO generator) each microwave channel to lower frequency to two of the inputs. I used the labview software that came with the scope to capture the two IF waveforms, and used a simple FFT process to figure out the phase angle. It worked very well, and was pretty accurate (much better than 1 degree) for phase.

- - - Updated - - -

thanks for your reply..
i am not talking about stereo broadcasts.
my intention is to measure phase difference between two identical channels of VHF freq range.

- - - Updated - - -

thanks for your reply..
i am not talking about stereo broadcasts.
my intention is to measure phase difference between two identical channels of VHF freq range.

Are you talking about stereo radio broadcasts? The left and right channel are combined and transmitted simultaneously.

The place where phase differences might occur is after the stereo detection stage, in unmatched lengths of wiring, or capacitive delay, etc.

If you want to observe phase differences with an oscilloscope, use the X-Y method. One signal for vertical, one signal for horizontal. Two matching signals, in phase, create a single diagonal line with the high end at top right onscreen.

This method also makes mismatched signals stand out.

You can tell when a broadcast is stereo if the image looks like rapidly changing drawings of chaotic steel wool pads.

Furthermore if the channels are not in phase, the diagonal line (or chaotic steel wool pads) will have its high end at upper left onscreen.

 

[biff44]

Subnyquist sampling, theoretically, would not change the phase measurement. You would have to carefully DSP filter, and choose sample rates, etc, so that you have a clear look at a single tone phase. You would not want one tone's phase with another tones information folded over due to too high a bin width, or spurious artifacts folding back in.

Also, to get accurate phase information, you have to worry about quantization noise, and that might be degraded by undersampling by la poorer ENOB.

Sounds like you need to ask these qestions on a DSP board where they know the details of filtering and sampling.