How to measure phase properly at home using REW?

[Krunok]

When trying to do room EQ measurements with REW you will find a lot of information on the Net. You have option to do RTA measurement with pink noise using moving microphone method. Another option is to make several sweeps at and around your LP and use "Average the response" feature under All SPL tab. If you want to preserve timing information you can rely on a single sweep or use Vector average (although that one tends to result in a weird frequency response, but there you have it).

However, with phase measurements things are getting a little tricky. let's assume for simpicity you have a done a single sweep and that it contains accurate info. You fixed the IR delay and your phase is shown wrapped in REW. But what now?

At the very beginning you will soon realise that it gets much harder to get phase response displayed correctly in REW as your distance between microphone and speaker gets bigger. If you measured from 10cm distance chances are you would not need to apply any gating and/or frequency dependent windowing to show pahse correctly.
If you measure from 0.5m or 2m you will probably need to make some adjustments under IR Window settings to get the phase displayed properly. But if you measure from your LP whivh is say 4 meters from your speakers and reflections in your room are pretty high you may realise that displaying phase correctly presents quite a challenge.

This is how it looks when I do it in my room with my left speaker, measured from app 2m, no gating:



With right window of 3ms (doesn't display correctly with higher value):


With FDW of 3 cycles (doesn't display correctly with higher value):


Here I would like to say that when I measure phase from 10cm no gating nor FDW is required to display phase correctly. As measuring distance increases the reflections are building up so I have to apply more and mroe gating to display phase.

Phase overlayed (3ms vs FDW 3cycl):


Step overlayed (3ms vs FDW 3cycl):


GD overlayed (3ms vs FDW 3cycl):


So, after all of these graphs a few question arises, like for example:

- what is the optimum distance to measure phase in your room? Very close to the speaker, say 10cm, or at your LP?
- what is the optimum way to adjust your measurement? Applaying right window time gating, FDW, or both?
- should we even bother with all this? Does getting the phase flat matters at all?

I would like to add here that, when i was doing manual correction of my room/speakers response I didn't rely much on measuring uncorrected phase response. Once I fixed amplitude response in rePhase I corrected phase of my filters so it would be as close to 0 as possible to stay minimum phase and I entered data about my crossover (LR 24dB/oct at 1800Hz) so the rePhase can take it into account. The graphs above are showing the results of that.

 

[dc655321]

Why would measuring phase be any more or less difficult than measuring amplitude magnitude?
They're both derived from the same source - sampling pressure amplitude.

AFAIU, near-field measurements are only accurate up to a certain frequency (dependent on transducer dimensions/geometry), and "far-field" measurements must be gated to achieve pseudo-anechoic status.

The windowing aspect (FDW) will affect compromises on amplitude/frequency accuracy.

IOW, both must be used.

 

[Krunok]

AFAIU, near-field measurements are only accurate up to a certain frequency (dependent on transducer dimensions/geometry), and "far-field" measurements must be gated to achieve pseudo-anechoic status.

Which frequencies are you saying are not measured correctly when i measure 10cm from a tweeter (on axis)?

 

[dc655321]

Which frequencies are you saying are not measured correctly when i measure 10cm from a tweeter (on axis)?

An oldie, but a goodie - http://www.xlrtechs.com/dbkeele.com/PDF/Keele (1974-04 AES Published) - Nearfield Paper.pdf

 

[Krunok]

An oldie, but a goodie - http://www.xlrtechs.com/dbkeele.com/PDF/Keele (1974-04 AES Published) - Nearfield Paper.pdf

Thank you, but I would more appreciate practical advices intended for rookies (such as myself) than a scientific paper full of formulas.

 

[dc655321]

Thank you, but I would more appreciate practical advices intended for rookies (such as myself) than a scientific paper full of formulas.

At some point in the past didn't you say you were an engineer of some type? ;-)

According to the info in the link above and for an assumed 2.54cm (1inch) tweeter, frequencies greater than approximately 4.5kHz will not be representative of far-field measurements when measured near-field. However, I also don't think that a 10cm measurement distance qualifies as near-field according to that paper...

 

[Krunok]

At some point in the past didn't you say you were an engineer of some type? ;-)

I am. But when you ask me for a practical advice how to setup your home WiFi network to be able to stream HD video I wouldn't send you scientific paper as you would probably have a hard time extracting practical advice to solve your problem from it unless you have some serious background in networking.

According to the info in the link above and for an assumed 2.54cm (1inch) tweeter, frequencies greater than approximately 4.5kHz will not be representative of far-field measurements when measured near-field. However, I also don't think that a 10cm measurement distance qualifies as near-field according to that paper...

Thank you!

Here is how it looks when I overlay measurements made from (10cm beige, 0.5m blue, 1m green and 2m violet) (All are gated to 3ms and FDW is 6 cycles):



Frequency response (non gated):



So, what distance is optimal for measuring phase response at home in echoic environment?
How many FDW cycles should be applied?

 

[KSTR]

@Krunok , you have to time-align the pulse before FFT'ing it, otherwise most of the phase wrap is delay, not phase. This is seen more easily with linear frequency axis, constant delay means a constant overal slope superimposed on the real phase response. This also seen as a group delay offset (doesn't go to zero a high freq). The windowing also doesn't produce the expected results because the center or left margin of the window must be right on the puls edge.

I don't use REW much for acoustic measurments, I prefer HolmImpulse which is very straightforward wrt offset compensation (automatically + manual fine tuning).

Absolute time offset is difficult to dial in, but with a specific assumption it can be done. The assumption is a) that the final HF roll-off is minimum phase which is certainly true for most tweeters, and b) that there is no crossover close to the finak roll-off that could skew the phase up there siginficantly which als is true for almost any speaker.
You roughly pre-align the pulse manually, derive the minimum-phase version of it and then you shift the original pulse around (sub-sample shifting required) until the phase at the roll-off is matching.
Don't know how to do that in REW but I think it should be possible.

Another strategy is to shift the offset until the phase starts to wrap in the other direction at 20kHz.

Yet another approach is to match the phase display more to what is actually percieved, that is, roll it back until you obtain the largest possible range where the phase is close to zero (+-20deg). For example, if you have a two-way with a 3kHz XO you can adjust the display phase until the whole range from 200Hz to 2kHz is flat. This matches both perception and measurement. Within that 2kHz bandwidth the speaker is transient perfect and reproduces a square wave nicely. The graph resembles the group delay plot a bit, the turnover point is at the same frequency. GD plot readout suffers from the linear Y-axis scaling which is used mostly, burying the HF content. The adjust phase plot is better in this regard.

 

[daftcombo]

Some people measure & flatten a frequency-dependant 1/3 octave windowed phase at LP and are very happy with it.

 

[Krunok]

@Krunok , you have to time-align the pulse before FFT'ing it, otherwise most of the phase wrap is delay, not phase.

Sure. I forgot to mention, but all the measurements have been time aligned with REWs estimate IR delay function.

 

[dc655321]

So, what distance is optimal for measuring phase response at home in echoic environment?

You're looking for an easy/simple answer to a complex problem. Not gonna happen
Requires study/investment of time on your part.

A gate of 3ms results in 333Hz resolution. Quite a coarse measurement.
Perhaps a hybrid approach is warranted? Measure near-field with no gate, measure with gate to capture the upper ranges, then marry the two into a single data set. Then perform analyses on the hybrid result.

 

[Krunok]

Here is the graph measured at 0.5 with minimum an excess phase shown:

 

[Krunok]

You're looking for an easy/simple answer to a complex problem. Not gonna happen
Requires study/investment of time on your part.

A gate of 3ms results in 333Hz resolution. Quite a coarse measurement.
Perhaps a hybrid approach is warranted? Measure near-field with no gate, measure with gate to capture the upper ranges, then marry the two into a single data set. Then perform analyses on the hybrid result.

Here is phase measured at 10cm (no gating) vs measured at 4m (LP):

 

[KSTR]

Sure. I forgot to mention, but all the measurements have been time aligned with REWs estimate IR delay function.

The first measurement shown does not look like that, I see a ton of delay contained. Even in a room measurement the phase never covers any more range that some ten or so rotations (and that would require a 4...5-way speaker with steep XO). Normally, we have some educated guess about how large the total phase rotation can possible be for a speaker of given construction and when the automatic alignment process doesn't yield something close things have gone wrong.

 

[Krunok]

The first measurement shown does not like that, I see a ton of delay contained. Even in a room measurement the phase never covers any more range that some ten or so rotations (and that would require a 4...5-way speaker with steep XO). Normally, we have some educated guess about how large the total phase rotation can possible be for a speaker of given construction and when the automatic alignment process doesn't yield something close things have gone wrong.

That is how it gets with my room and my speakers. The bigger the dsistance the more rolls I'm getting. Here is the graph taken at 0.5m, no gating:

 

[Krunok]

Some people measure & flatten a frequency-dependant 1/3 octave windowed phase at LP and are very happy with it.

Here is how it looks, measured at 0.5m:



Step:



GD:



And yes, I'm also quite happy with the sound.

 

[daftcombo]

That is how it gets with my room and my speakers. The bigger the dsistance the more rolls I'm getting. Here is the graph taken at 0.5m, no gating:

View attachment 24733

Try to put a frequency dependant windows, 1/3 octave and post the graph again please.

Click IR Windows, tick "frequency dependant window" and put 1/3 in the last line.

 

[KSTR]

Will go offline now, will try to comment later this evening...

 

[daftcombo]

Here is how it looks, measured at 0.5m:



Step:



GD:



And yes, I'm also quite happy with the sound.

I can't see the picture. I could see the one in the previous post.

 

[Krunok]

Try to put a frequency dependant windows, 1/3 octave and post the graph again please.

Click IR Windows, tick "frequency dependant window" and put 1/3 in the last line.

I just did. Is it showing correctly now?