TOPIC:

Hello all,

I am currently trying to follow the tutorial provided by MiniDSP for etting the delay for speakers using the calculation in REW:

www.minidsp.com/applications/auto-eq-wit...eaker-time-alignment

What I struggle now with is the fact that the calculated delay differs when using different sweep length in REW. 

To some extend that sounds logical to me, but then I struggle on how to use this tutorial in the right way.

Looking forward to some useful input here.


Thanks and regards,
Matthias

I suspect you may be experiencing the result of an effect that I reported on the REW forum (see thread below). Essentially the 'answer' was to use the results of a short duration measurement.

www.avnirvana.com/threads/measurement-le...ation-question.4475/

Very interesting, so there is some kind of interference.

Matthias182 wrote:

Very interesting, so there is some kind of interference.

Not really what I'd describe as interference. I'm not sure if you read through the thread I linked to but there it was described that the cause was the digital clocks for the UMIK and 2x4 HD signals being very slightly different. The longer the measurement, the more significan tthe difference becomes, producing the effect I found.

What are you trying to time-align? I'm just asking in case it's a subwoofer and main speakers, because if it is I would do it using by matching phase response rather than just using impuse response data.

I think I wrongly used the word interference (not a native speaker). What I meant is that there is some kind of coherence between the signals coming from the Umik and the MiniDSP.

What I did not understand from reading the topic is, if it is only a problem for the acoustic reference (it seems because we are using the Umik). What happens if we use a USB audio interface and measure the delay using the loopback?

Regarding alignment, I am basically trying to align all my speakers, so Fronts, Center, Surround and Subwoofer following the same method. What is the difference between time and phase alignment and what woudl I do differently?

What miniDSP do you own?

A loop-back via USB where the miniDSP is set as the 'microphone' in REW can't be used to judge acoustic time-delays. It is possible to make measurements in REW that avoid the issue I highlighted but not when using a miniDSP. The way it can be done is to use a single sound-card as both DAC and to recive the signals from the microphone.

There are two reasons that phase matching is better for setting delays to subs that looking for impulse response peaks:

1. in-room subwoofer impulse responses are a bit of a mess, and it's not clear quite what point to align to.
2. What we hear at low bass freqencies is not dominated by the direct signal straight from the sub to our ears, but rather this combined by the multiple reflections we get as the sound bounces around the room. 

What you want at the crossover frequency between main speakers and subwoofer is for the two contributions to add up perfectly. If they don't they cancel each other out to some degree, and give a dip in the frequency response. The responses sum perfectly when the phases for each source match. You can get the delay roughly right from impulse response but if you use the 'Overlays' feature in REW you can compare the phases, with the difference at the crossover frequency being the important one.

I use the MiniDSP 2x4 (no HD).
I am not trying to do any kind of USB loopback or so. Maybe that was wrongly interpreted.

Let us maybe start from the beginning. This is my setup:

Denon AVR-3500x
Front: XTZ 99.25 MK3
Center: XTZ 99.25 MK3
Surround: Quadral Phase 16
Sub: SVS SB2000
MiniDSP 2x4
For measurements: Umik-1

For measurements I connect my PC via HDMI to the AVR to play the sweeps and record using the Umik-1 connected to the PC. As a reference I just use the acoustic reference. I was considering to use a loopback instead, because of the issues reproted here. I was asking if the loopback would actually resolve those problems or if it would be the same as with the acoustic reference.

My aim currently is to best align all speakers in my setup

So you are measuring delays to be applied on your AVR, not the minDSP? I guess your 2x4 is connected to the sub output on your AVR, and then on to your sub?

I don't think you can avoid the delay issue with your setup, with the only way to do so being the single sound-card example I described above. Assuming you can't do the measurements with the UMIK connected to the AVR rather than your computer? However, if you want an answer from a real expert I would suggest you ask you question on the REW forum that I linked to above. 

To be honest, I would rather time align based on the amplitude. I with phase and Impulse before and this was much easier. Play a test tone at the cut off frequency and dial the sub in to have the highest amplitude. You can then verify if the alignment matches, if you like.