REW time alignment clarification needed

[mamba76]

Hi all. Iv been using REW on and off for a long time now and Im reproaching my cars 4 way active system. I wanted to clear up some grey points starting with time alignment of a pair of drivers. Anything in bold is a quote from REWs own tutorial. I have loopback timing which simplifies taking impulse readings. Its quite easy to align all speakers the way REW reccommends which is bypassing all filters ('PEQs can be left as is') and take measurements using a sweep range 'above which both speakers are flat', I struggled to interpret this but got all the measurements done and adjusted the plugin so all my impulse settings measurements are now aligned. Perfect! Turns out this way all my speakers are within 5 milli seconds of each other. This sounds correct to me as the speed of sound is quite fast ). It also seems easier to select a crossover slope & point. If I activate the filters and take impulse readings again it appears that there is a 10 - 15 ms miss alignment on the impulse graph (peak to peak), should I ignore this and why does this happen?

If there was a magic speaker that could play all frequencies at equal amplitude they would all eminate from a single point (the speaker). In the real world we time align multiple speakers so they APPEAR to eminate from the same point (the mic position). So with filters off they are all aligned. With filters on we are just antenuating frequencies the speaker can't play but why does the impulse graph suggest they are mis aligned? Im guessing its a product of the FFT but am struggling with this. Are they really mis aligned?

 

[John Mulcahy]

Can you link the tutorial you are referring to? Crossover filters will affect the impulse response. For example, low pass filters will broaden and delay the impulse response peak. If you post mdat files with your measurements it will be easier to comment on any alignment or misalignment.

 

[mamba76]

Actually I just realized this tutorial is from MiniDSP -
https://www.minidsp.com/applications/auto-eq-with-rew/measuring-time-delay
The crossovers between these 2 speakers are LR48 @ 200hz
Aligning after the crossovers are set is what I thought and confirms my discovery a couple of years ago that I DID need more delay than one of my dsp's could give. The CDSP6x8 is made for cars but only has 15ms max delay which was not enough so I feed that with a 2x4HD which has 80ms max delay for a total possible 95ms, in reality I used upto 25-30ms in total but that depends where the impulse biggest peeks are. I now try to adjust/select the crossovers so the first major wave is positive and of the highest amplitude (I don't really like inverting the drivers) then do a listening test to fine tune alignment and repeat for any 2 speakers. Im liking the new clarity tab BTW! Iv not finished my new setup yet but crossing my sub at 53hz sounds really impresive and decays much more quickly. Tighter control needs higher gain /level but thats not a problem. What a difference!

 

[mamba76]

If any admin reading this, please feel free do remove this thread as I have solved my original issue and don't think this is of any help to anyone. Thanks!