Correct calibration of my active two-way speakers with REW

[Alpi]

I would like to discuss the correct procedure about calibrating my active speakers with REW.
here are my measurements of the left LS and my double bass array

 

[Alpi]

I cant Attach files

 

[John Mulcahy]

Perhaps the files are too large.

 

[Alpi]

36299 KB - too large?

 

[John Mulcahy]

Yes, limit per file is 20,000 KB. You could save some of the measurements in separate files.

 

[Alpi]

Ok - thank you

 

[Alpi]

So now it must be ready

 

[Alpi]

I have here measurements of my left speaker at my seat. The dimensions of the room and my seat position can be seen in the REW file under P5 in Roomsim.
The measurements were all made with an acoustic time reference except for the woofer alone because the tweeter was muted and could not play a reference signal.
You can see the measurements before and after. I also measured the drivers alone (I have an active LS system).
Attached is the mdat file - what do you say?
The irregularities at 1.2 to 2.1 Khz of the tweeter is a Filter in the DCX i have forgot to delete. I will show you the new measure.

Later on, I would also like to re-correct the other two speakers and talk about my bass system.

I would like to know if it is done correctly or if you have to do something different with the equalization.
I have also created two filters as a template for my rectification.

My source is a HTPC and my measurement computer is an old Sony laptop (2014) with a Tascam US144II with a calibrated ECM 8000 microphone. For measuring I use the ASIO drivers.

My Cinema technology:
Yamaha CX-A 5000, power amplifier MX-A 5000, 2 DCX 2496 for equalizing.
My speakers (Beckersounds B300) are designed as inwall home theater speakers with subwoofer support.
They are fully active and are powered by my Yamaha MX-A5000.

Since I am very interested in REW, I would like to see how to tune the system even better.

I am curious about your assessment.

 

[Alpi]

A specific question regarding my DBA.
I measured the front and rear arrays separately. I think the rear one was inverted during the measurement.

How would you adjust the DBA? In the Alignemt tool I come up with different solutions.

Attached is the measurement.

Thank you

P.S you use the newest Version of REW? I use V5.20.13

 

[John Mulcahy]

Looks good. The tweeter level before EQ looks high. The DBA summed response looks good as measured without any delay changes, the rear delay should correspond to the distance from the front array.

 

[Alpi]

Thanks for your assessment, but what I don't understand is the fact that my DBA without delay, without inverting and without EQ up to 80hz is a straight line with minimal waterfall. This is also the case in the alignment tool - can you explain this?
I show you my measurements.

 

[John Mulcahy]

Symmetric placement of 4 subs is quite effective at suppressing modal resonances when the subs are the same and all playing the same signal without delays. You can see that effect in the room simulator.

 

[ddude003]

If you go down the multi-sub rabbit hole this paper by Todd Welti might be interesting... https://www.harman.com/documents/multsubs_0.pdf
And it looks like your room has a good amount of treatment...
Look at the Overlays Window and Impulse Response to see what is going on with your subs interaction...

I try to keep up with the latest and greatest version of REW... Here: https://www.avnirvana.com/threads/v5-20-14-early-access-build.11154/

 

[Alpi]

Thank you i read and i work now with 5.20.14 - looks nice.
But a Question -
the front and rear arrays do not add up optimally, but the frequency response and waterfall are extremely good. Are the subwoofers not working against each other? Is the bass possibly not as energetic as a result?
Is that what I see in the measurements?
I'll measure the DBA again today or tomorrow, as the last measurement was not the most recent.
I'll measure once without delay, without inversion, without level adjustment at the rear and without EQ and once with delay (room length), inversion, level adjustment and with EQ.
Then you can compare better.

It would be great if you could help me a little to get the best out of it.

 

[ddude003]

The theory is that the rear array is a virtual absorber when delayed precisely and inverted... "In theory there is no difference between theory and practice - in practice there is" - Yogi Berra... And yes, of course, they are working against each other... That's the point.. And because there are a bunch of moving pieces, things may not always add up to an energetic result one may be looking for... Your looking for a synergy where the "The Whole is Greater than the Sum of its Parts" - Aristotle... Also realize your working with your taste, what sounds good to you, not some pretty lines on a chart... As far as I can see you have, and are doing, everything more or less by the book... It all looks pretty good... And yet your still looking for some extra "magic"... Because something does not quite add up for your ears/brain... I get it... So here are a few things to think about...

Is the SPL level the same in the front array as the rear array... Did you measure the front array SPL at the back array position and not the MLP? Is it possible that the back array is too loud?
Have you stepped thru several distance/delay settings (forward/back, constructive/destructive) on the back array to see how that affects the sound...
Did you include a House Curve with your overall system? This is a Home Theatre, not a Mix/Mastering Studio...
How does your overall system sound with only 2 subs as the back array? Is it possible that with your room treatments and the DBA you have too much Bass trapping/absorption?

 

[Alpi]

Good morning,


1) Is the SPL level in the front array the same as in the rear array?

2) Did you measure the SPL of the front array at the position of the rear array and not at the MLP?

3) Is it possible that the rear array is too loud?

4) Have you made different distance/delay settings (forward/backward, constructive/destructive) for the back array to see how this affects the sound...?

5) Have you added a house curve to your overall system?

This is a home theatre, not a mixing/mastering studio....
6) How does your overall system sound with only 2 subwoofers as a back array?
7) Is it possible that you are capturing/absorbing too much bass with your room solutions and DBA?

to 1) Yes the front array is set equally loud

to 2) I actually haven't yet, but I don't understand the sense of this measurement.

to 3) I have already measured many different delays and levels and simulated a lot with Roomsim.

to 4) yes, from a measurement point of view, I still have to listen to different measurements/presets.

to5) yes I did and I know approximately where I want to go

I know - home cinema

to 6) I can't test because I have 4 SUBs (2 serial, 2 parallel) connected to one power amp channel. Maybe a the inuke NU6000 have not enouph Power for my eight LAB12?

To 7) I can't imagine that, in the front the SUBs are built into a baffle wall (1 foot deep) and in the back I have about 2 feet of absorber in the corners but not all the way through (see photo) and the door is covered with 0.65 feet of absorber.

I will take detailed measurements of different DBA configurations and listen to them. I will also post them here.
I have been listening for 2 years now with the "classic" installation (rear array inverted, 14ms delay and rear -1db level). Sounds good but I would like to have a bit more punch (kick?).

Classic DBA - rear array sucks up the front wave and thus cancels out the standing wave.

Multisub - in my case all SUBs play without inverting delay or level difference only the rear ones arrive earlier at my seat than the front ones. In my case it just happens to add up as a straight line at the master listening position and the waterfall is much better than with the classic method - I will show this later. At the moment the weather is nice and I don't get to measure much, but I really want to do the measurements in the next few days!

Greetings from Tirol

 

[ddude003]

Hello Horst,

Thank you for answering above... Sorry I was unable to convey my thoughts regarding number 2... I'll try again and hopefully it might make a little more sense... We both know the object of a DBA is to precisely match the incoming source sound from the front array, sound pressure level, timing and phase inverted, with the back array... My thought is that, because of the law of inverse square, the sound pressure level of the wave from the front array is down several db as it is reaches the back array... Therefore the back array is not precisely matching the sound pressure level and spilling its additional sound into the room, timing would really off at the Main Listening Position...

Now where did I leave my Maxwell's equations pocket calculator...

Greeting from Northwest Missouri... The show me state...

 

[Alpi]

Hello

I already know how a DBA works in principle and during my measurements I usually set the rear array -1 to -2db lower to achieve a better frequency response.
I just have to "coincide" that my DBA looks so good without inverting, delay and level (mode at 33Hz much better) - I just have to listen to it in comparison to my classically calibrated DBA.
I will also send you measurements later, maybe you will see if it fits.

I can measure the front array at the very back of the wall, but I don't know how that helps me - I'm not sitting there. I don't quite understand that yet.

 

[ddude003]

I can measure the front array at the very back of the wall, but I don't know how that helps me - I'm not sitting there. I don't quite understand that yet.

Because you want to Equally match the Sound Pressure Level _as it arrives_ at the back array _not_ your Main Listening Position...
As an example lets say that your SPL 2 feet from your front array is SPL 85db and at your MPL its 80db and at your back array its 76db... The correct level for your back array SPL would be 76db... You can hear the difference in a +4db shelf from 30hz to 60hz can't you... And who knows what that extra +4bd might do to the sound as it arrives at the MLP...

 

[Alpi]

Hello,
it took me a while but I have a new measurement PC and had to get everything working again.
Dell / i5 (9th) / 16GB RAM / Geforce 1060 / Tascam US144 mkII / Mic ECM 8000 kalibriert

Another question which audio driver in REW do you recommend?
Java - Asio4all or Flexasio and why? Advantages and disadvantages and what should I pay special attention to (sampling rate?) - Thanks!

 

[ddude003]

Hi,
That looks like a reasonable system... I do my room sampling with a MacBookPro and a UMK1 USB mic... So I won't say which audio driver is best... I think Flexasio works pretty fine for the other things I do on my windoz PC... I do all my room sampling at 48kHz... Which I believe is the native rate for the UMK1 USB mic... 48kHz gets me a sweep to the 24kHz Nyquist mark for my ML ESL speakers which go up to 22kHz... A longer sweep, 2 or 4 meg, may get you a better take on the noise floor...

 

[Alpi]

Ok thanks,
I will take the left speaker as a time reference (you should take the speaker furthest away, right?) and enter +2 or +3db in "ref level trim". First I will try it with the ASIO4ALL driver. As sample rate I take 48kHz and the length of the measurement sweep 2M.
Measurement level 75db/C
I hope I have some time to measure today

 

[Alpi]

I am measuring but with 48 kHz I have strange sounds / interferences with my Tascam US144mkII.
it can do 44,1khz but ASIO4ALL can convert the signal from 44,1kHz to 48kHz. Unfortunately, I have strange tones there, so now I measure as always with 44.1 kHz.
Or is the Tascam and my ECM 8000 possibly already somewhat outdated?

 

[ddude003]

Good morning Horst... I have no experience with the Tascam... So I won't be much help with that... I took a quick look at their web site and see that its discontinued in the Americas... I also saw that it should be able to do 48kHz... And if you have issues with 48kHz then 44.1 will have to do... I wonder if writing the sweep tones to a .wav file (Generator) and playing them thru your kit might avoid the strange tones issue... That way you are just using the Tascam for mic input... This is something I do to include my entire signal chain in my measurements...

I noticed there are firmware updates and drivers to the Tascam... Are you using the latest?

 

[thothsong]

Try using the Java driver and see if you have the same issue with 48 kHz.