Speaker placement and subwoofer integration for a small living room using MiniDSP SHD + REW

[xovaan]

Ahh, I see! For my prior tests, I got them pretty close to 80dB with my amp at 47/70, SHD at -17, and REW at -10. So I'm gonna use the amp and these metrics as a starting point and turn the SHD wayyyy down and use it for my tests, yeah?

@xovaan Btw put it at -6 dB not at -3dB, i messed up the math xD.

Noted!


Testing now

 

[abdo123]

Ahh, I see! For my prior tests, I got them pretty close to 80dB with my amp at 47/70, SHD at -17, and REW at -10. So I'm gonna use the amp and these metrics as a starting point and turn the SHD wayyyy down and use it for my tests, yeah?

I can't really follow here, maybe provide some extra context? what's 47/70?

As for the SHD master volume control and the REW level it doesn't matter where you start (as long as the mic doesn't clip during measurements).

but keeping SHD master volume control on -20 or below is a good idea.

your previous tests are SPL matching, not gain matching, because they weren't anechoic. With this method you will gain more subwoofer headroom.

 

[xovaan]

Oh apologies, my SMSL amp has a volume range from 0 to 70 and it's at 47 and staying at 47 forever now haha.

For the subwoofers, they have an unremovable fabric cover and I don't know if I can safely remove it-- the anechoic guide says mic distance should be 0.05x the diameter of the speaker, so 0.6", however I think it's about 1" from fabric to cone. Any idea what to do?

 

[abdo123]

Oh apologies, my SMSL amp has a volume range from 0 to 70 and it's at 47 and staying at 47 forever now haha.

For the subwoofers, they have an unremovable fabric cover and I don't know if I can safely remove it-- the anechoic guide says mic distance should be 0.05x the diameter of the speaker, so 0.6", however I think it's about 1" from fabric to cone. Any idea what to do?

1" is okay. but if you still have some energy left do the speaker measurements on a 1" distance too, for the sake of accuracy.

 

[xovaan]

kk, at 1" distance for speakers and subs, and -6dB applied in the SHD, all four are ~79.8dB-80.1db (fluctuating) -- speakers ran with the 20-300hz pink noise and subs ran with the "sub" 30-80, not sure if that matters for the sub or not?

 

[abdo123]

kk, at 1" distance for speakers and subs, and -6dB applied in the SHD, all four are ~79.8dB-80.1db (fluctuating) -- speakers ran with the 20-300hz pink noise and subs ran with the "sub" 30-80, not sure if that matters for the sub or not?

Unfortunately you can’t do baffle step adjustment outside of sweeps

I know it sucks but you only have to do it once in your life. This is the subwoofer integration end game.

 

[xovaan]

kk, here's the first sweeps for left and right, no baffle step adjustment yet-- are you saying I need to get the baffle step for my subwoofers as well, as I did with the speaker towers?

 

[abdo123]

kk, here's the first sweeps for left and right, no baffle step adjustment yet-- are you saying I need to get the baffle step for my subwoofers as well, as I did with the speaker towers?

View attachment 159771

yes, exactly, two different baffle step adjustment for each. Also why are there 4 measurements? sanity check?

 

[xovaan]

Yeah hahah

So these are the settings for my subwoofer-- is the observation point the distance from cone to mic, or listening position to speaker?

 

[abdo123]

Yeah hahah

View attachment 159772

So these are the settings for my subwoofer-- is the observation point the distance from cone to mic, or listening position to speaker?

just put 9999 (far-field summation) it shouldn't matter as much.

 

[xovaan]

kk, sending these from my PC to my laptop:

Left speaker BDR from calculator (with 9999ft this time):
https://drive.google.com/file/d/1wYR5yoCxd5sE3j0OoebRpo3RmlVo-vdg/view?usp=sharing

Left sub BDR from calculator:

https://drive.google.com/file/d/1YRUVwxP46uksUOKXFZW2X0C5-RLz2c7A/view?usp=sharing

Together with the speakers and subs:

https://drive.google.com/file/d/1pzo34J9nqK9vCpS2uV5Tu0CogSTn9O6S/view?usp=sharing

edit: I'm confused-- am I running the anechoic calculations on the sub + speaker instead of speaker + port this time?

 

[abdo123]

kk, sending these from my PC to my laptop:

Left speaker BDR from calculator (with 9999ft this time):
https://drive.google.com/file/d/1wYR5yoCxd5sE3j0OoebRpo3RmlVo-vdg/view?usp=sharing

Left sub BDR from calculator:

https://drive.google.com/file/d/1YRUVwxP46uksUOKXFZW2X0C5-RLz2c7A/view?usp=sharing

Together with the speakers and subs:

https://drive.google.com/file/d/1pzo34J9nqK9vCpS2uV5Tu0CogSTn9O6S/view?usp=sharing

edit: I'm confused-- am I running the anechoic calculations on the sub + speaker instead of speaker + port this time?

you should compare Sub * Sub.frd to Speaker * speaker frd.

the two should meet at exactly 100Hz, manipulate the gain knob on the subwoofer accordingly.

 

[xovaan]

Ah I see-- so this is Trace Arithmetic Sub * Sub.frd and Speaker * speaker FRD

So I increase gain on the subs until it hits 100 at the A*B crossover points?

 

[abdo123]

Ah I see-- so this is Trace Arithmetic Sub * Sub.frd and Speaker * speaker FRD

View attachment 159784

So I increase gain on the subs until it hits 100 at the A*B crossover points?

Great! know move the gain knob up on the subwoofer, measure, do the multiplication. keep repeating till the A*B (sub) and A*B (speaker) meet at 100Hz.

 

[xovaan]

Like this?

 

[abdo123]

View attachment 159786

Like this?

yep perfect, now take the -6 dB out, and put the routing matrix like this (mono sub-bass).

Give it a listen for a couple of minutes and then wrap it all up with a fresh Dirac calibration. Congrats you just completed your reference quality system!

Feel free to try my own Dirac Target curve

https://drive.google.com/file/d/1MLiD3J_DM2-T7UG6mk3TDhK_DsRhuFvI/view?usp=sharing

 

[xovaan]

Ahhh awesome!! I am so stoked!

Should I do the timing adjustments again prior to the Dirac and listening?

 

[abdo123]

Should I do the timing adjustments again prior to the Dirac and listening?

You mean the trick thing to push Dirac to set the delays for all the output?

Your choice, the phase might have changed a bit (in a good way) since the previous one, but the difference will be very subtle.

 

[xovaan]

Here's the measurements for left and right channels using your matrix after timing alignment. It's crazy how close both channels match!

Gonna give it a listen and then try your Dirac curve

 

[abdo123]

Here's the measurements for left and right channels using your matrix after timing alignment. It's crazy how close both channels match!

Gonna give it a listen and then try your Dirac curve

View attachment 159789

I checked that null and it seems to be the combination of two room modes X and Z! But that's a topic for a different day. I'm heading to bed now lol. Let Dirac do its thing and let me know how it sounds. Also experiment with limiting correction only to 250Hz (your room's transition frequency).