REW MMM measurements

[JD1]

Hello everyone I had another thread about the subwoofer however now that I got the KEF R5 Meta I'm finally ready to EQ and then do sweeps to take care of phase.

SETUP:
ETHERNET/TV > Wiim Ultra connected to AMP (Audiophonics AP300-S250NC) which is connected to KEF R5 Meta and sub out which is connected to arendal 1961 1s.

Room is 4.25 meter wide by 5.47 meter long with 2.56 meter ceiling.

- MMM using pink periodic noise using -12 DBFS at 64k sequence lenght (I am using 96kHz sample rate) and asio driver.
- L+R level balanced by making the wiim ultra volume balance +0.12 to the right speaker.
- Sub unfortunately has to be at +12 DB in sub out page (which I'm wondering if this is ok?) since with a volume limit of 95% (in wiim ultra) I dont really go above 30 volume for listening). check sub at 0db picture.
- Test was done using 40 volume which spl meter hints at 74/5 I calibrated SPL using c weighted averaging (S) and I have an external SPL calibrator.
- Audio interface is Focusrite scarlett gen 3 using a beyerdynamic MM1 (90 degree cal file).
- I did measure left plus right plus sub individually.

This was first EQ, some advice or opinions would be highly appreciated and especially an answer to the +12 in sub out page if this is ok since I dont use high volume or if it would be better to lower/raise gain in SUB (arendal 1961 1s), I will do an average sweep using 9 points asap and post it, aswell as single sweeps using 0 degree cal file of each speaker/sub individually.

Also soldered a loopback cable which I use for timing reference in loopback.

Best regards and thanks in advance

EDIT: the +6 db in eq filters i toned them down to +3 instead.

 

[JD1]

The 110 Hz dip is where REW room sim says it happens which does match nicely, however it is quite wide ;<

 

[boxerfan88]

If physically possible to locate the subwoofer in the middle of the front wall, try it at that location and check if the bass evens out a little and reduces the range of the dip (before EQ).

If testing at the middle is possible, try with driver cone facing listening position vs. facing front wall, check and see if it makes any difference. (The experience with my setup was that facing the wall was marginally “more even” in my measurement)

If you have dual driver push-pull subwoofer, you can also try side-firing orientation.

Basically by trial & error find a location & orientation that gives the most even bass with least amount of dips. Then EQ to tame the peak resonances.

Of course measure the results and post your observations/results.

 

[JD1]

The sub is facing side wall (room.jpg) which from measurements seemed most stable, all locations are pretty crappy tbh in the middle of front wall is impossible. It has a sideboard with the tv.

 

[Eleo]

The sub is facing side wall (room.jpg) which from measurements seemed most stable, all locations are pretty crappy tbh in the middle of front wall is impossible. It has a sideboard with the tv.

If possibile, try the middle of side wall. In my case this worked

 

[JD1]

Hello again, here is sweeps, single point, 1 sweep of ALL SPEAKERS together and "Sweeps 1M" is sweeps of each speaker alone, also attached MMM of EQ1 and EQ2. (eq2 fixes the bump near 400hz).

If anyone could check the speaker timing and ALL waterfall and give me some info since I'm no expert at reading those and especially about their phase I would appreciate it.

Thanks alot.

 

[JD1]

I think sub was 7 ms too early so I reduced main speakers delay from 16 to 9 ms will do 3 more single point sweeps tomorrow.

 

[OCA]

I looked at Sweeps 1M.zip measurements, Assuming they are pure measurements (with all filters off), it seems you need to crossover the speakers with the sub at 120Hz, apply a 17.86ms delay to the sub (it should fire earlier) and boost sub volume by at least another dB before you EQ the summation of sub + speaker. For PEQ, you can comfortably use "None" or "1/48" smoothing up to 120Hz and maybe go up to 200-250Hz from there with "Var" smoothing. I see an odd rise around 2-3Khz region with the speakers which I don't think is typical with a KEF speaker, they might be close to some strange surface they don't like. Not the most reliable thoughts based on draft measurements but could give you an idea.

 

[JD1]

I looked at Sweeps 1M.zip measurements, Assuming they are pure measurements (with all filters off), it seems you need to crossover the speakers with the sub at 120Hz, apply a 17.86ms delay to the sub (it should fire earlier) and boost sub volume by at least another dB before you EQ the summation of sub + speaker. For PEQ, you can comfortably use "None" or "1/48" smoothing up to 120Hz and maybe go up to 200-250Hz from there with "Var" smoothing. I see an odd rise around 2-3Khz region with the speakers which I don't think is typical with a KEF speaker, they might be close to some strange surface they don't like. Not the most reliable thoughts based on draft measurements but could give you an idea.

View attachment 472774View attachment 472775

Thank you so much for the input, today before reading this I did change the crossover of subout/mains to 115 hz since I thought it could work better, also faced the subwoofer to LP instead of it side firing to wall, the new eq does seem to bring more clarity into the sound (less boomy and more tamed mids) but I didn't compare it thoroughly, the bass does seem more tamed, I usually use var when I click the EQ button/window guess I should use 1/48 till 120hz and then the rest with var but I did do EQ this time from 16 to 500 hz (by telling rew that range, aiming at 76 db target and I did change the house curve variables; again using var smoothing).

I would love to learn how to time align from what I understood it should'nt have EQ or crossovers? I've been doing it with xover/eq and setting in wiim (output bass from main speakers) and making the sweep from 60 to 150hz. also I'm applying EQ to everything not per speaker. although I did put 2 of the new eq bands on subwoofer (the first 2) since wiim only has 10 peq bands .

If you tell me how I should really make the sweep for you to help me out a bit more with the TA and phase i'd do it. (from what I've been getting with sweeps 15ms main speaker delayed so sub and main speakers reach mic at same time). which doesnt seem far from the 17.86 ms i can only adjust by 1ms +/- due to wiim ultra.

I'm thankful for the help already given!

The new eq and crossover does seem better and I'm kinda glad I reached this conclusion before reading you're post, which does confirm I was not that wrong with my approach today.

 

[JD1]

Although I also dont understand the bump from 10k to 15k guess I could move the main speakers around a bit. the ports are plugged since they are 15.5 cm away from backwall.

 

[OCA]

Thank you so much for the input, today before reading this I did change the crossover of subout/mains to 115 hz since I thought it could work better, also faced the subwoofer to LP instead of it side firing to wall, the new eq does seem to bring more clarity into the sound (less boomy and more tamed mids) but I didn't compare it thoroughly, the bass does seem more tamed, I usually use var when I click the EQ button/window guess I should use 1/48 till 120hz and then the rest with var but I did do EQ this time from 16 to 500 hz (by telling rew that range, aiming at 76 db target and I did change the house curve variables; again using var smoothing).

I would love to learn how to time align from what I understood it should'nt have EQ or crossovers? I've been doing it with xover/eq and setting in wiim (output bass from main speakers) and making the sweep from 60 to 150hz. also I'm applying EQ to everything not per speaker. although I did put 2 of the new eq bands on subwoofer (the first 2) since wiim only has 10 peq bands .

If you tell me how I should really make the sweep for you to help me out a bit more with the TA and phase i'd do it. (from what I've been getting with sweeps 15ms main speaker delayed so sub and main speakers reach mic at same time). which doesnt seem far from the 17.86 ms i can only adjust by 1ms +/- due to wiim ultra.

I'm thankful for the help already given!

The new eq and crossover does seem better and I'm kinda glad I reached this conclusion before reading you're post, which does confirm I was not that wrong with my approach today.

One descent xover method is to apply Butterworth 2nd order hpf to the speaker and Linkwitz 4th order lpf to the sub at the xo frequency. Most speakers are considered have natural BW2 roll off and when this adds up with your applied BW2, it's equal to LR4 of the sub and will sum flat. The exact xo frequency is dependent on the target curve used but yours seem around 120hz. So you apply a 120hz BU2 hpf to each speaker, vector average these two. Apply 120hz lpf LR4 to sub and use alignment tool to align this sub response with the lpf to vector average of speaker (use impulse alignment at 120hz). Rew will give you exact timing and inversion requirement of the sub. Then use vector sums of aligned sub + speaker (with lpf and hpf applied) for eq filtering.

 

[OCA]

Although I also dont understand the bump from 10k to 15k guess I could move the main speakers around a bit.

The very high end response is often toe in or mic tip position problem.

 

[JD1]

Unfortunately I think wiim ultra only uses LR4 for crossovers I know the sub uses LR4 (arendal 1961 1s) if wiim adds a way to change crossover types which I think has been requested then I could use LPF from sub and hpf BU2 to speakers.

Wiim also has individual speaker eq but I haven't tried it. it has stereo mode or L/R but I'm not sure how the sub works with the L/R eq mode.

 

[OCA]

Unfortunately I think wiim ultra only uses LR4 for crossovers I know the sub uses LR4 (arendal 1961 1s) if wiim adds a way to change crossover types which I think has been requested then I could use LPF from sub and hpf BU2 to speakers.

Wiim also has individual speaker eq but I haven't tried it. it has stereo mode or L/R but I'm not sure how the sub works with the L/R eq mode.

Sorry no experience with Wiim but I'd be surprised if crossover doesn't have BU2.

 

[JD1]

Sorry no experience with Wiim but I'd be surprised if crossover doesn't have BU2.

they dont let you change crossover except hzs from 30 to 250

 

[JD1]

I looked at Sweeps 1M.zip measurements, Assuming they are pure measurements (with all filters off), it seems you need to crossover the speakers with the sub at 120Hz, apply a 17.86ms delay to the sub (it should fire earlier) and boost sub volume by at least another dB before you EQ the summation of sub + speaker. For PEQ, you can comfortably use "None" or "1/48" smoothing up to 120Hz and maybe go up to 200-250Hz from there with "Var" smoothing. I see an odd rise around 2-3Khz region with the speakers which I don't think is typical with a KEF speaker, they might be close to some strange surface they don't like. Not the most reliable thoughts based on draft measurements but could give you an idea.

View attachment 472774View attachment 472775

Hello I did new sweeps and I think maybe im doing TA wrong by looking at the above graph (1) however yesterday (in youtube) I also saw to go by 100% peak but maybe it's wrong I did however match phase and TA if I did it right although I feel perhaps not, I also eqed with only sweep I know its wrong I just wanted to test it, I will do MMM now.

I did make minimum phase from cal files into new cal files with phase data.

However would appreciate it if you can give an opinion on what I did today.

 

[OCA]

Time alignment seems spot on. How will you sort crossover in Wiim?

I would not eq all the way to 500Hz. Maybe if you apply some windowing to the response first but it will usually sound boxed if you exceed about 200-250Hz. Less is more in DSP.

 

[JD1]

last MMM with EQ.

 

[JD1]

Time alignment seems spot on. How will you sort crossover in Wiim?

I would not eq all the way to 500Hz. Maybe if you apply some windowing to the response first but it will usually sound boxed if you exceed about 200-250Hz. Less is more in DSP.

I guess my last eq is better then up to 262 hz.

 

[JD1]

Can I do TA+phase using mains together vs sub or should I pick the 2 most distant speakers
Sub is farthest then right speaker is next to sub and left speaker is closest to me, do I do left vs SUB or mains vs SUB?