• WANTED: Happy members who like to discuss audio and other topics related to our interest. Desire to learn and share knowledge of science required as is 20 years of participation in forums (not all true). Come here to have fun, be ready to be teased and not take online life too seriously. We now measure and review equipment for free! Click here for details.

KEF LS50 WII in Room Measurements

jstownse

Member
Joined
Jun 4, 2021
Messages
15
Likes
20
Hi everyone,

I am an audio novice, and even less experienced with REW and all of its features and technical measurements. I am hopeful that this thread will still provide some interest to some, and perhaps some advice for me regarding how to improve my experience.

I’ve been enjoying my LS50 WII’s for a few months now, but bought a UMIK-1 and downloaded REW anyways. I’m hoping to just get a basic understanding of what I’m hearing now, and perhaps eventually try to EQ/room correction.

Depending on how my experience with the measurements and EQ proceeds, I’m hoping to settle on one of these options:

  • Stick with a “stock” LS50 WII experience, if the measurements and REW prove too challenging
  • Create an informed, basic, EQ profile using the KEF Connect app’s EQ functions that is good enough
  • Create a more robust EQ profile in Equalizer APO and use my computer to stream Spotify
  • Create a more robust EQ profile for a Roon and Qobuz trial to avoid the computer streaming solution

I also thought it might be interesting to “test” the effects of KEF’s different EQ functions built into their KEF Connect app. They include: “Bass Extension”, “Wall Mode”, “Desk Mode”, “Treble Trim” and a “Phase Correction”. I figured these tests will help me learn my way around REW and might produce some interesting data. I’m also hoping that you guys might find the data interesting and incentivized to suggest improvements, tips, or just basic commentary.

Anyways, I’ve read some of the REW guides, understood some parts, while much of it going way over my head. I did still manage to record several measurements on Day 1, and as far as I know, it was set up appropriately.

My plan to start is to post the frequency response graphs for the full range measurement, and then the overlays to some of my “tests”. If there’s any feedback on how to improve the measurement graphs in terms of usability, comments on the modes or any requests, please let me know and I’ll incorporate them into the future measurements.

Well here we go, here's my fist export of what was measured from the listening position:

Full Measurement.jpg


And here's the Wall Mode tests, this is how KEF describes it.

“Distance to the wall (Expert Mode: Wall Mode): This setting adjusts all frequencies from around 500 Hz downwards, causing broader changes than Desk Mode. A dip in these frequencies causes a tinny sound, while too much of these frequencies can overwhelm the overall presentation with lower frequencies.“

Wall Mode Test Export.jpg


And the Treble trim:


“How is your room? (Expert Mode: Treble Trim): This adjusts the frequencies above 500 Hz. A damped room (lots of curtains, soft furnishings) can make sound seem muted, while a lively room (high ceiling, flat surfaces, glass) can make it sound harsh. Move towards damped to reduce the muting effect, or move towards lively to tame harshness”

Treble Test Export.jpg


I hope these prove interesting and generate some discussion.
 

theyellowspecial

Active Member
Forum Donor
Joined
Sep 21, 2020
Messages
145
Likes
157
Probably the most balanced recommendation between simplicity and effectiveness is to use an auto-EQ solution for below room transition frequencies, then use REW for anechoic measurements and adjustments for higher frequencies.

Dirac has a 14-day trial.
 

mugbot

Member
Forum Donor
Joined
Jul 3, 2020
Messages
63
Likes
43
Location
Adelaide
What's your preferred source? I use wavelet for eq on android, and it's PEQ is pretty powerful.
You'd still use REW to generate the filters, but could apply these in wavelet to save using a pc.
Not sure if there's a similar solution for IOS sorry.
 

trl

Major Contributor
King of Mods
Joined
Feb 28, 2018
Messages
1,575
Likes
1,643
Location
Iasi, RO
The two peaks from around 86 Hz and 240 Hz should be notched down with a Parametric EQ, but the dips from around 70 Hz and 130 Hz need both speakers to be repositioned in the room due to a possible phase cancellation in the listener's position. You may need to experiment more with distance to the walls or near furniture, but you also need to add some damping in corners too.
 

Senior NEET Engineer

Addicted to Fun and Learning
Joined
Jan 6, 2020
Messages
515
Likes
541
Location
San Diego
Try moving the speaker closer to the floor. Might help with that 100-200hz dip.
 

jgiannakas

Member
Joined
Jun 22, 2021
Messages
45
Likes
51
Are this stationary mic measurements or moving mic? Some of the cancellations may very well be due to where the mic was. If not tried, I’d recommend a MMM measurement to compare :)
 
OP
J

jstownse

Member
Joined
Jun 4, 2021
Messages
15
Likes
20
The two peaks from around 86 Hz and 240 Hz should be notched down with a Parametric EQ, but the dips from around 70 Hz and 130 Hz need both speakers to be repositioned in the room due to a possible phase cancellation in the listener's position. You may need to experiment more with distance to the walls or near furniture, but you also need to add some damping in corners too.

Thanks for the comment. I don't have a ton of latitude for changing placement, but could probably move another 6 inches from the wall or laterally. So certainly not significant movement. I'll test it just out of the sake of interest, but could these type of movements yield much improvement?
 
OP
J

jstownse

Member
Joined
Jun 4, 2021
Messages
15
Likes
20
Are this stationary mic measurements or moving mic? Some of the cancellations may very well be due to where the mic was. If not tried, I’d recommend a MMM measurement to compare :)

Thanks for the comment, these were recorded with a stationary mic on the little mini tripod that came with the microphone, then raised up on essentially a box to be at approximate ear height.

I will look into the MMM and try that.
 

jgiannakas

Member
Joined
Jun 22, 2021
Messages
45
Likes
51
Thanks for the comment. I don't have a ton of latitude for changing placement, but could probably move another 6 inches from the wall or laterally. So certainly not significant movement. I'll test it just out of the sake of interest, but could these type of movements yield much improvement?
Also try moving them closer to the rear wall. It may be phase cancellation due to SBIR. Moving closer resolves it.
 
OP
J

jstownse

Member
Joined
Jun 4, 2021
Messages
15
Likes
20
Also try moving them closer to the rear wall. It may be phase cancellation due to SBIR. Moving closer resolves it.

I've tested three different distances from the rear wall, not really sure what to make of the results. When I conduct the measurements, it's just a few second sweep, is that generally sufficient or when people are posting their measurements is that an average of many measurements or longer measurements?

General theme seems to be that the closest to wall measurement has larger peaks and dips at the lower frequencies and seemingly tighter as the frequencies get higher.

Rear wall test.jpg
 
OP
J

jstownse

Member
Joined
Jun 4, 2021
Messages
15
Likes
20
Are this stationary mic measurements or moving mic? Some of the cancellations may very well be due to where the mic was. If not tried, I’d recommend a MMM measurement to compare :)

I've performed three measurements doing a general swirl with the microphone and have averaged the results. I realized the first full range was at 1/6, then my subsequent graphs were at 1/24, so I've done both here.
moving mic average 1.24.jpg


Moving mic average.jpg
 

napilopez

Major Contributor
Forum Donor
Joined
Oct 17, 2018
Messages
1,918
Likes
7,253
Location
NYC
I've tested three different distances from the rear wall, not really sure what to make of the results. When I conduct the measurements, it's just a few second sweep, is that generally sufficient or when people are posting their measurements is that an average of many measurements or longer measurements?

General theme seems to be that the closest to wall measurement has larger peaks and dips at the lower frequencies and seemingly tighter as the frequencies get higher.

View attachment 144835

I'm generally a fan of keeping the speakers close to the rear wall if you are able to do EQ. Genelec explains it best:


Curiously, it should be the opposite. The closer you are to the wall, the smaller the SBIR dips will be. So either that's a room node thing, or it's just how you're taking sweeps. since they are not averaged.

Other than the MMM method, which I find I sometimes need to to dial in my settings a bit to get proper results in the bass, you can try performing 5-10 sweeps at slightly different microphone positions and averaging them.
 

richard12511

Major Contributor
Forum Donor
Joined
Jan 23, 2020
Messages
4,024
Likes
5,801
I've performed three measurements doing a general swirl with the microphone and have averaged the results. I realized the first full range was at 1/6, then my subsequent graphs were at 1/24, so I've done both here.View attachment 144837

View attachment 144836

Have you tried the moving mic method yet? I find that method works best with 60+ measurements averaged together. Luckily, you have a great speaker, so it's only really the bass and low mids you need to worry about EQing. You're response right now actually looks pretty good for a pre-eq response.
 

jgiannakas

Member
Joined
Jun 22, 2021
Messages
45
Likes
51
I've performed three measurements doing a general swirl with the microphone and have averaged the results. I realized the first full range was at 1/6, then my subsequent graphs were at 1/24, so I've done both here.View attachment 144837

View attachment 144836
How far away are your speakers from the side walls?
Maybe post a picture of the front of the listening space if you can?

cancellations can happen either from the rear wall, the side walls or the back wall if your sofa is at the “right” distance to it.

ideally you want your speakers as close as possible to the front wall, at least 3.5 meters away from the side walls (to avoid any cancellations above 80hz) and same for the sofa (3.5 meters away from the rear wall or place it right up against it.

in genera cancellations happen at 1/4 (quarter) wavelength distances from boundaries against specific frequencies. So you can quite easily calculate where your speakers should go to either move the cancellations up where sound is directive and they don’t matter much or low enough that the speakers don’t reach them. Hence my recommendation above for 3.5 meters (the 1/4 wavelength of 80hz)
 
OP
J

jstownse

Member
Joined
Jun 4, 2021
Messages
15
Likes
20
How far away are your speakers from the side walls?
Maybe post a picture of the front of the listening space if you can?

cancellations can happen either from the rear wall, the side walls or the back wall if your sofa is at the “right” distance to it.

ideally you want your speakers as close as possible to the front wall, at least 3.5 meters away from the side walls (to avoid any cancellations above 80hz) and same for the sofa (3.5 meters away from the rear wall or place it right up against it.

in genera cancellations happen at 1/4 (quarter) wavelength distances from boundaries against specific frequencies. So you can quite easily calculate where your speakers should go to either move the cancellations up where sound is directive and they don’t matter much or low enough that the speakers don’t reach them. Hence my recommendation above for 3.5 meters (the 1/4 wavelength of 80hz)

Thanks for the help. Unfortunately don't have a ton of room for significant movement changes. Nor can I veto the the blanket basket (who knew that was something that is needed). I did do a simple measurement test to see if the basket had more impact on the measurements if it was really close to the speaker, I'm thinking sufficiently underneath it, versus closer to the listening position. Closer to the speaker seemed to be better.

Also excuse the wire mess, this was a photo soon after moving in.

Sofa happens to be right up against the back wall as well.

PXL_20210604_211436710.jpg
 
OP
J

jstownse

Member
Joined
Jun 4, 2021
Messages
15
Likes
20
Have you tried the moving mic method yet? I find that method works best with 60+ measurements averaged together. Luckily, you have a great speaker, so it's only really the bass and low mids you need to worry about EQing. You're response right now actually looks pretty good for a pre-eq response.


Thanks. I'll try the moving mic method again with additional measurements to average, my first attempt was just three. From an average with that many more samples, I ought to have a pretty good baseline with which to try some EQing.
 
OP
J

jstownse

Member
Joined
Jun 4, 2021
Messages
15
Likes
20
I have no idea if there is any interest in this "test" but here are my results of triggering the different base extension settings in the KEF Connect app.

“How large is your room (Expert Mode: Bass Extension): This setting adjusts the bass extension of the speakers. The larger the room selected, the lower the roll-off point.“

Bass Extension.jpg
 

jgiannakas

Member
Joined
Jun 22, 2021
Messages
45
Likes
51
Thanks for the help. Unfortunately don't have a ton of room for significant movement changes. Nor can I veto the the blanket basket (who knew that was something that is needed). I did do a simple measurement test to see if the basket had more impact on the measurements if it was really close to the speaker, I'm thinking sufficiently underneath it, versus closer to the listening position. Closer to the speaker seemed to be better.

Also excuse the wire mess, this was a photo soon after moving in.

Sofa happens to be right up against the back wall as well.

View attachment 145371
I’m in the same position as you, fixed sub optimal placement with very little wiggle room :)
Do try and get three sets of moving mic measurements (left only, right only, both) without any smoothing to see what’s going on. Try to take them for about 1 minute each to get enough samples (basically until the line stops moving!). More likely than not your mic placement in single measurements is also causing cancellations due to the rear wall being close to the sofa.

Your most likely next steps to bring improvement in audio quality will result from a sub and a mini dsp to eq it.
 
Last edited:

justinkhuong

Member
Joined
Aug 8, 2021
Messages
6
Likes
2
Location
Vietnam
Hi. Im very interested in doing this myself. I also have a pair of KEF LS50W ii. But I can't buy the microphone recommended by @amirm, which is the UMIK- 1 (I live in Vietnam). I'm not sure if I can use another USB microphone that is available such as some ATH or Rode...
 
Top Bottom