• WANTED: Happy members who like to discuss audio and other topics related to our interest. Desire to learn and share knowledge of science required. There are many reviews of audio hardware and expert members to help answer your questions. Click here to have your audio equipment measured for free!

Murphy Corner Line Array

BDE

Active Member
Joined
Nov 11, 2020
Messages
134
Likes
273
The major issue with line arrays is that you have to keep the distance between the driver centers as small as possible and this isn`t going to be changed just by puting it in a corner. So with the speed of sound and the distance you can calculate when side lobes arise. The spacing with the Dayton ND90 is at least 90mm, so at 4kHz. If you take some usual tweeters the distance is around 40mm -> 8,5kHz.
 
Last edited:

fluid

Addicted to Fun and Learning
Joined
Apr 19, 2021
Messages
691
Likes
1,196
Eventually, you realise you are dealing with fans.
It is always difficult to decide if you can take the opinions and information provided by others to make a decision for yourself. I hope I can find a speaker that is a better compromise and addresses some of the things I wish the array did better without losing anything that it does well. In the build or design threads I have posted I try and include as much data, measurement and simulation as possible so that others can make up their own minds.
 

kipman725

Active Member
Joined
Jul 6, 2020
Messages
255
Likes
224
ok but if you push the comb filtering high up enough in frequency the notch filters formed will have way less bandwidth than the critical bandwidth of the ear and won't be audible?

A high cost option to reduce comb filtering would be to use line array waveguides:
http://www.bmsspeakers.com/fileadmin/bms-data/product_data_2013/bms_4508nd_t_data.pdf
comb filtering should be minimized as the waveguide is designed for 0 degree vertical dispersion.

I use horns in my system, they are not all created equaly so some sound bad and some good, if you have had bad experience in the past its worth checking out the state of the art:
https://www.diyaudio.com/forums/multi-way/338806-acoustic-horn-design-easy-ath4.html
 

hex168

Senior Member
Joined
May 29, 2020
Messages
396
Likes
338
I've been thinking about the line array / multichannel-and-upmixing problem. Could this work?

Floor to ceiling line arrays of Peerless TC9s for Front Left, FR, Surround L, SR. Split the arrays into top and bottom for height and main channels. Bessel array of the same drivers, horizontal, for center. Possibly with switching to run the L & R fronts un-split for stereo. Possibly substitute shorter Bessel arrays for the SL and SR. Optimize positioning for Auro 3D.

I'm eagerly awaiting results on this thread:
https://www.audiosciencereview.com/...aper-experiment-of-a-bessel-line-array.24841/

Reference thread on TC9 array (very, very long):
https://www.diyaudio.com/forums/full-range/242171-towers-25-driver-range-line-array-193.html

Fluid, I do not want to post your thread since you chose not to, but it is very good as well.
 

hollis

Active Member
Joined
Sep 3, 2018
Messages
117
Likes
84
I've been thinking about the line array / multichannel-and-upmixing problem.
I have been thinking about that too, they would be excellent for up to 7.1, they will not work for Atmos, Auro3D or systems with heights.

Split the arrays into top and bottom for height and main channels.
I have experimented with 4-driver and short line arrays. We built a test box with only 4 drivers to look at seal, offaxis response and internal vs external driver mounting. Splitting the Murphy Corner Line Array in half, in my opinion, would ruin both halves. Splitting into 2/3rd, 1/3rd leaving less drivers for the heights may work, but the top 3rd would be arraying above you. You would need an extraordinary amount of eq to hear any highs from a sofa from the heights.

Now if one wants a very narrow (maybe 15deg?) vertical response, and is willing to angle the heights down at a sofa, then yes, 4-6 driver mini arrays would work for heights I feel.

One warning I can give you is that full length MCLA, with some basic furniture (two couches at a right angle, coffee tables and AV rack) will image slightly higher than your ears. This has been confirmed by three different listeners in my room. I feel this is due to less line of sight paths to the drivers lower than your ears, vs the drivers higher than your ears. But that is me guessing.

Thank you for the Bessel link, I'll check that out.
 

Attachments

  • 2020-05-25 16.03.41.jpg
    2020-05-25 16.03.41.jpg
    364.2 KB · Views: 123

hollis

Active Member
Joined
Sep 3, 2018
Messages
117
Likes
84
I thought about making an attempt, but with a different driver.
We are deeply considering a rebuild with the ND91. More powerhanding and slightly different response +5kHz

Wiring the whole array as 4-ohm vs 8-ohm is also up for debate. Some feel amps perform better (and measure better) at 8 vs 4. One thing for sure is that you will have sensitivity for days. 100-200w will go a long way. I am using my MCLA with a VTV build with Purifi modules, but I have not heard any differences vs SHD power.
 

ppataki

Major Contributor
Joined
Aug 7, 2019
Messages
1,216
Likes
1,355
Location
Budapest
Maybe a dummy question but would this concept work if using bigger drivers?
For example 4" Dayton RS100 or even 6" PS180? I guess beaming would be more but any other drawbacks? (let's assume appropriate room EQ like Dirac is in place)
Thank you
 

Zvu

Addicted to Fun and Learning
Joined
May 1, 2020
Messages
830
Likes
1,416
Location
Serbia
Either you stick to the ideals, and go for it, or start clattering your way down tin can alley with compromises that ‘sound fine’.

What are the speakers you are using now ? Being thorough, as you seem to be, can be quite a problem when juggling with compromises (ubiquitous for loudspeakers).
 

fluid

Addicted to Fun and Learning
Joined
Apr 19, 2021
Messages
691
Likes
1,196
Maybe a dummy question but would this concept work if using bigger drivers?
For example 4" Dayton RS100 or even 6" PS180? I guess beaming would be more but any other drawbacks? (let's assume appropriate room EQ like Dirac is in place)
Thank you
The centre to centre distance of the drivers dictates the frequency where comb filtering will start. Bigger drivers pushes this lower, where it starts to become more audible. There is a trade off when going smaller that the volume displacement of the drivers is less and so the low frequency output gets limited. This prompts the idea of a two way system which has it's own set of compromises.
 

ppataki

Major Contributor
Joined
Aug 7, 2019
Messages
1,216
Likes
1,355
Location
Budapest
The centre to centre distance of the drivers dictates the frequency where comb filtering will start. Bigger drivers pushes this lower, where it starts to become more audible. There is a trade off when going smaller that the volume displacement of the drivers is less and so the low frequency output gets limited. This prompts the idea of a two way system which has it's own set of compromises.

How will comb filtering be manifested? (I mean in the frequency and/or time domain)
Can it be measured?
 

fluid

Addicted to Fun and Learning
Joined
Apr 19, 2021
Messages
691
Likes
1,196
How will comb filtering be manifested? (I mean in the frequency and/or time domain)
Can it be measured?
You can see it in both and is easily measured, it looks much worse than it sounds.

You can see in this image the difference between an outdoor and indoor measurement. The combing is more visible outdoors where there are no reflections to fill in the gaps. Inside the room modes and interference are more obvious. The wiggle at 5K is where the vertical directivity is lost and there is a ceiling splash in a simulated polar plot. At 8k the combing becomes obvious due to the interference pattern generated by the line of drivers.

Outdoor vs Indoor.jpg


In the Impulse you can the initial peak is very sharp and then the train of later arrivals from the separated drivers is visible. Ignore the soft ramp to the impulse that is an effect of clock drift in that measurement.

2m no felt impulse.jpg
 

ppataki

Major Contributor
Joined
Aug 7, 2019
Messages
1,216
Likes
1,355
Location
Budapest
Thank you @fluid, much appreciated!
The indoor measurement looks good enough to me - I suppose no room correction DSP was applied. Did you try one of those (like Dirac)? Did it help?
I would have another point rather from a practical perspective: normally people place bass absorbers in the front (and back) corners of the room; if you put the line arrays in the corners where would you put your bass absorbers? I guess I am just trying to understand if the corner line arrays have any impact on room modes (if they behave differently from that perspective compared to regular floorstanding speakers where you can play around with their placement in the room)
 

fluid

Addicted to Fun and Learning
Joined
Apr 19, 2021
Messages
691
Likes
1,196
The graph above is a raw array without any EQ, it doesn't sound that good, too much low mid and no bass.

Yes I use DRC-FIR amongst other EQ and processing. DRC allows me to adjust the window length by frequency band so there is relatively longer windows in the bass and treble than in the midrange. The automatic correction is set to flat and comes out something like this.

R channel no psycho smoothing.jpg


Then an IIR EQ is used to set the tonal balance. I use a cascade of shelving filters so they can all be adjusted individually to get the desired effect. I have used target responses in DRC to have it all in one filter but it isn't quite as good somehow. You can get an idea of the amount of processing from the tick boxes to the left.

House Curve.JPG


I have mine free standing and not in the corners. I can understand the idea behind the corner position but it forces compromises I don't want to make. You can see the corner and first reflection absorbers in the image I posted on the first page. Even without any absorbers the spread of drivers in height creates an averaging effect so with some EQ to take down the worst modal peaks the bass can be very even.
 

BDE

Active Member
Joined
Nov 11, 2020
Messages
134
Likes
273
Thanks for the measurements!
I wondered about the early drop in the low frequency however after you wrote they are not in a corner nor on-wall it explains everything.
The FR looks fine but how about distortion? With small drives the distortions normally skyrocket beginning from 200-100Hz.
 

fluid

Addicted to Fun and Learning
Joined
Apr 19, 2021
Messages
691
Likes
1,196
They are 50cm from the wall where I have them now for some reinforcement. The room they are in in very open at either end and does not give that much assistance. No real issue with harmonic distortion. I don't have many distortion measurements I use them to check that nothing is broken.

Outdoor with no EQ
Distortion Outdoor no EQ.jpg


Outdoor with EQ
Distortion Outdoor EQ.jpg


Indoor above 200Hz the THD struggles to get above the noise floor of the room.

HD Wave Panel Proc.jpg
 

ppataki

Major Contributor
Joined
Aug 7, 2019
Messages
1,216
Likes
1,355
Location
Budapest
The graph above is a raw array without any EQ, it doesn't sound that good, too much low mid and no bass.

Yes I use DRC-FIR amongst other EQ and processing. DRC allows me to adjust the window length by frequency band so there is relatively longer windows in the bass and treble than in the midrange. The automatic correction is set to flat and comes out something like this.

View attachment 145834

Then an IIR EQ is used to set the tonal balance. I use a cascade of shelving filters so they can all be adjusted individually to get the desired effect. I have used target responses in DRC to have it all in one filter but it isn't quite as good somehow. You can get an idea of the amount of processing from the tick boxes to the left.

View attachment 145835

I have mine free standing and not in the corners. I can understand the idea behind the corner position but it forces compromises I don't want to make. You can see the corner and first reflection absorbers in the image I posted on the first page. Even without any absorbers the spread of drivers in height creates an averaging effect so with some EQ to take down the worst modal peaks the bass can be very even.

This is impressive indeed! :)
I am also doing DSP in Jriver using Dirac and other plugins so I am in the same boat too ;)
The difference is that I don't (yet!) have a line array speaker but a pair of 12" fullrange with a pair of 12" sub combo (that also needs heavy correction)

You mentioned some compromises for the corner placement - can you please elaborate what those are?
I see in the FR graph that apparently you would not need a sub to attain really low frequencies - is this really the case? I mean do you also feel the power of the bass? (like if you try watching a movie)
Thank you
 

ppataki

Major Contributor
Joined
Aug 7, 2019
Messages
1,216
Likes
1,355
Location
Budapest
One thing I just figured out is that the corner placement will force a 45 degree listening position which would put me very close to the front wall
(360cm room width --> listening position would be 177cm away from the front wall, pretty impossible)
So either the sweet spot will be like 1 meter away from me or I need to change the angle of the front of the speaker to 60 degrees...I am wondering if that would jeopardize the whole design
 

fluid

Addicted to Fun and Learning
Joined
Apr 19, 2021
Messages
691
Likes
1,196
This is impressive indeed! :)
I am also doing DSP in Jriver using Dirac and other plugins so I am in the same boat too ;)
Thanks, I've seen some of your other posts, I have EQuilibrium buried inside Metaplugin for Mid Side EQ to combat phantom centre tonal differences from removing early reflections. I like DRC-FIR because I can change every setting, Dirac won't let me do that. I understand why they don't let you because with that power comes the ability to destroy your sound too, not good for a commercial product.

You mentioned some compromises for the corner placement - can you please elaborate what those are?
I see in the FR graph that apparently you would not need a sub to attain really low frequencies - is this really the case? I mean do you also feel the power of the bass? (like if you try watching a movie)
Thank you
The room does need to be the right size and shape for any corner speaker to work. Most rooms don't fit what is needed. The other two main issues are cabinet volume and diffraction / reflection from the nearby surface. When placed in a corner the cabinet volume is lower than can be had in a freestanding speaker, whether it is triangular to go tightly in the corner or the shape used in the Murphy cabinet. This means that more EQ and amp power is needed to compensate for the earlier rolloff of the smaller cabinet. This limits even more how loud and low the speaker can be EQ'd. If a speaker has a 90 degree or narrower angle waveguide placed in a corner then there will be limited issues from that placement. Using a driver with a wide horizontal directivity causes very early reflections from any vertical surfaces.

You do not need a subwoofer for music, that was part of the attraction to me in building them. These go right down and when measuring in room I have to start the sweep above 10Hz to avoid shaking the structure of the house. There is of course a trade off from EQing a speaker for more low frequency output and that is maximum level. I don't listen loud, 70 to 80 dB average. The speakers can go louder but for reference level dinosaur footprints subwoofers would be advised, as they would be for almost any speaker.
 
Last edited:

Newman

Major Contributor
Joined
Jan 6, 2017
Messages
3,448
Likes
4,209
One thing I just figured out is that the corner placement will force a 45 degree listening position…

I’m not so sure. Assume the standard listening position is 60 degrees (it is). Options include angling them 10-15 degrees towards the listener, with very little down side wrt the corner‘s acoustic function, or listen to them 10-15 degrees off axis, which most speakers are very tolerant of and is routinely done with many speakers as a “cross the speakers a little in front of the listener” layout. Remember, it is quite common to fire speakers straight down the room and listen in a 60 degree layout, which is 30 degrees off axis, not the 15 degrees off axis of a 45 degree corner speaker listened to at 60 degrees.

Having said that, if the Murphy line is executed with a single full-range driver, it will get beamy above 5 kHz and might perform worse than most speakers when listening off axis.

cheers
 
Top Bottom