Jump to content

An attempt to build K-402-MEH clone in wood


StabMe

Recommended Posts

18 hours ago, gnarly said:

Super nice project ! Looks very well made!  Looking forward to your listening impressions 😀

 

Thanks! Will definitely share those along with measurements i will be able to take.

 

18 hours ago, gnarly said:

It looks like your port size is about 1/10 Sd.  Is that correct?  1/10 has been my stand ratio on a handful of MEH builds, using 12", 10", and 8" low/mids.

How high do you plan to run the 15"s ?

 

Correct, the ratio is about 1/10th - as per Chris' experience and recommendation. My plan is to run mids up to about 450-500 or somewhere in the middle. Will decide upon measurements.

 

18 hours ago, gnarly said:

I guess i maybe got inspired the same way you did, by ChrisA's praise of his K-402 MEH, and the clone-like build by the Australian fellow named Oohms........

My build uses curved XPS foamboard secondary flares, wrapped with glass cloth and epoxy, to try to match the geometry  Oohm layed out. 

Think I got reasonably close, but yours is no doubt a better match.

Anyway, mine use a bms4594he CD, crossed to a pair of faital 10pr300's at 500Hz.  All my speaker builds are designed to cross to subs at 100Hz, so I didn't  need anything bigger than 10"s.

 

I like your build! Looks impressive. So you've made curved parts out of XPS, covered them with fiberglass and epoxied them as to make them sturdier and put them inside of the box? Or is curved flare detachable? Those almost fridge sized boxed below speakers - are those subwoofers?

 

 

There is a guy over at DIYAudio who made a few Synergy builds, who likes to cross them at 100Hz, likes single injection port per driver and likes them not to be offset. Is it you by chance? Can't remember the nickname at that forum.

 

 

Link to comment
Share on other sites

32 minutes ago, StabMe said:

Do those windows pose a real problem as far as corner loading goes?

It's only a problem if it affects the sound quality.  The best way to deal with it is assume it isn't a problem, then put absorption over them, listening and measuring (looking at the filtered IR [dbFS], and spectrogram plots at the time associated with the loudspeaker-->window reflection-->microphone distance).  The Schroeder integral on the filtered IR plot should be so far away from the decay trail that you're probably not going to have strong effects from the window, but the adjacent walls (side and front walls) will have some effect on the phantom center imaging.  It's best just to try it in A-B fashion using absorption material to hear the difference.  I adjusted the amount of absorption squares  and their placement based on using my ears and those measurements.

 

32 minutes ago, StabMe said:

Maybe some other problems? Any obvious places to add room treatments?

You can look at the "RT60" plots (mainly using RT20, RT30, and EDT plots) to keep the reverberations times below ~0.5 seconds across the listening bands.  For instance, this is where I ended up with my listening room as measured 1m in front of the right loudspeaker with foam rubber on the floor between the right Jubilee and the microphone:

 

Chris A Listening Room - 1 M Right.jpg

 

734051168_RightJubilee-Axi20501mFilteredImpulseResponse.jpg.2fa4bdf03224efc1e72f01389a43b7be.jpg

 

265469362_RightJubilee-Axi20501mSpectrogram.jpg.6000912f2d790888645b965adf67a2ef.jpg

 

Chris

 

  • Like 1
Link to comment
Share on other sites

The spectrogram plot shows the dog leg of the peak energy time curve centered on the Axi2050 crossover frequency, something that I have yet to smooth out using the Xilica crossover, but other than that, much of what you see is facilitated by the position of the microphone (1 m from the horn mouths--not at the listening position) and foam rubber on the floor between the microphone and horn mouths to catch the floor bounce. 

 

Here's the same thing with the K-402-MEH prototype, showing a floor reflection or listening chair reflection at ~6.5 ms:

 

379123656_K-402-MEH-BMS4592ND1mFitleredIR(centered).jpg.244b4bb69db86b4b72a257a46d3788fa.jpg

 

1057610190_K-402-MEH-BMS4592ND1mSpectrogram(centered).jpg.bc812c372de238d8825ae88a783d60b0.jpg

 

Chris

  • Like 1
  • Thanks 1
Link to comment
Share on other sites

[mention=62555]Chris[/mention] Would it be better to assemble some wall acoustic treatment panels to make a box around the front of the speaker, and have the mic inside of the acoustic panel box? Sorta like a acoustic panel mini anaholic chamber for tuning your crossover?

 

Sent from my SM-G985F using Tapatalk

 

 

 

 

 

 

 

 

 

Link to comment
Share on other sites

4 hours ago, StabMe said:

@ChrisA, what is the reason for measuring at 1m from the speaker as opposed to at the listening position? Intuitively i would assume that what matters is what's happening at where my ears at. 

I recently responded to the same question on a PM a couple of days ago.  Here is the text of that response:

 

Quote

First, let's think about the "measurement from the listening position (LP)" situation this way: 

 

If you had a perfect loudspeaker (anechoically) and placed it in a real listening room, you would have the same measurement issues at the listening position that you're having presently?  Is that the fault of the loudspeaker or room?  How about the placement of the loudspeakers in-room?  How can perfect loudspeakers be faulted?

 

Now, with the perfect loudspeaker in your room, what are you going to "correct"?  Flattening the peaks in the SPL response only by using attenuating EQ filters? 

 

Are you going to try to boost the nulls or valleys in SPL response?  What if those nulls are due to reflections in-room?  You can't boost the SPL response of those room-related issues, because the destructive reflections get proportionately stronger as you turn up the drive level at those frequencies.  If you move your listening position around or the loudspeakers, those problematic frequencies change.  So are you correcting the SPL response for your head at one precise location and the loudspeakers in their positions (i.e., head in a vice)?  What about someone else in the room?  Does the SPL response suck for them? 

 

What if you set up the loudspeakers for near perfect SPL response in-room, measured only at one metre on-axis, i.e. without the measurements including room reflections?  Would that provide the best compromise for all listeners wherever they are in-room? Think carefully.

 

Now let's think about phase response.  The phase response (vs. frequency) of direct arrivals from a perfect loudspeaker in-room would be flat phase response.  If any [room] reflections make it into the measurements are included, the phase response is destroyed, and is a function of where you are listening within the room.  Can the human hearing system pick out the reflections from the direct arrivals (as it is designed by nature do do)?  The answer is, "it depends on how bad those in-room reflections are."  If the reflections are minimized relative to direct arrivals, it been my distinct experience that I can hear the flat loudspeaker phase response from the LP.  How can that be? 

 

I take measurements at one metre (only), and leave the "measurements at the listening position" to someone else.  I find that all measurements at the LP are actually garbage, and never give me the best subjective sound from the setup.  I learned this the hard way over a lot of trials and errors--and years.

 

By the way, the only time that I would take acoustic measurements at the listening position is for understanding deep subwoofer response below 60 Hz (SPL and phase).  At that point, room modes become dominant, and understanding what you've got in terms of placement of the subwoofers and the listening positions in-room is the only real adjustment that you've got.  You can also flatten the peaking SPL response, but flattening phase response is very difficult to achieve below 100 Hz.

The deal with apparently all "room correction software" packages is that they all are trying to make the dialing-in process automated so that the user is insulated from the complexities of that process and doesn't have to know anything. 

 

The problem is that all the measurements are made from the listening position [using "room correction software" apps], not at the much more optimal 1 metre on-axis position in front of each loudspeaker.  This means that the most important capability of these apps is compromised, or at least focused exclusively on subwoofer SPL response. The greatest complaint that I've experienced personally and heard is that the resulting SPL response of the loudspeakers is pretty seriously degraded.  This is the one task that it needs to have almost perfect performance, and they all seem to fail--even Dirac [in my experience].

 

All of the above is "in my experience/in my opinion"...so if you disagree with anything, you can do it your way and I'm not going to get in your way. 

 

What I wrote above is how I feel about taking measurements and is (I believe) how Roy basically views the situation of (at least) flattening the SPL response and the reason why anechoic measurements are so important--at least.  

 

But perhaps the phase flattening portion of the discussion isn't relevant to how he views the importance of phase flatness.   

 

YMMV.

 

Chris

  • Like 1
  • Thanks 2
Link to comment
Share on other sites

Should measurements be taken from 1 meter when blending a sub with the main speakers?  Or should that be done from the Listening Position?

 

Would it make sense to take measurements from the Listening Position to determine whether there are any room reflections to reduce?

 

Thanks.

 

Mark

Edited by Marks
Link to comment
Share on other sites

9 hours ago, Chris A said:

All of the above is "in my experience/in my opinion"...so if you disagree with anything, you can do it your way and I'm not going to get in your way. 

 

What I wrote above is how I feel about taking measurements and is (I believe) how Roy basically views the situation of (at least) flattening the SPL response and the reason why anechoic measurements are so important--at least.  

 

But perhaps the phase flattening portion of the discussion isn't relevant to how he views the importance of phase flatness.   

 

 

Chris, this is great information and i will definitely use that approach when the time comes and compare the result with correction over the listening position. Lately i've been using a movic mic measurement technique with REW which gave me nice results. It is gonna be very interesting to A/B the results.

 

There is also an old ConEq software and newer software called APL by the same guy who was the main developer behind ConEq. This software also uses a nearfield measurements, although they take about 80-300 measurements per time by playing and recording short sweeps:

 

841868010_ConEqmeasurements.gif.7a9f5ca45149964c465da548f6534bc3.gif

 

But when measuring the reverberation and effect of room treatments, do we really need to take the room out? Isn't the room and its behavior in fact what ought to measure? This part is confusing.

Link to comment
Share on other sites

4 hours ago, StabMe said:

But when measuring the reverberation and effect of room treatments, do we really need to take the room out? Isn't the room and its behavior in fact what ought to measure? This part is confusing.

There is something that needs to be corrected, but generally speaking most people couldn't (or wouldn't) tell you what that "something" is.  PWK talked about it, and generally was almost silent on the other things that can't really be corrected.  A hint: we've already talked about the things that cannot be corrected using DSP or other signal processing methods, and we must arrange our room and listening positions to minimize their effects, as well as use room acoustic treatments to minimize them. 


So what does that leave? 

Link to comment
Share on other sites

If you have a single listening location, it seems one should do all the filtering at the listening point. Breaking it up into two components (1 meter and listening point) cannot do better in theory,  and in general would be expected to do worse. With multiple listening positions the situation is different. The other caveat is the ability to interpret the measurements and constrain the filtering appropriately (this issue would have to be addressed to some extent regardless). 

 

There are many successful implementations of single listening point filtering with psychoacoustic criteria , such as Acourate. It has been used successfully with Cornscala-type speakers. I would like to try it with my Klipschorns, but the PC-based audio interface is a barrier for me. 

Link to comment
Share on other sites

21 hours ago, StabMe said:

 

I like your build! Looks impressive. So you've made curved parts out of XPS, covered them with fiberglass and epoxied them as to make them sturdier and put them inside of the box? Or is curved flare detachable? Those almost fridge sized boxed below speakers - are those subwoofers?

Thanks ! !

Yes, the horn is regular flat baltic birch, and the curved flares are detachable.  They bolt onto the outside, and would not have been strong enough at all, without being fiber glassed.

Wood horn has mouth of 33x19, flares add up to about 49x29  (90Hx60V)

 

And yep, they are double 18" slot-loaded bass-reflex subs.

21 hours ago, StabMe said:

 

 

There is a guy over at DIYAudio who made a few Synergy builds, who likes to cross them at 100Hz, likes single injection port per driver and likes them not to be offset. Is it you by chance? Can't remember the nickname at that forum.

 

 

One more yes, 😀 i'm mark100 on DIY. 

100Hz and single round centered port has worked well on several tries. So have corner ports but I haven't found any advantage to them yet.

 

Where are you planning to cross to the subs you have in mind? 

Link to comment
Share on other sites

1 hour ago, Chris A said:

There is something that needs to be corrected, but generally speaking most people couldn't (or wouldn't) tell you what that "something" is.  PWK talked about it, and generally was almost silent on the other things that can't really be corrected.  A hint: we've already talked about the things that cannot be corrected using DSP or other signal processing methods, and we must arrange our room and listening positions to minimize their effects, as well as use room acoustic treatments to minimize them. 


So what does that leave? 

 

Well, what's left is the speaker itself. So by measuring from 1m distance we are correcting speakers, not the speaker-room interaction? Or, even while the mic is placed 1m from the speaker, we still have the room effects, only we see less of the room in the measurements. My understanding. 

  • Like 1
Link to comment
Share on other sites

13 minutes ago, gnarly said:

Yes, the horn is regular flat baltic birch, and the curved flares are detachable.  They bolt onto the outside, and would not have been strong enough at all, without being fiber glassed.

Wood horn has mouth of 33x19, flares add up to about 49x29  (90Hx60V)

 

After seeing a few builds with detachable second flare, i was contemplating on the idea of making mine like this, but then i decided to make it full range and used double 15", which kinda made the detachable flare impossible.

 

13 minutes ago, gnarly said:

One more yes, 😀 i'm mark100 on DIY. 

100Hz and single round centered port has worked well on several tries. So have corner ports but I haven't found any advantage to them yet.

 

Where are you planning to cross to the subs you have in mind? 

 

You didn't document this one on the forum, did you? Found your older threads, but not one on this buid.

 

I will first try it without the sub. If I feel that the sub is needed, i will make two of them based on the Kicker driver I have (mentioned earlier in this thread) and will cross them at around 30-40Hz.

 

Quote

And yep, they are double 18" slot-loaded bass-reflex subs.

 

They are monstrous :)

 

BTW, Chris, what slope do you use on your subs? LR24?

Edited by StabMe
Link to comment
Share on other sites

51 minutes ago, StabMe said:

Well, what's left is the speaker itself.

Try again.  You missed.

 

What loudspeaker + room acoustics + listener characteristic has minimum phase characteristics (i.e., can be corrected properly using DSP)--assuming the loudspeaker was perfect anechoically?  This is the only thing that can be "corrected".

 

And how do you separate the all-pass (or non-minimum phase) issues from the minimum phase ones, using something like REW or other in-room measurement setup?

 

Chris

  • Like 1
Link to comment
Share on other sites

1 hour ago, StabMe said:

 

After seeing a few builds with detachable second flare, i was contemplating on the idea of making mine like this, but then i decided to make it full range and used double 15", which kinda made the detachable flare impossible.

Yeah, that would have made it really huge with double 15"s

1 hour ago, StabMe said:

 

 

You didn't document this one on the forum, did you? Found your older threads, but not one on this buid.

 

 

I will first try it without the sub. If I feel that the sub is needed, i will make two of them based on the Kicker driver I have (mentioned earlier in this thread) and will cross them at around 30-40Hz.

 

 

They are monstrous :)

 

It was titled "synergy...take#7.  The foamboard flares don't enter the thread until post 30 something. https://www.diyaudio.com/forums/multi-way/351670-synergy-7-a.html

A good picture of how the curved foamboard bolts to the horn is at bottom of this post.  https://www.diyaudio.com/forums/multi-way/362087-somewhat-easier-build-synergy.html

 

Thx again, and good luck...hope you don't even need the subs😀

1 hour ago, StabMe said:

 

 

  • Like 1
Link to comment
Share on other sites

1 hour ago, StabMe said:

Well, you are probably correcting direct speaker response as opposed to a mix of direct and reflected sound.

In this example, the loudspeaker is perfect anechoically. 

 

What's different and not part of the "reflected sound" (the portion of the interior acoustics to rooms that's not correctable using PEQs due to 1/4 wavelength boundary reflection cancellations and to some degree, 1/2 wavelength resonances, etc.)? 

 

You can't blame this on the loudspeaker (as most people try to do). 

 

Chris

  • Like 1
Link to comment
Share on other sites

Join the conversation

You can post now and register later. If you have an account, sign in now to post with your account.
Note: Your post will require moderator approval before it will be visible.

Guest
Reply to this topic...

×   Pasted as rich text.   Paste as plain text instead

  Only 75 emoji are allowed.

×   Your link has been automatically embedded.   Display as a link instead

×   Your previous content has been restored.   Clear editor

×   You cannot paste images directly. Upload or insert images from URL.

×
×
  • Create New...