3-way for bookshelf (in-wall) - Page 12 - diyAudio
Go Back   Home > Forums > Loudspeakers > Multi-Way

Multi-Way Conventional loudspeakers with crossovers

Please consider donating to help us continue to serve you.

Ads on/off / Custom Title / More PMs / More album space / Advanced printing & mass image saving
Reply
 
Thread Tools Search this Thread
Old 18th April 2014, 01:12 PM   #111
soren5 is offline soren5  Denmark
diyAudio Member
 
Join Date: Jun 2011
Location: Copenhagen
Quote:
Originally Posted by andy19191 View Post
Whether it is a better tweeter is for you to decide. The primary purpose of the waveguide is to control directivity and match the directivity of the mid and tweeter at crossover. You can see this in the example design. For a tweeter without a waveguide there is a jump in directivity between the tweeter and mid. This results in a step change in the frequency response of a substantial proportion of the reflected sound.
And that’s not an easy decision to make since I have no experience with waveguides. On the contrary I already know the SB29RDNC from my current speakers and I really like their sound, even off-axis.
I have read up on waveguides, trying to understand how they work and what benefits they have. Waveguides do indeed help to control the directivity of the higher frequency, which means better off-axis response. But waveguides also give a rise in the FR, the “waveguide hump”, which I assume you will have to deal with somehow? Zaph has a page discussing waveguides here. I have searched for opinions on the SEAS DXT and from what I’ve found people don’t seem overly impressed by it, see e.g. this PE thread.

I have looked at the floor plan and speaker placement I tried to illustrate the horizontal off-axis response, see attached drawing showing 15* and 30* off-axis. If you believe off-axis response is a great concern for this project I guess there are other candidates to consider, e.g. metal dome tweeters such as Vifa NE19-VTA or SB26ADC. Again, I’m not sure how well either of those would integrate with the SB mid and woofer. The SB29RDNC+SB15NRXC30-8 OTOH is a tried and tested combination.

Quote:
Originally Posted by andy19191 View Post
The simulation programs are not exact and experienced designers often opt for ports somewhat longer than that for maximum bass extension. I would suggest going for an arrangement that allows you to modify the length based on what you find.
Good idea. I have looked at these ports from Monacor, which are adjustable. Two of those will give me an area of 40.8 cm2. Do you think that’s enough for the 10” SB?

Quote:
Originally Posted by andy19191 View Post
The best shape is probably a cylinder with a bulge to get round the magnet while maintaining a constant(ish) cross sectional area and then gently tapering down. A square box with a cross section at the magnet slightly greater than that of the cone is likely to be OK. What you want to avoid is a sharp change in cross sectional area in the direction the rear wave is propagating because this will reflect the sound. I am not sure I can see the purpose of the current shape since a normal tube or box may well give a longer path length through the dissipating material before what is left of the sound returns to the cone. But I doubt this sort of thing is of great importance so long as the chamber is appropriately stuffed.
I have looked at using large diameter PVC tube instead. See the attached drawing with a 160mm tube. Internal volume is now a little over 4L. Last picture is a view from inside the tube showing a 10mm chamfer on hole for the mid.

Quote:
Originally Posted by andy19191 View Post
Why throw away all that volume? What is wrong with a gap of a few mm? If you seal the entrance with a soft rubber strip what is the purpose of putting material between the bookshelf and the speaker?
So you would only put absorbing material in the gaps around the front of the speaker? I thought I would need to have material on all the surface between the speakers and the bookshelf to avoid having two hard surfaces touching each other.
Anyway, I have updated the dimensions of the speaker. It is now only a tiny bit smaller than the shelving unit, just 2mm all the way around. I have also changed to 16mm MDF and simplified the bracing. The updates give my the following Vb calculations:

Volume: 69 L
Midrange: -4.5 L (4 L internal)
Bracing: -2.3 L
Ports: -2.0
Woofer: -2.0
Vb: 58.2

Thoughts?
Attached Images
File Type: png Off-axis floor plan.png (191.0 KB, 80 views)
File Type: png Sigurd Stereo 05.png (73.8 KB, 78 views)
File Type: png Sigurd Stereo 06.png (70.7 KB, 78 views)
File Type: png Sigurd Stereo 07.png (76.3 KB, 76 views)
  Reply With Quote
Old 18th April 2014, 06:17 PM   #112
jReave is offline jReave  Canada
diyAudio Member
 
Join Date: Oct 2012
A couple of questions.

I'm wondering if the listening room actually has a couch or a chair in it across from the speakers that would be a central listening spot if and when your friend ever wants to sit down to enjoy the music. If so, then what is that distance? I ask because that is the position I would model the xo for.

Also, at the risk of creating more decisions, I haven't used an active xo but I'm assuming that if you use one that you can eq the bottom end. If this is the case, then we no longer have to work with the same set of restrictions that were in place in originally selecting the woofer (that only took me a few days to figure out sheesh). In other words, you can choose a woofer of different sensitivity and impedance if you want and you could possibly run it sealed and then eq the bottom end to taste. This will eliminate the problem of port tuning and could result in a less expensive woofer and/or amp (smaller power requirement to reach the same SPL with a higher sensitivity driver). Also volume constraints could be relaxed and you get the benefit of what some people consider to be better bass with a sealed box.

The SB29 actually works sealed in 30L to 104dB/1m with 70W before exceeding xmax.
The Peerless 10" works sealed in 45-50L to 104.5dB with 40W.
And looking at what else Falcon had available,
the Seas CA26RE4X 10" will work sealed in 25-30L to 108dB with 50W. http://www.falconacoustics.co.uk/dow..._datasheet.pdf
All require eq though to effectively replace what the ports would be doing. Both alternatives to the SB are less expensive btw.

I have worked out xo's for the SB15NRXC-8 and both the 29RDC and the 29RDNC. Both combinations go together more easily and more simply than with the 17NRXC-4 and are very very similar in response and in values. I'll post them shortly. Personally I would stick with either of these tweeters - I've just been hearing so many good things about them lately and their off-axis responses are actually quite good. If making the choice, I would go with the 29RDC because of the slightly lower distortion and the more symetrical impedance response. Perfectionist stuff.

I do actually agree with andy about the mid enclosure and think the 160mm tube would be an improvement. About 4.5L internal with heavy stuffing in the back half or 2 thirds should work nicely. I actually tend to like to give the mid even more room, so Qtc a little lower, but that might be tough to do in your case. Depends on your above decision.

Your bracing was better before. Turn them back 90* so the longer axis is perpendicular to the panels again. More stiffness.
  Reply With Quote
Old 18th April 2014, 09:18 PM   #113
jReave is offline jReave  Canada
diyAudio Member
 
Join Date: Oct 2012
Let me just address directivity for a second. First attachment below tries to show in graphical terms what the horizontal directivity looks like for each of the SB drivers you're looking at. With the mid/tweeter xo at 2Khz and with a 4th order acoustic slope, the continuity of off-axis behavior as one driver hands off to the other is excellent. Thirty degrees off-axis for the whole speaker with xo's does not in fact drop below the on-axis response by any significant degree until up around 10kHz. If your ears tell you to, you can lower the padding a touch on the tweeter if you think it sounds better.

The next chart is what your Piccolo looks like when I sim it with the files available to me. That looks pretty good and is a useful standard of comparison when looking at the next sims with your baffle dimensions. For these I used the center of the tweeter at about 7.6 to 9cm below the top of your shelving unit and they are based on a listening axis that is 2.5m away, 25cm above the middle of the tweeter and about 20* horizontally off-axis, which would be a seated position in between the speakers. I haven't included the driver phases because they get so messy at these listening angles but you can see how well they align by the good reverse nulls in each graph.

The last attachment shows the xo values for each tweeter and mid. I can post them as schematics and let you know which values you might want to change in terms of voicing it if you like. There is quite a difference between these values and what the Piccolo uses which is what you should expect with such a huge difference in baffle configuration and placement.

(Second chart is mislabeled - the SB29NRX30-8 should be 15NRXC30-8. This is the Piccolo.)

Cheers
Attached Images
File Type: gif directivity summation.GIF (69.3 KB, 63 views)
File Type: gif summed FR Piccolo.gif (34.5 KB, 18 views)
File Type: gif summed FR w 29RDNC.gif (32.0 KB, 17 views)
File Type: gif summed FR w 29RDC.gif (31.8 KB, 16 views)
File Type: gif xo values for 2 x SB29 twtrs.GIF (73.2 KB, 15 views)
  Reply With Quote
Old 18th April 2014, 10:46 PM   #114
jReave is offline jReave  Canada
diyAudio Member
 
Join Date: Oct 2012
I think I need to amend my sealed woofer suggestion.

I forgot to include the extra amp power that you need in order to do the LF boosting. So with the Peerless 10" for eg, it will need about 6-8dB of boost at about 25Hz to reach the same level as the SB29RNX ported at 25Hz. So that's 40W doubled at least twice again, so 80W and then 160W plus headroom if you want to make it work at those SPL levels. And it's only rated at 140W max. The Seas needs even a little more boost than that. Probably best to stick with the SB ported. Oh well........
  Reply With Quote
Old 19th April 2014, 06:25 PM   #115
diyAudio Member
 
Join Date: Apr 2005
Location: -
Quote:
Originally Posted by soren5 View Post
I have looked at these ports from Monacor, which are adjustable. Two of those will give me an area of 40.8 cm2. Do you think that’s enough for the 10” SB?
It looks to be in the right ball park. Air velocities look OK for longer port lengths with a conservative high pass filter but perhaps a bit marginal for shorter port lengths.

Quote:
Originally Posted by soren5 View Post
So you would only put absorbing material in the gaps around the front of the speaker? I thought I would need to have material on all the surface between the speakers and the bookshelf to avoid having two hard surfaces touching each other.
I am not sure I understand. The speaker is probably best attached to the rear wall or floor via rubber isolators without touching the bookshelf anywhere. A soft rubber ring or felt can used to keep air out of the cavity and keep a smooth baffle surface. Putting some fibreglass or speaker damping material in a significant sized cavity might be a good idea but it should avoid transferring the vibrations of the speaker walls to the bookshelf.
  Reply With Quote
Old 19th April 2014, 11:42 PM   #116
soren5 is offline soren5  Denmark
diyAudio Member
 
Join Date: Jun 2011
Location: Copenhagen
Quote:
Originally Posted by jReave View Post
I'm wondering if the listening room actually has a couch or a chair in it across from the speakers that would be a central listening spot if and when your friend ever wants to sit down to enjoy the music. If so, then what is that distance? I ask because that is the position I would model the xo for.
Yes. There will be an ottoman placed in the living room, almost in centre of the speakers, about 3.5 meters away. This spot will be 21* horizontally off-axis. By the way, we have to move the right speaker one shelving unit to the right as there are some radiator pipes that would otherwise limit the depth of the speakers. The centre of right speaker is now 56 cm away from the right wall while the centre of the left speaker is 78 cm away from the left wall.

Quote:
Originally Posted by jReave View Post
I do actually agree with andy about the mid enclosure and think the 160mm tube would be an improvement. About 4.5L internal with heavy stuffing in the back half or 2 thirds should work nicely. I actually tend to like to give the mid even more room, so Qtc a little lower, but that might be tough to do in your case. Depends on your above decision.

Your bracing was better before. Turn them back 90* so the longer axis is perpendicular to the panels again. More stiffness.
Okay, I have increased to mid volume to 4.5 and changed the bracing slightly. See attached drawing. I save a little volume on the bracing but loose some on the mid. Vb is still around 58 L.

Quote:
Originally Posted by jReave View Post
Let me just address directivity for a second. First attachment below tries to show in graphical terms what the horizontal directivity looks like for each of the SB drivers you're looking at. With the mid/tweeter xo at 2Khz and with a 4th order acoustic slope, the continuity of off-axis behavior as one driver hands off to the other is excellent. Thirty degrees off-axis for the whole speaker with xo's does not in fact drop below the on-axis response by any significant degree until up around 10kHz. If your ears tell you to, you can lower the padding a touch on the tweeter if you think it sounds better.
That looks good to me

Quote:
Originally Posted by jReave View Post
The next chart is what your Piccolo looks like when I sim it with the files available to me. That looks pretty good and is a useful standard of comparison when looking at the next sims with your baffle dimensions. For these I used the center of the tweeter at about 7.6 to 9cm below the top of your shelving unit and they are based on a listening axis that is 2.5m away, 25cm above the middle of the tweeter and about 20* horizontally off-axis, which would be a seated position in between the speakers. I haven't included the driver phases because they get so messy at these listening angles but you can see how well they align by the good reverse nulls in each graph.
Thanks for the help! That looks really good. If we take the driver layout from the attached drawing, the centre of the tweeter is going to be 9.5cm below the top of the shelving unit (and 8.3cm below the bottom of the top, which is 12mm and sticks out 10mm). The centre-to-centre distance between tweeter (29RDNC) and mid is 12cm. If the 29RDC is used these measurements may change slightly. I think we should just stick with the 29RDNC.

Quote:
Originally Posted by jReave View Post
The last attachment shows the xo values for each tweeter and mid. I can post them as schematics and let you know which values you might want to change in terms of voicing it if you like. There is quite a difference between these values and what the Piccolo uses which is what you should expect with such a huge difference in baffle configuration and placement.
Please do! That would be very helpful. Looks like this project is getting closer to being realised This is all very exiting! I promise to post many photos when I start to build.
Attached Images
File Type: png Sigurd Stereo 08.png (90.9 KB, 15 views)
  Reply With Quote
Old 20th April 2014, 12:03 AM   #117
soren5 is offline soren5  Denmark
diyAudio Member
 
Join Date: Jun 2011
Location: Copenhagen
Quote:
Originally Posted by andy19191 View Post
It looks to be in the right ball park. Air velocities look OK for longer port lengths with a conservative high pass filter but perhaps a bit marginal for shorter port lengths.
I did some WinISD sims and with the two Monacor ports I get a max air velocity of almost 38 m/s (123 ft/s) at 135W input, which is where cone excursion reaches Xmax. Will these ports chuff if the volume is cranked up?
Attached Images
File Type: png Screen Shot 2014-04-20 at 00.54.17.png (145.3 KB, 17 views)
File Type: png Screen Shot 2014-04-20 at 00.54.48.png (137.8 KB, 11 views)
  Reply With Quote
Old 20th April 2014, 10:03 AM   #118
diyAudio Member
 
Join Date: Apr 2005
Location: -
Quote:
Originally Posted by soren5 View Post
I did some WinISD sims and with the two Monacor ports I get a max air velocity of almost 38 m/s (123 ft/s) at 135W input, which is where cone excursion reaches Xmax. Will these ports chuff if the volume is cranked up?
Your cone excursion plot goes off the scale and so playing some film rumblings is likely to break the driver. For ported speakers what is normally done is to include a high pass filter to bring the cone excursion back to safe levels. This will reduce the maximum air velocity as will lengthening the port. Chuffing is also reduced by flaring the ends. Some measurements to help gauge what is likely to chuff can be found here and other places on the web.

If peak velocities are at, say, 20Hz and you are listening to conventional music then nothing is going to be driving the speakers to maximum SPL at this frequency. This is not necessarily the case with films and some non-conventional music.
  Reply With Quote
Old 20th April 2014, 02:04 PM   #119
diyAudio Member
 
Join Date: Apr 2005
Location: -
I seem to have chopped off the first part of my reply in an earlier post.

Quote:
Originally Posted by soren5 View Post
And that’s not an easy decision to make since I have no experience with waveguides. On the contrary I already know the SB29RDNC from my current speakers and I really like their sound, even off-axis.
I have read up on waveguides, trying to understand how they work and what benefits they have. Waveguides do indeed help to control the directivity of the higher frequency, which means better off-axis response. But waveguides also give a rise in the FR, the “waveguide hump”, which I assume you will have to deal with somehow? Zaph has a page discussing waveguides here. I have searched for opinions on the SEAS DXT and from what I’ve found people don’t seem overly impressed by it, see e.g. this PE thread.
Reading forum posts requires a degree of interpretation. There is good advice and there is bad advice. Sorting out which is which when the subject is relatively new to you can be difficult. Now there is criticism in the cited thread but there is also criticism of the criticism. How to weight the contributions?

The DXT was suggested because it was the simplest way to get a waveguide. If you do not believe a waveguide will bring significant benefits to the project then the DXT is unlikely to be a good choice. The use of sharp edges to aid the diffraction creates a small package but almost all high quality commercial speakers with waveguides have a larger smooth walled elliptical shape. There are reasons for this.

Quote:
Originally Posted by soren5 View Post
I have looked at the floor plan and speaker placement I tried to illustrate the horizontal off-axis response, see attached drawing showing 15* and 30* off-axis. If you believe off-axis response is a great concern for this project I guess there are other candidates to consider, e.g. metal dome tweeters such as Vifa NE19-VTA or SB26ADC. Again, I’m not sure how well either of those would integrate with the SB mid and woofer. The SB29RDNC+SB15NRXC30-8 OTOH is a tried and tested combination.
What does integration between tweeter and midrange mean to you? Does the tweeter+midrange in your 3 way project require the same design decisions as a tweeter+midwoofer in a 2 way? Think about efficiency, bass extension, cone displacement, directivity and the like.

The first design decision concerning tweeter/midrange integration is waveguide or no waveguide. The choices you would make after this are likely to be different. If you opt for no waveguide I am not sure why you are considering the tweeters you cite if you already like the tweeter you have used before.
  Reply With Quote
Old 20th April 2014, 07:36 PM   #120
jReave is offline jReave  Canada
diyAudio Member
 
Join Date: Oct 2012
The box program I use suggests that port noise is likely to become noticeable somewhere between 17 and 26m/s depending on flaring or no flaring. Your eg is up at 38m/s. Too high. If you really want to play loud, you may need to go with a slightly larger port diameter.

Port length with those Monacors will be fine as long as you are able to fine tune the LF with whatever active xo you choose to go with. Typically when tuning down this low, you might want to tune a little lower than usual to avoid boomy bass due to room gain so that means longer ports than those Monacors. So with those ports, you might actually have to pull the response down just a touch around 30 or 40Hz. You'll have to listen and/or measure to know for sure.

And if you can also use a HP filter at about 20Hz on the woofer, that will help to control those very high cone excursions down at those frequencies at large SPL's.

What I was sort of thinking was that you could actually use damping material around the outside of the speaker cabinets, perhaps something like this - BITUMEX/FG2 - BITUMEX FG 2 SK 245 x 325 x 2,0 mm - Europe Audio - to decrease resonances and sound transmission but given that there's another set of panels surrounding them, I'm not sure that it's 100% necessary. I'd probably do it if I could find something that wasn't too expensive. Otherwise a strip or 2 of something like felt all around the front perimeter to seal it off and then just small pieces elsewhere to prevent wood to wood contact might be sufficient.
  Reply With Quote

Reply


Hide this!Advertise here!
Thread Tools Search this Thread
Search this Thread:

Advanced Search

Posting Rules
You may not post new threads
You may not post replies
You may not post attachments
You may not edit your posts

BB code is On
Smilies are On
[IMG] code is On
HTML code is Off
Trackbacks are Off
Pingbacks are Off
Refbacks are Off


Similar Threads
Thread Thread Starter Forum Replies Last Post
To in-wall or not to in-wall, that is the question... head_unit Multi-Way 7 22nd January 2012 11:59 AM
Designing for on-wall/near-wall speaker placement morbo Multi-Way 4 27th November 2010 05:52 PM
in-wall/on-wall speaker? (axiom w-22 clone?) Scott_fx Multi-Way 0 22nd February 2007 03:08 AM
In-wall vs bookshelf spk flush with wall darascott Multi-Way 3 1st October 2005 04:16 PM
Opinions on wall to wall subwoofer array jmiyake Subwoofers 35 20th July 2003 12:26 AM


New To Site? Need Help?

All times are GMT. The time now is 12:06 AM.


vBulletin Optimisation provided by vB Optimise (Pro) - vBulletin Mods & Addons Copyright © 2014 DragonByte Technologies Ltd.
Copyright ©1999-2014 diyAudio

Content Relevant URLs by vBSEO 3.3.2