Feedback on drivers for 3-way bookshelf - Page 5 - 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 17th June 2013, 04:38 PM   #41
diyAudio Member
 
Join Date: Jun 2007
Define the Project -

- Five speakers in a 5.0 system for both music and movies.

- Bookshelf Front with a single 6.5" bass driver in a 3-way configuration. Front Ported or Sealed; ie: no rear facing ports.

- Center Speaker - One or two 5" or 6.5" bass drivers, 3-way configuration. May or may not be ported, unknown at this point.

- Surround Speakers - 5" or 6.5" bass drivers (unknown), 2-way or 3-way (unknown), any special configuration (unknown), On-wall, in-wall, Di-pole/Bi-pole, on stands (unknown)

- Room size (unknown); Room dimensions (unknown); placement of equipment within the yet to be defined space (unknown)

By actually clearly defining the system, we see what we know and what we don't know.

As to the speaker literally being up against the wall, that is both unnecessary and unwise regardless of the placement of the ports. Most bookshelf speakers, depending on how much bass they have, need to be roughly 6" to 8" from the wall behind. In some cases 10" wouldn't hurt. Most floorstanding need to be 10" to 12" from the wall, and in some bass heavy speakers, 18" to 24" is necessary, though not for most speakers.

Only in the rarest of circumstance do speaker literally need to be against the wall, and even rear ported bookshelf don't need huge space behind them. If you get them any distance from the wall, they will mellow out and sound fine. Again, 6" to 10" for most bookshelf is sufficient, and achieving that is not as difficult as you have convinced yourself it is.

I had a pair of JBL Venue Stadium speakers with TWIN 8" woofers each in a 3.5-way system. Because the bass was already prominent, the speakers were just barely functional 12" from the wall with the rear ports plugged with Foam. I moved them to a new location, and with substantial room behind them, they settled down nicely, I was almost reluctant to sell them.

I generally find that as the speaker is closer to the wall, the bass increase, but it does so at the expense of mid-range clarity, which is something you don't want to sacrifice.

I've also found that literally right up against the wall is unnecessary. In my case, I move the speaker even with the front of my equipment cabinet, and that gives me the 10" to 12" I need for my floorstanding speakers. Many people have their speaker right along side the equipment cabinet, and are OK with the equipment cabinet reaching a certain distance into the room, but despite being side-by-side, they are not OK with the speakers being on the same front plane. I fail to see the logic.

So, with the limited (best guess) information we have, we assume -

- 6 woofers of various or the same size
- 3 or possibly 5 Mid-range Drivers
- 6 Tweeters

Given that you are having the cabinets and crossovers custom built, I would say 1/3rd to 1/2 of the budget would go to drivers. With a budget in the $3500 to $5000 range, that means you have between $1167 and $2500 for drivers.

Though you should probably stay in the 1200 to 1500 range. At least until you see what is possible. You need 17 drivers in the largest configuration, which means about $70 to $90 per driver. Though in reality, you can adjust individual drivers up or down as needed to fit the design. This just establishes perspective.

That should at least give you a frame of reference to get you started.

Now we have defined the problem. I would say the next step is to consider physical reality. Estimate the cabinet volume, figure out rough dimension, and see if you can actually get the drivers to fit in that box - woofer, mid-range, tweeter, plus port. Draw it up in a proportional sketch, and see the physical reality. Physical reality might dictate that the port has to be on the back. If you can make it work - fine, if not - then redesign.

Just a few thoughts.

Steve
  Reply With Quote
Old 17th June 2013, 07:55 PM   #42
jReave is offline jReave  Canada
diyAudio Member
 
Join Date: Oct 2012
Quote:
Originally Posted by rick240 View Post
The only way that could really work is with an AT screen. Any other way and you'd have to raise the screen too high. I think the compromise of a well designed centre (hence all of this effort) will make for a better HT experience than having the screen a few feet higher than it should be.
Agreed, 100%

Quote:
Originally Posted by rick240 View Post
The Seas are really nice - I'm just a bit concerned about the sensitivity. 2 together just might do the trick.
Ok, let's talk sensitivity, xmax, xmechanical and max power input for a second.

Sensitivity = SPL with 1 watt at 1 meter at, I believe, 1000Hz.

So a driver with 86dB sensitivity does reach 105dB at about 75W (actually I took a guess, 79W is correct) at 1 meter and with no regards for amp headroom. Now move back from the front of that single speaker another 2 m and you're looking at about 95dB with those same 75W. I'm using this calculator Amplifier Power Required Calculator. Now, 95dB is LOUD. If you have an SPL meter, try playing with your volume control and measure from your listening position. For normal listening I don't need anything louder than 85-90dB and even that is loud (no capitals this time ).

xmax = the maximum distance (1 way) the voice coil travels within the gap. Outside the gap, magnetic forces decline, the cone is not controlled as well and distortion occurs.

x mechanical = the maximum physically distance the voice coil/cone can travel. Usually more than xmax but varies with the driver.

max power = the max wattage that the voice coil can electrically handle. Pretty straight forward.

Applying these to your choices, let's start with the Audax. Clearly, you want to keep distortion to a minimum, so you don't want to exceed xmax. First graph below shows the FR of the Audax in 2L with 30W. SPL at 1 m is 107.7dB. Left of the red line shows the frequencies at which xmax is exceeded. The limit here is the point at which the lines intersect, about 400Hz. Using a steep xo (3rd or 4th order) at 500Hz should hopefully diminish output below that point enough so that xmax is not exceeded. Second graph shows the same thing, peak excursion, in a different way - below the red line=good; above=bad. Again, point of intersection is about 400Hz.

So, this effectively limits your speaker to a max SPL of about 107dB at 1 m. Maybe 105 to be more on the safe side. Happily 30W is also below the Audax's max power of 40W. Of course if you move the xo up to say, 600Hz, that should allow a little bit more SPL. But if you also want to keep the xo at the baffle step point (not necessary, but helpful), now the baffle needs to be a little narrower- closer to 7.5" for 600Hz instead of 9" for 500Hz. And that makes a difference to your woofer selection.

Next graph is the cone excursion of 2 Seas in 16L, sealed with 76W and SPL of 106dB (1 m). xmax is exceeded at about 66Hz but it will be crossed at 80Hz so this is probably about your max SPL here with these. X mechanical though has an additional 5mm each way.

The 4th graph is 2 Discoveries in 24L, sealed with 38W and SPL of 106dB. Again, xmax is exceeded at about 66Hz. X mechanical has an extra 4mm.

So, the Seas and Discoveries look about the same in terms of max SPL. It's how you use your mid however that's going to limit how loud your speakers will go. But it's your woofer choice that's going to determine how much power your amps are going to need to get you there. And if you want some amp headroom for the peaks at this level then the advantage goes to the Discoveries. A 150W amp will have 3dB headroom with the Seas and 6dB of headroom with the Discoveries. The Scans will need a larger box though and the 9" wide baffle.

Have to go out right now but I have 2 more woofer suggestions to post later.
  Reply With Quote
Old 17th June 2013, 08:06 PM   #43
jReave is offline jReave  Canada
diyAudio Member
 
Join Date: Oct 2012
Actually, just quickly here are 4 distortion measurements from Zaph. Three are 180mm drivers and 1 is a 15mm. Which would be your 1st choice from 80 to 500Hz?
Attached Images
File Type: gif Scan-Speak-18W8434G00-HD.gif (21.8 KB, 51 views)
File Type: gif Scanspeak-15W8530K00-HD.gif (20.5 KB, 8 views)
File Type: gif 8945A-HD.gif (22.8 KB, 7 views)
File Type: gif 18W8531G-HD.gif (22.8 KB, 7 views)
  Reply With Quote
Old 17th June 2013, 08:11 PM   #44
diyAudio Member
 
Join Date: Jun 2013
Wow - thanks for the effort and education!

Quote:
Originally Posted by jReave View Post
Agreed, 100%



Ok, let's talk sensitivity, xmax, xmechanical and max power input for a second.

Sensitivity = SPL with 1 watt at 1 meter at, I believe, 1000Hz.

So a driver with 86dB sensitivity does reach 105dB at about 75W (actually I took a guess, 79W is correct) at 1 meter and with no regards for amp headroom. Now move back from the front of that single speaker another 2 m and you're looking at about 95dB with those same 75W. I'm using this calculator Amplifier Power Required Calculator. Now, 95dB is LOUD. If you have an SPL meter, try playing with your volume control and measure from your listening position. For normal listening I don't need anything louder than 85-90dB and even that is loud (no capitals this time ).

xmax = the maximum distance (1 way) the voice coil travels within the gap. Outside the gap, magnetic forces decline, the cone is not controlled as well and distortion occurs.

x mechanical = the maximum physically distance the voice coil/cone can travel. Usually more than xmax but varies with the driver.

max power = the max wattage that the voice coil can electrically handle. Pretty straight forward.

Applying these to your choices, let's start with the Audax. Clearly, you want to keep distortion to a minimum, so you don't want to exceed xmax. First graph below shows the FR of the Audax in 2L with 30W. SPL at 1 m is 107.7dB. Left of the red line shows the frequencies at which xmax is exceeded. The limit here is the point at which the lines intersect, about 400Hz. Using a steep xo (3rd or 4th order) at 500Hz should hopefully diminish output below that point enough so that xmax is not exceeded. Second graph shows the same thing, peak excursion, in a different way - below the red line=good; above=bad. Again, point of intersection is about 400Hz.

So, this effectively limits your speaker to a max SPL of about 107dB at 1 m. Maybe 105 to be more on the safe side. Happily 30W is also below the Audax's max power of 40W. Of course if you move the xo up to say, 600Hz, that should allow a little bit more SPL. But if you also want to keep the xo at the baffle step point (not necessary, but helpful), now the baffle needs to be a little narrower- closer to 7.5" for 600Hz instead of 9" for 500Hz. And that makes a difference to your woofer selection.

Next graph is the cone excursion of 2 Seas in 16L, sealed with 76W and SPL of 106dB (1 m). xmax is exceeded at about 66Hz but it will be crossed at 80Hz so this is probably about your max SPL here with these. X mechanical though has an additional 5mm each way.

The 4th graph is 2 Discoveries in 24L, sealed with 38W and SPL of 106dB. Again, xmax is exceeded at about 66Hz. X mechanical has an extra 4mm.

So, the Seas and Discoveries look about the same in terms of max SPL. It's how you use your mid however that's going to limit how loud your speakers will go. But it's your woofer choice that's going to determine how much power your amps are going to need to get you there. And if you want some amp headroom for the peaks at this level then the advantage goes to the Discoveries. A 150W amp will have 3dB headroom with the Seas and 6dB of headroom with the Discoveries. The Scans will need a larger box though and the 9" wide baffle.

Have to go out right now but I have 2 more woofer suggestions to post later.
  Reply With Quote
Old 17th June 2013, 08:57 PM   #45
diyAudio Member
 
Join Date: Jun 2007
Here are some bits of random information that you might find helpful.

I did a test on my system to determine the volume difference between my CD and my Turntable; the CD was louder, but by how much?

For What It's Worth - LP vs CD SPL | AVForums.com - UK Online

With a 100w/ch amp and speakers with TWO 8" bass drivers each in a 3.5-way configuration, I was able to reach a 100dB average level at 1 o'clock on the volume dial. That is LOUD in my room, but not really straining the amp to excess. (See post #13 w/ photo of my system, only ONE set of speakers used during this test.)

Here are a few other threads that might be generally helpful. Mostly they relate to manufactured equipment, but the principles apply to any case were you are acquiring new equipment, whether bought or built.

Primer: How much do I have to spend to get a good System?

Primer: Acoustics - Absorption & Diffusion

Of more minor but still valuable interest -

Quality Dry Speaker Stand FILLER - CHEAP

Interactive Frequency and Sensitivity Charts

A discussion of exactly what constitutes Bass, Mid-Range, and Treble -

Bass, Midrange, and Treble? What?

Again, this is more general information stuff, but you might find it helpful in evaluating where you are going and what you are doing.

Here is another hand chart that shows the Power Distribution across the Frequency spectrum. For orchestral music, the peak power band is between 250hz and 500hz.

Click the image to open in full size.

This can be a useful guide in determining relative power of various drivers, crossover points, and other useful things. Notice above 2000hz, you have 1/10th the power you have at 355hz.

I think you are getting good technical advice from others.

Steve

Last edited by BlueWizard; 17th June 2013 at 09:01 PM.
  Reply With Quote
Old 17th June 2013, 11:19 PM   #46
AllenB is offline AllenB  Australia
diyAudio Member
 
Join Date: Oct 2008
Quote:
Originally Posted by BlueWizard View Post
I generally find that as the speaker is closer to the wall, the bass increase, but it does so at the expense of mid-range clarity, which is something you don't want to sacrifice.
I'd agree with this. You either need to join a wall or move away from it. At the least you could use columns of absorbent beside the speakers against the wall or build wings onto the sides that chamfer toward the wall.
  Reply With Quote
Old 18th June 2013, 12:56 AM   #47
jReave is offline jReave  Canada
diyAudio Member
 
Join Date: Oct 2012
I'm going to preface this post by saying that I hesitate to rely solely upon harmonic distortion graphs but without hearing drivers personally, I don't think recommendations can be made without some kind of frame of reference. And so far people aren't jumping in with personal experiences re these drivers.

So in post #43, the graphs go like this: Scan Discovery 18W8434G, Scan Revelator 15W8530K00, Usher 8945A and Scan Revelator 18W8531G.

I posted the 15W8530K00 because that's my frame of reference. That's what I listen to from 80 - 2800Hz and can recommend almost unreservedly. My 1 caveat is that's it's only a 6" driver, so I find the LF (my guess is 80 - 200Hz) just a tiny little bit thin (I cross to a sub at 80), but a single cone with less than a 5" diameter can only push so much air. Otherwise, it's a superb driver as I mentioned before.

So looking at it's graph more closely, 3rd order distortion stays below -70 from about 150 to 1200Hz. Fifth order almost mirrors this a little lower but with a bump at 120 - 200Hz. Second order is higher but is considered more benign.

Looking at the 7" Revelator, its graphs have always confused me a little. Everyone raves about its sound (well, almost everyone), yet below 200Hz its 2nd order distortion does not stay below -70. I've heard it said that we don't hear distortion in the LF quite as well, so maybe this has something to do with it. Or perhaps it's everything taken as whole. Or perhaps it's over-hyped.

In terms of the Discovery, 3rd order is only below -70 from about 125 to 350Hz altho 2nd order is also nice and low in the 150 - 700Hz range. Rick, for your purposes, I'm now thinking this is inadequate. This fails to meet the benchmark I'm trying to set with my 15W's.

So, another ScanSpeak that is well regarded is the 18W8545K http://www.solen.ca/pdf/scan/18w8545k.pdf, $172 at Solen but they only have 2 in right now. Also, sorry no distortion measurements. Sensitivity is 87dB.

However, the Usher 8945A is virtually the exact same driver for $125 (that is from the States tho so it'll cost us a little more) Usher 8945A 7" Carbon Fiber/Paper Woofer 296-600. Not sure if that's getting too expensive for you but it might be the best 180mm performer I've found from 80 to 800Hz. Sensitivity of 2 in parallel after 4dB BSC will be about 88dB and 2 of them sealed in 13L, Qtc of .703 gives an f3 of 75Hz and max SPL of 110dB. Third order distortion stays below -70 from about 80 to 800 Hz, 2nd order below -60 all across the FR and now that's better than my 15W's. So according to my criteria, highly, highly recommended.

Lastly, here's the SB Acoustics 17MFC35-8, $64 http://www.solen.ca/pdf/sb/sb17mfc358.pdf. I really like what this company is doing. Value is amazing. Sensitivity of 2 in parallel after 4dB BSC will be about 89dB and 2 of them sealed in 18L, Qtc of .702 gives an f3 of 70Hz and max SPL of 107dB. Third order distortion stays below -70 from about 160 to 1200 Hz, which is about exactly the same as my 15W but also with lower 2nd order all across the usable range. Totally comparable to my 15W's but just a little poorer distortion in the bottom octave, 80 to 160 Hz. But value is hard to beat. And good sensitivity. Driver width is also only 171mm. SB makes the Satori too btw, which I highly recommended before but isn't really good for your application. Quality trickle down and all that.....

You might also want to think of this. If your surrounds are much closer to you than your front channels, you may just be able to get away with a single woofer in these. Same timbre but they just won't play quite as loud which they may not need to. That could save you a couple of hundred bucks, altho the way I look at these things is in the long run, a couple of hundred bucks really doesn't matter over the life span of the speakers.

Maybe some others can comment on any of these selections too.

Graphs: Rev15W, Rev18W, Scan Discovery, Usher 8945A, Usher 17MFC35-8
Attached Images
File Type: gif Scanspeak-15W8530K00-HD.gif (20.5 KB, 2 views)
File Type: gif 18W8531G-HD.gif (22.8 KB, 3 views)
File Type: gif Scan-Speak-18W8434G00-HD.gif (21.8 KB, 3 views)
File Type: gif 8945A-HD.gif (22.8 KB, 5 views)
File Type: gif SBAcoustics-SB17MFC35-8-HD.gif (22.7 KB, 7 views)
  Reply With Quote
Old 18th June 2013, 03:56 PM   #48
diyAudio Member
 
Join Date: Jun 2013
Thanks for all of the comments there. Where does the L16 fit in there?

And with the Seas, I'm trying to understand the difference between L16RN-SL and L16RNX.

Quote:
Originally Posted by jReave View Post
You might also want to think of this. If your surrounds are much closer to you than your front channels, you may just be able to get away with a single woofer in these. Same timbre but they just won't play quite as loud which they may not need to. That could save you a couple of hundred bucks, altho the way I look at these things is in the long run, a couple of hundred bucks really doesn't matter over the life span of the speakers.
I'd always been thinking a single woofer for the rears. That's had me thinking an 8ohm for the fronts (two in parallel giving 4 ohms) and a 4ohm for the rears (typically higher sensitivity) so that the mid and tweeter aren't excessively padded in the rears.

Talking about padding - does it impact power handling?

With the Audax mid, it has 93db sensitivity, which if the woofers end up at 88db would need a lot of padding. Although, there is the C0 model (as opposed to the Z0) which is under the 4" woofers at Madisound instead of 4" midrange, and it shows lower sensitivity (so less padding).
  Reply With Quote
Old 18th June 2013, 05:54 PM   #49
jReave is offline jReave  Canada
diyAudio Member
 
Join Date: Oct 2012
Quote:
Originally Posted by rick240 View Post
Thanks for all of the comments there. Where does the L16 fit in there?
Sorry, the L16 and Discovery distortion responses are very, very similar so I just posted the 1, but I guess the Seas might have very slightly better 3rd and 5th distortion from 80 to 500Hz. Advantage in 2nd order goes to the Scan. Graphs below: Seas then Scan. Again, not having heard these I'm trying to find a driver with similar measurements to what I have heard and can recommend. Both of these are close but fall short of the Scan 15W.

Quote:
Originally Posted by rick240 View Post
And with the Seas, I'm trying to understand the difference between L16RN-SL and L16RNX.
L16RN is good in both ported or sealed. L16RNX is designed to work better in smaller, ported boxes. (Man, I so want to get you into something like Unibox. You'll learn so much thru the process of playing around with all of this stuff.)

So, the RN hits 67Hz in 8L with heavy stuffing (not absolutely necessary, but my preference). The RNX-8 hits 78Hz in 5.4L but only if you stuff it lightly. The RNX-4 only hits 86Hz in 8L with light stuffing. All boxes sealed with Qtc close to .7 (which is desirable). RNX-4 will go lower ported in about 8L.

Quote:
Originally Posted by rick240 View Post
..... and a 4ohm for the rears (typically higher sensitivity) so that the mid and tweeter aren't excessively padded in the rears.
That's a good idea. I hadn't thought of that. But I would think that for what you are looking for, sound quality would be the 1st priority. Still, all drivers mentioned so far except the Usher also come in 4ohms

Quote:
Originally Posted by rick240 View Post
Talking about padding - does it impact power handling?

With the Audax mid, it has 93db sensitivity, which if the woofers end up at 88db would need a lot of padding. Although, there is the C0 model (as opposed to the Z0) which is under the 4" woofers at Madisound instead of 4" midrange, and it shows lower sensitivity (so less padding).
Quick answer is no - power handling of the driver stays unchanged. The resistors just eat up a certain percentage of the power before it hits the driver and so it doesn't play as loud until you feed it more watts. Tweeters and mids in 3-ways are padded down all the time, sometimes quite a lot. I think the sensitivity of my speakers is about 82dB which means the tweeter is padded down about 10dB. Again, it just becomes a consideration of how loud you want them to play and how much power you have to play with. Altho, I'm a very practical kind of person and I hate to waste things so I'm with you in trying to keep the sensitivity high. But it wouldn't be my 1st priority.

Actually, now that I think about it, there are 2 advantages to higher sensitivity. Less power in the voice coil means good electrical things like less heat generated and other stuff that's beyond me. And the cone also has to move less for a given SPL which means less inter-modulation distortion. Very subtle again so good if achievable but still not my 1st priority.

Also, stick with the Aerogel cone. To my knowledge it's got more magic in it for the midrange than the carbon fibre. Moving mass 25g vs 69g for example.
Attached Images
File Type: gif Seas-L16RN-SL-HD.gif (20.1 KB, 6 views)
File Type: gif Scan-Speak-18W8434G00-HD.gif (21.8 KB, 2 views)
  Reply With Quote
Old 18th June 2013, 06:03 PM   #50
diyAudio Member
 
Join Date: Jun 2013
Quote:
Originally Posted by jReave View Post
Man, I so want to get you into something like Unibox. You'll learn so much thru the process of playing around with all of this stuff.
That could be a next step - but not for a couple of weeks (studying hard for a wine exam, and too much work to do in my software company - sigh).
  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
How the negative feedback really works and an alternative feedback question Amit_112dB Solid State 37 21st April 2014 03:16 PM
Modify amp from voltage to current feedback for low Qts drivers nicotin78 Solid State 13 22nd March 2013 12:28 PM
Feedback on wholesale drivers Thims Swap Meet 9 14th May 2008 10:17 PM
'Local Feedback and Nonlinearity of Multistage Feedback Amplifiers' mikeks Solid State 62 20th July 2005 08:03 AM


New To Site? Need Help?

All times are GMT. The time now is 03:04 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