speaker build, should drivers be off center

I thought maybe he was referring to a time factor, when sound waves roll off the baffle and around the sides of the cabinet, on the outside.  I didn't consider that he might have meant the inside of the enclosure.  But I think you're right about the aesthetics and all the other speakers that are produced.  And I don't doubt these speakers will have plenty of little irregularities anyway.  ;)  concept sketch with 1/2" added to one side...I'd rather have the red stripe running through the center - meaning that the drivers are centered - at the top and just off center at the wider bottom part.  My original idea was to have that odd off-center look at the bottom anyway.

Z66_withtv.jpg

kfr01:  You wrote some very good reviews.  You have found your calling.  With your big kit build, how did you bevel the edges?  Table saw cut and sanding?

stevdart wrote:

---

I thought maybe he was referring to a time factor, when sound waves roll off the baffle and around the sides of the cabinet, on the outside.  I didn't consider that he might have meant the inside of the enclosure.  But I think you're right about the aesthetics and all the other speakers that are produced.  And I don't doubt these speakers will have plenty of little irregularities anyway.  ;)  concept sketch with 1/2" added to one side...I'd rather have the red stripe running through the center - meaning that the drivers are centered - at the top and just off center at the wider bottom part.  My original idea was to have that odd off-center look at the bottom anyway.

Z66_withtv.jpg

kfr01:  You wrote some very good reviews.  You have found your calling.  With your big kit build, how did you bevel the edges?  Table saw cut and sanding?

---

Stevdart:

On the road right now, but I can look up the answer to your off-center question in the loudspeaker cookbook by vance dickason when I get home on Sunday.  It actually has a relatively decent section explaining edge diffraction.  Stay tuned.

Kevin and Dan W.  of Exodus/Adire said they didn't offset the tweeter because it didn't make any practical difference with regards to final frequency response. Ideed, they tried it both ways, see the prototype here: https://www.diycable.com/main/product_info.php?cPath=24_92&products_id=539

I guess if I was starting from scratch and making specific left/right speakers, I would offset the tweeter.  The theory in Dickason's book seems to be good.

Beveled edges?  Cheapo ryobi router and the right 45 degree chamfer bit. 

Cheapest I found at the size I wanted:

https://www.routerbits.com/cgi-routerbits/sr.cgi?1140272577_5539+33

#2306.  It worked relatively well, but I'd invest in a large roundover before buying another chamfer.  A large roundover should reduce diffraction effects even more (in theory) and provide an easier edge to wrap veneer around.

Cheers!   Meeting w/ bank + 10 home showings yesterday.   20 home showings today.  Ahhhh! 

kfr01] wrote:

---

It actually has a relatively decent section explaining edge diffraction. Stay tuned.

---

The theory of offsetting drivers (especially high-frequency drivers) is to minimize edge diffraction effects.  Edge diffraction is a type of distortion caused by the high frequency sound waves moving at a different angle when they reach the edges of the enclosure.  It can be measured as a "ringing" or harmonic of the wavelength equal to the width of the baffle.  If the sound source is centered in the baffle, the ringing is more pronounced since it emanates from two sources (left and right edge) at the same time.  In severe cases it creates a "sound of the cabinet," a constant tone that never changes like a resonance, and can be very annoying.  By offsetting the tweeter two different edge tones are created, each at much lower SPL which is easier to get lost in the overall sound.

The same effect in woofers is called baffle step and creates a reduction in bass frequencies proportional to the wavelength of the baffle's width.  Reducing edge diffraction effects is the main reason for beveled or rounded cabinet edges, and for the strange shapes of some higher-end speaker systems.

Now while it can sometimes be detected and often times measured, is it important?  Most people cannot hear it as most times the SPL of the HF edge diffraction is low compared to the main signal.  Also, depending on the width of the baffle, the wavelength may be outside any critical musical areas or outside hiuman hearing (and unimportant) or right in the middle of the main information area (and completely covered up.)  In most cases, centering the drivers will not cause sufficient HF edge diffraction to be worth making them offset.  Bass baffle step diffraction, on the other hand, can cause a system to sound like it has too much mid range or weak or soft bass response, and often must be compensated for.  Offsetting the woofer can help reduce this effect, but most times you have to compensate by including a BSC (baffle step correction) filter in your crossover.  Or by increasing the low frequency efficiency (like using two woofers) to overcome it.  Or by making the baffle wider.

There is a more practical reason for offsetting tweeters in home audio speakers.  Since high-frequency sound is directional, it is sometimes useful to move the left tweeter further left and the right tweeter further right to help widen the sound field.  But again, while the effect may be measurable, it is not always significant.

Some people just like the look, of course.  This is also a legitimate reason for offsetting drivers.  I fall into this camp.    Bottom line: center your drivers unless you can measure or hear any cabinet ringing caused by edge diffraction.  Then try rounding the edges first to eliminate it.  Unless you like the way offsetting looks, then go for it.

As far as time alignment, that is a completely different thing and is corrected by aligning the acoustic centers from front to back rather than from side to side.  There are many papers written on this...

Thanks for the explanation, DYohn.  You make these things so much easier to understand when you tell it. 

I looked at Barry's Handy Freq-to-Wavelength chart to get an idea of how center-aligned MTM could cause problems.  This baffle, at 7" wide, looks like it might be a bit narrow to work with unless I take some precautions against these possible problems.  It looks like 7" wavelength is about 1900 Hz, half of 7" would be about 3900 Hz.  Multiples of 7" would hit at about 950 and 650 Hz.  The mids are crossed at 500 Hz but still outputting well below that, so I could see how their output might be affected a couple more places down the scale.

I will round over the edges for sure, and I think I should go ahead and do the offset and push the tweeter out to the outer edge, the mids toward the inner.  Doing that, there wouldn't be a whole lot of difference in the width of the baffle sides for either driver, but I'd rather take reasonable precautions since I'm scientific only to a degree in this hobby (testing is crude, for one)  ;).  My present MTM center channel is 7" wide and there was plenty of room for the offset that I used on it...and it sounds really good.  I'll copy that look on these mains and keep the driver alignment as it is on the center channel, just doing a rebuild of the cabinet as planned anyway.

Gotcha.  I wonder if Barry has a handy harmonics table ;) ?  The tweeter is about 4" total diam while the woofers width is 5", so they may be a bit too close in diameter to align like that.  centerchannel.jpg is the way I offset them in my current piece that so desperately needs a facelift (if I'd built this cab in 8th grade shop class I probably would have got my usual 'D' ;)  But, in this case, would that little bit of a center offset created by aligning the upper edge of these drivers as you described amount to enough baffle difference to make a change? 

(BTW, the offset look doesn't bother me and I'm using a paint scheme to throw the eye off anyway, with the bright red stripe aligned with the mids toward the innermost part of the cabinet.  My wife wants red and if that's all it takes to get the coveted SAF them I'm all for it!)

I'll use the offset variation in this current center channel config as the working model.

:)  The third harmonic is simply 3X the fundamental.  If your baffle is 7" wide, the fundamental is 1937Hz, so 3rd harmonic would be 5811Hz.  You might hear that as a shrillness or sharp sound that is present no matter what is playing.  IMO if you use a round-over on the edges of at least 1/2" you will minimize the effect as much as offsetting would probably do on that narrow a baffle.  Also, with a 4" and 5" OD, you will offset the center about 1/2", which is 1/2 the diameter of your soft dome.  From what I recall from the few cases where I actually read anything definitive, you want the offset to be at least 1X the OD of the dome... but don't quote me on that.  Oh yes, I remember your black and red design.  Looks great, and I like that offset.  I built a CC similar to yours, except I front ported the woofers under the tweeter. 

BTW, a 7" baffle might cause you some grief with baffle step losses.  Your baffle step will occur around 600Hz (meaning you will lose somewhere between 3-6db of all information below 600Hz) unless you push the speakers all the way against a wall, or include a BSC circuit.  I hate the Xover complexity introduced by BSC, actually, so I always try to increase bass output using porting or by attenuating the tweet so it is a little less efficient than the mid/woof so as to compensate.  Or by using EQ on the channel...

I think you might notice the results of flush mounting your tweeters and surface mounting the woofers (which will help with time alignment by pushing the tweet "back" in relation to the other drivers) a lot more than you would ever notice the possible edge diffraction.  Hell, align the centers and use a round over and be done with it! 

Audio is a very complicated thing, being full of small nuances that can be measured and are often discussed in the DIY croud, as well as in books like Mr. Dickason's.  Unfortunately these topics seem to get blown way out of proportion in their importance like what has happened here.  I mean, to say that "A really well designed crossover is of no use if........[it] has drivers mounted equidistant from two or more edges."  is like saying that a bug smear at the bottom left of your windshield makes driving impossible.  If you are lucky enough to see it, is it really going to effect your driving that much?  That's one of my pet peeves about the Loudspeaker Design Cookbook is that it discusses quite a few of these ideas, but doesn't measure their importance or audibility.  I've been trying to talk Dan Wiggins into writing a similar book as he's so good at explaining the technical stuff so even the beginner can understand (something the LDC could use some help with). 

Oh, I totally agree.  Just as long as they know whether it is worth the cost (being time, money, or cosmetic in nature).  That's why I like to see an accurate description of how important these phenomenon are, to prevent people from taking it too far and getting into stuff like the speaker wire tripods, or jars of acoustic rocks.

Also, it will provide people with a better foundation to make better compromises in their design.  For instance, one popular topic in the DIY world is to use shallower crossover slopes to reduce phase changes (something else from the Cookbook).  But as Dan has argued in the past, this produces other much more audible problems that must be dealt with.  Especially since the phase issue the shallower slope is correcting for probably couldn't be heard unless you listen to sine waves on the system.

item to scratch from my materials list:  jar of acoustic rocks...

All you gentlemen gave valuable feedback on this topic and now I have a much clearer perspective going forward. 

Ah, and one last thing stevedart: make sure YOU are not off-center when you start cutting MDF! 

Oh, and those acoustic rocks must be cryo-treated then activated under a crystal pyramid during the spring equinox before they will work.

Looks like this was wrapped up prior to me getting back from home hunting land. 

Seller accepted our offer.  Close date April 28th!  Now we just need to clean + sell our condo. 

A basement room of 24'x13'   It isn't quite the right dimensions for a big HT room.  A little long and narrow.  But with the right treatments might make an O.K. listening room.  Important thing is that the wife likes the rest of the home.

p.s.   I would buy Dan's book in a heartbeat. 

Congrats on choosing a house!  It sounds like you've worked out the model for negotiations with your wife re: SAF from early on...so you should be good to go. 

My HT room is similar to yours, even a little narrower.  I found that working the sound system into it worked out well, sub is 1/3 across the end wall and under the TV, rear surrounds are 2/3 back and the last back surrounds-to-be will go in the rear corners.  It's the seating in such a narrow room that presents the challenge.  Do you furnish it like a theater, two rows in tiers, which is probably the easiest to do but limits the way you can use the room.....or do you find just the right size, style and mix of comfortable chairs so that they can either all face the screen or work as a social entertainment area?

My house has limited space so I have to use this room for all of the above.  When I moved the viewing toward the end wall, the traditional furniture pieces suddenly were the wrong size for the room.  Eventually, I'll do a furnishings makeover with slimmer individual chairs that can semi-recline, swivel and have headrests.  And maybe a lower futon-type comfy thing in the front for the kids.  A small bar-type arrangement with stools of the right height will probably occupy an area behind the main seating.  It'll take me a while but it'll get there eventually.

I'm in line for that book, too!

Good idea.  The fund will be started when I get the tax return, which gives me extra incentive to get the taxes done!  (...and those are some really nice looking drivers, too.)

This is funny...I had that PartsExpress page loaded that shows that Vifa midwoofer.  The price jumped from $4 to $14.25 within the last day, but this item has been out of stock, never to return, for months!  Go figure.

stevdart wrote:

---

This is funny...I had that PartsExpress page loaded that shows that Vifa midwoofer.  The price jumped from $4 to $14.25 within the last day, but this item has been out of stock, never to return, for months!  Go figure.

---

I bet it's because they are sold out, so the web price reverted to the normal list.  By the way, Vifa is no more.  From now on, what used to be Vifa will be sold as Peerless V-line.  Price changes remain to be seen...