A few stupid speaker questions... (1 Viewer)

[Chuck Bogie]

Okay - I'm going to stick with the same two main types of drivers... Both shielded - leaning sorta toward some 7" Vifa woofs and some Morel tweeters... But will probably change my mind at least 3-4 more times... Right now I'm figuring that "bigger is better." Speakers must be shielded, of course...

Why do folks use two smaller woofers rather than one larger one? Can you make the box smaller that way? Or go lower? Do they set them up as "three ways" with one woof acting as a mid and the other going lower? How do you calculate the box size for using two?

I am thinking myself of using two woofs/speaker, simply because the center will have two - With the center, should I line everything up horizontally, or should I "triangle" 'em with the tweeter above 'em? Or is there some esoteric formula at work here?

Likewise, I'm wondering - I've seen speakers with both vertically aligned woof/tweet/woofs and vertically aligned tweet/woof/woof combinations... Is there an advantage to one type?

Now, where to put the port... I see that I essentially have 18 or so options (grin...). I'm narrowing it down to toward the bottom of the speaker, but wondering what different effect it'll have pointing toward the front vs. the rear...

From what I've been reading, crossovers are almost as important as speaker choice and box construction - But I -seriously- lack the skills to design/build my own - what are my options?

 

[Dan Wesnor]

Actually, these are good questions.

From what I've been reading, crossovers are almost as important as speaker choice and box construction - But I -seriously- lack the skills to design/build my own - what are my options?

Leap services from Madisound. Not the best, but passable. Make sure you tell them you want baffle step compensation (if they ask how much, 4dB is a good generic number). They will design an XO for any drivers they sell for not much $$, I think $25 for a 2-way?

 

[Aaron_Smith]

It is generally thought that rear-facing ports are better because midrange leakage is not in the direction of the listener. There's an exception to every rule, of course; if you need to put the speakers right up against a wall, for example, it may be better to do front ported.
I would highly recommend reading the Loudspeaker Design Cookbook before attempting your own speaker design.

 

[Chuck Bogie]

Okay... This is getting interesting...

So I can essentially get a pair of woofs that'll run from 50 to 4,000 or so, use a 3-way crossover, and run the lower one from 50-500 or so, the upper one from 500 to 4,000, and then the tweeter will irritate the local puppies?

What about the center? I suspect that there I'll have to put a tweeter in between two woofs? Should it be in between 'em, or a little above or below?

Have any of y'all tried the pre-built crossovers that Partsexpress sells?

Definitely gonna get the cookbook - I'm also wondering about how to figure out box size, given use of two woofs - I'm guessing that simply doubling what I saw in the PE catalog probably won't be the way to go...

Semi-tempted to try to construct some essentially two-part towers, with the signal going through a PE plate amp for the extreme low end before hitting the speaker crossover...

 

[Anthony_Gomez]

In a TMM: (using generic XO points)
the tweeter goes from 3k-on up
1 mid/woof goes from 3k on down
0.5 woof goes from 500hz on down


generic XO's will give mediocre sound at best (even with the best drivers).

a center channel MTM on side has a very limited horizontal dispersion. this is partially solved with 3rd order XO's. proper centers are going to usually be a TM or a WTMW like with the audax HT kit.


from what I am reading, you know very little about speakers (which is perfectly ok) and you want to dive head first. If you really want to design your own, you should read through the cookbook first to see if you really want to go that route. Designing your own is more than numbers in a calculator. you need to get measurement software, mic and pre-amp.

if in the end you are possitive you want to build your own, here is the BEST advice I give any newbie.
Choose a project that used drivers from a well established kit. If your XO doesn't work, then you can atleast still easily use the drivers

(FYI, XO's are VERY dependent on how big of a baffle you have the drivers on AND WHERE on the baffle the driver sare)

 

[Patrick Sun]

I would echo the sentiment that novices should stick to building a proven speaker kit, and then as the absorption of the material and the understanding as to what really goes into making a good sound speaker increases, only then would I suggest that someone should try to build their own speakers from scratch.

If you don't have access to measurement equipment and crossover modelling software (either by finding someone who does, or buying the necessary gear and software), and the know-how in using the tools, you'll be at a huge disadvantage in building a smooth sounding speaker that meets your listening requirements/criteria.

You'll need to understand how to use the driver's T/S parameters to model the enclosure size needed, and what to expect with a given driver's specs in terms of performance, and its intended frequency range of operation so that you don't ask each driver to do more than it's capable of doing without straining itself and adding distortion to the final output of the speaker.

You'll have to get a handle on integrating baffle step compensation so that your speakers don't sound "thin" due to the loss of bass from 500Hz and below.

You'll have to look out for impedance swings, making sure the overall speaker present an impedance profile that is friendly to the amp powering it. Coming up with a speaker that presents a very consistent load to the amp is a desirable quality.

You'll have to check out the off-axis response of the speaker output throughout the audio frequency band, and design for good dispersion. Different crossover topologies help in this area.

You'll have to deal with phase tracking of the drivers to prevent delays and nulls in the response.

And you'll also have to worry about smoothing out the frequency response, and deal with peakiness if present using traps to smooth things out.

Crossovers can make or break a speaker, even if you use the best drivers in the world, a poor crossover will rob the speaker of its full potential. A great crossover design and make decent drivers sound good provided enough attention is paid to the shortcomings of the drivers.

Building a speaker is not just building a box, popping in some holes for the driver, slapping on a "pre-made" crossover, and praying to the deity of your choice for a good sounding speaker. It might happen, but the odds are long and hard.

 

[Chuck Bogie]

Well, I guess this is out then, since I'm not the world's greatest at assembling circuit boards. I thought that one could locate pre-built crossovers that would work...

 

[Anthony_Gomez]

it is really easy to do. There isn't any tricky or "fine" soldering involved.

Some DIY kits will come with pre-assembled XO's.

 

[Dan Wesnor]

I quit soldering a long time ago. Twist the wires together and cover with electrical tape. No problems.

The pre-built crossovers from PE or anyplace else WILL NOT WORK. They will provide much less satisfactory results than from using a real design.

 

[Stephen Dodds]

Having just spent many pleasant hours making my first set of a DIY speakers a custom design, I would strongly agree with choosing a kit.

There are many fine sounding kits that would undoubtedly sound better than you could currently design atthe moment.

If you still want to charge ahead, I would equally strongly suggest using an active crossover such as the Behringer CX3400 or even better the Behringer DCX2496. These will give you a choose of crossover points, and the latter a choice of slopes, EQ and a bunch more.

The downside is that you will need six channels of smplification for a three way, and the DCX costs $350.

If you can't afford this, steal or buy someone elses design.

Steve

 

[Henry_W]

Chuck -

Don't be dissuaded to try. While all this advice is true, it is a little bit like golf - You probably won't start out as real good at the game, but unless you start you won't know how much you like it.

Lots of folks tried to tell me my first project was too ambitious. In many ways they were right, but through patience and the good education from here and other places I was able to forge ahead. My advice:

1. Listen, learn and read
2. Use proven design software (Boxplot, Unibox, etc.)
3. Remember - all designs are compromises - your setup, room, drivers, asthetics, etc, all play into the final product. Learn where you are willing to accept those 'less than perfect' issues and design accordingly.
4. Forge ahead...

Regardless of your inexperience, you may find you like your results - I most certainly did and I am now officially hooked. Good Luck...

 

[Henry_W]

BTW Steve - Great website and sharp wit...

 

[Chuck Bogie]

Well, I'm not going to be able to take the time for the electronics courses, etc., to learn how to design a crossover, and I don't want to spend more than what the dang speakers will cost to buy the electronics testing stuff, and all that stuff, then it seems I'll need to build and discard several crossovers before I get it right, just to make between one pair or a half-dozen speakers.

$700 worth of crossovers for a $300 pair of speakers? (my original estimate of $100/speak has already started getting some creep as I read - thinking 3-way/D'appolitto instead of 2-way, but that is just ridiculous).

I know woodworking. And I know CAD. I'm a graphic designer by trade. I _thought_ that I'd be able to buy a few speakers, such as the Vifa MG speaks, and a crossover that'd work, and then, with aid of maybe one book, figure out the volume/shape of the box for 'em.

I don't want to do a kit. I have the tools and capability to build my own box, thankyouverymuch.

 

[Anthony_Gomez]

Chuck, a kit does not mean that you can't build your own box.

If you decide to choose drivers and slap together an XO, you might as well go to your local best buy and get a KLH speaker.

If you want to do the XO right (depends on EVERYTHING) prepare to spend a chunck on money between measurement stuff and trial and error parts/enclosures...AND plan on spending a substantial amount of time learning the software, and all the many complexities of XO design.

Not trying to discourage you, but to prepare you for what you WILL have to face if you want to do this completely on your own...and right.


yourwelcomeverymuch.

 

[Brian Fellmeth]

What you want is a speaker kit project, but you have misunderstood the term "kit" in this context. What we mean by "kit" is that the very difficult crossover and driver integration work has already been done by patient, knoledgable, well equiped individuals. The "kit" is just a recipe of driver choice, crossover circuit, and cabinet volume/alignment. You are still free to design and build your cabinet any way you like (withen reason). Just make sure the internal volume ends up close to the designer's specification, and that there are no major changes in the driver position and baffel shape. You seem very concerned about being able to solder together the crossover components (I DO NOT recommend Dan's idea of relying on twisted and taped connections). This is really easy- the people here will help you and a soldering iron is cheap.

 

[Chuck Bogie]

I already own a soldering iron - That's how I installed the stereo in my truck. And lemme tell you, it wasn't pretty. Thank Bob for shrink wrap. But the connections are still there...

These "kits..." So they're basically just recipes - Take two Acme woofs, an Epitome tweet, stir well, then bake in a 1.56 cubic foot pan for 20 minutes, stick a port in it, and it's done?

So, what I'm looking to do is make something that sounds a bit better/fuller than my existing Infinity Entra Two cute widdle towers and Entra One center... Any suggestions as to which sorta brownies I oughta bake?

 

[Aaron_Smith]

In my opinion a good speaker kit lets you do all the fun stuff involved in building a speaker (cabinet construction and finishing, crossover assembly, and LISTENING) without all the not-so-fun stuff (crossover design with 82 unsuccessful iterations, testing and measurement, etc.) I say go kit all the way. The GR Research AV1+, Adire 281, and sound clearing house kits are all highly recommended by people in this group.
And really- anyone capable of building a speaker cabinet should be able to solder just fine. It takes a little practice but it's the easiest thing in the world once you get it down. Do a google search for soldering tutorial and you'll learn exactly what to do. I think twisting and covering with tape is a bad idea- period.

 

[Dave Poehlman]

I think twisting and covering with tape is a bad idea- period.

Agreed.. a drop of solder will give you a much better connection. Practice on a few pieces of wire if you need to. I used some circuit perf-board I bought at Radio Shack on my first few X-overs. It helps in laying out the parts and holds them in place while you solder them (especially if you get a cap with a short lead).

You'd be surprised my first X-over even works if you saw the solder job on it. :b

 

[Brian Fellmeth]

So, what I'm looking to do is make something that sounds a bit better/fuller than my existing Infinity Entra Two cute widdle towers and Entra One center... Any suggestions as to which sorta brownies I oughta bake?

Aaron's suggestions are good. We need to know how much you want to spend. If your willing to spend $400 per box, there are many choices with high end scanspeaks or Seas excell etc. These will sound as good as $800 plus per box in retail equivlent. At $100 per box, still good options but they are different.

 

[Stephen Dodds]

There are kits to suit any budget here:

http://www.zalytron.com

All give you option of building your own box.

Steve