My first "done right" build...

[bzzeigler]

New to these forums, so I'm a bit overwhelmed with the amount of information presented to me.

So I am looking to build a pair of bookshelf 3way speakers using these drivers:
Vifa XT25TG30-04s HIGHs
HiVi F5s MIDs
HiVi F8s LOWs

I will be building the crossover myself as well. I plan on making these speakers bi-amp ready and I plan on utilizing an external crossover enclosure.

I plan on utilizing a slanted front baffle with facets and an offset tweeter but I wanted to keep the drivers facing parallel to the ground in order to keep the tweeters as close to ear height as possible.
The goal for the slanted baffle is to try and time align the drivers without having to do a stepped baffle, though I'm not sure how well the angled baffle will pull that off. It will also help reduce the number of parallel panels as the sides will also be angled. The rear panel will be perpendicular to the ground and the top/bottom panels will be parallel to each other.

I planned on putting the F5 in a sealed sub-enclosure with the XT25 in the same airspace, unless that sends up red flags for anyone?
The F8 would be in a ported enclosure.

The primary use of these speakers would be for listening to music in a 2ch/2.1ch configuration depending on the music I'm listening to, but would also be used in a 5.1config for movies and games.

I've got access to plenty of woodworking tools and plenty of time on my hands as this will be a summer project to do with my father.

Any and ALL input/opinions/suggestions you guys might have would be greatly appreciated!

[Bear]

If you've never built a speaker before, let's start with the obligatory read:

http://sites.google.com/site/undefinition/diy-faqs-provendesigns

I've no experience with any of the above drivers, though the Vifa tweeter is very popular. My only real advice is to plan on this taking a LOT longer than just a summer. For example, what software are you using? What size cabinets? When you say "parallel to the ground" is that from the perspective of the driver faceplate/frame or the direction sound is emitted from the cone/ring?

[fbov]

There's a reason you're overwhelmed.

Designing speaker from scratch (i.e. let's make a speaker from "these" drivers) requires measurement gear, simulation skills, and a basic understanding of electronics and acoustic wave phenomena in order to design the crossover. Read a couple books, buy a few $$$ in test equipment and you'll be set with a lifelong hobby.

The sane alternative, for someone like you (today), or me a few years ago, is to build a "proven design." Find someone who's already done the hard part (crossover design, based on selected drivers and planned layout), and copy their work if you're goal is to build a pair of good speakers. Here's the master thread:

http://techtalk.parts-express.com/showthread.php?t=211558

You've got some really good ideas here. I don't want to discourage you so much as steer you into a low-risk, high probability of success path. If I'm wrong, and you really want to learn speaker design, start here.

http://www.pvconsultants.com/audio/frdgroup.htm

Drill down into the top, How-to links to see what's involved in detail. If you're still interested, I bet you can find driver data and model your way into the design you describe. The more-expensive, but more reliable method is to measure FR and impendence in the intended application (the true goal of all that modeling and simulation).

But I still recommend finding a proven design that fits your needs as a first project.

Have fun,
Frank

[bzzeigler]

I've built a pair of speakers before, as well as a handful of sub enclosures for both automotive and home use. I've got a decent grasp on electrical components and circuit design, and I've received training as an audio engineer so I've got a pretty good grasp on the physics of sound waves. I don't mind this project taking a long time, provided at the end I've got a product that looks and, more importantly, sounds great.

I plan on buying an EMC8000 and having it calibrated. I have an Mbox 2 mini which will give me a good dac/adc preamp to work with.

I understand there is a lot of work ahead of me, but I think I'm ready.

[Undefinition]

Just my $.02...

I can tell that you're trying to do things "right." However, I think a lot of what you're focusing on is minutiae, and not nearly as important as THE BIG 3 aspects of speaker design
1. Drivers - actually, those are really nice drivers you've chosen. The XT25 isn't the easiest guy to work with, but it's a great performer when handled properly
2. Enclosure - The size of the enclosure, and its tuning, will make or break bass performance, as you probably know since you've built subs. I know the HiVi F8 needs at least a good solid 1 cu ft to really get going. Unless you're doing a sealed design, in which case you can go smaller, but the bass extension will probably not be all that inspiring (in fact, I'd wager that the F5 vented can go lower than the F8 sealed).
3. Crossover - Well, even if the first two are done well, if the crossover's not right, the speaker can fail. Doing a good crossover doesn't necessarily require measurement (though it makes a world of difference), but if you don't have measurement capabilities, I'd suggest at least modeling the measurements . From there, you can use your choice of crossover design software.

If you have THE BIG 3 in place, you're at least 90% of the way to an excellent speaker. The last few percent is just details. And this is where a lot of music lovers are misguided... focusing on things like offsetting drivers, roundovers/facets/chamfers, type of stuffing used, capacitor quality, speaker wire, etc..

You're talking about faceted cabs and slanted baffles, and the truth of it is that those things are really just details (the last few percent). So much of those issues can be accounted for in the crossover using just a plain rectangular box! Also, non-paralleled sides are also not terribly important, unless the cabinet is extraordinarily long, and it becomes a transmission line. I think people spend too much time picturing sound waves like light waves, which they are not.

I'm not trying to discourage you. Like I said, you have chosen good drivers, and you know how to build a cabinet. But I think that refocusing your priorities to THE BIG 3 will cut to the heart of what makes a great speaker, and you'll get to that goal much quicker.

[bzzeigler]

I'm not trying to discourage you. Like I said, you have chosen good drivers, and you know how to build a cabinet. But I think that refocusing your priorities to THE BIG 3 will cut to the heart of what makes a great speaker, and you'll get to that goal much quicker.

Not at all discouraging! I've actually come to this forum to be told things to do and what not to do.

As for the dimensions - I don't have anything set in stone but was planning on at least 1cu.ft. for the F8, and it was going to be vented. The F5 I planned on putting in a sealed enclosure seeing as I don't need the bass extension with the F8 being in there.

Of the BIG 3, the thing I still need to really focus on is the crossover, but I need to first get the drivers, measure the drivers, design the enclosure, build the enclosure, measure the drivers in the enclosure then I can focus on the crossover.

Speaking of measurements, I plan on buying an ECM8000 and having it calibrated, unless anyone has a better suggestion?
I've already downloaded all the FRD Consortium tools I could, however Unibox, Subwoofer simulator and Crossover Simulator don't seem to be hosted anymore, if anyone knows of a way I can get a hold of Unibox and Crossover Simulator, that would be great!
I'd love to be able to afford a CLIO system, but that is way out of my budget.

[Undefinition]

Unibox, Passive Crossover Designer, Response Modeler, etc. are all hosted here

http://audio.claub.net/software.html

A calibrated mic is a good thing, yes. You can get very, very far in your case, however, using simulated measurements (linked in my previous post). HiVi's measurements are very accurate, as are Vifa's. You could trace their graphs, model the responses, and would be modeling a speaker that is very close to its real-world counterpart. And don't get me wrong, I'm not advising against measurements. It's just that--and I've said this many times--measuring loudspeakers is a whole science in itself. It's hard enough learning crossover design, but to learn a whole another discipline at the same time... well, it's an exercise in frustration. It's like trying to graduate from all four years of college at the same time.
Please, if nothing else, read my link above about "simulating measurements."

[bzzeigler]

I definitely will be reading through the Simulated Measurements page a few times. I'm also looking at buying a calibrated microphone in order to make measurements, at least for the crossover. A $100 microphone isn't too terribly expensive, not when you look at how much time/money you end up spending on the rest of the project, plus its a one time expense that you can use for multiple projects...