Need Advice on building speakers for 7.1

[mintos]

Hi All,

I've built some tower speakers in HS as a class project years ago. Recently, my sony towers broke, and I found my old tower in the attic and decided to use them. I was very happy they still work, in fact they sound better than the Sonys. I had forgotten how they sounded. Now I'm thinking of building an entire 7.1 system and I need some advice.

My design priorities are:

1. Price
2. Performace
3. Looks

If they can look like a polk monitor 50, cost less, and perform better, I'd be really happy.

However, I can't do any thing too complicated since I've no woodworking skills. I can read a circuit diagram, plug in some equations, and soilder.

Step 1

I've picked out some speakers already in the below configuration...

L-----R------C-----LS------RS------BLS------BRS

T-----T------T-----MT-----TM------MT------TM
M-----M--W_M_W---W------W-------W-------W
W----W
W----W

They will all be ported. I tried to pick out three speakers with similar looks and ok cost wise.

Tang Band 25-1719S 1" Ceramic Dome Tweeter x 7 T = $133
Tang Band W3-315E 3" Aluminum/Magnesium Driver x 7 M = $150.99
Dayton DA175-8 7" Aluminum Cone Woofer x10 W = $245.60

Total = 529.59

Any advice on my choices?

Step 2: Cross Over design

This will be a major challenge for me. I need to know where I should start. With those speakers I'm thinking of a 3-way at 500/5k.

Step 3: Encloser design

I think I can use software to get me a basic box design. Maybe I can get a head start here with help on what type of wood working tools I would need.

Hope to hear back soon

Thank You All!

[johnathanwinter]

basic woodworking tools.
clamps, circular saw, router with bits, circle template, clamps (what i have seen you need lots of clamps), jig saw, some type of sander, probably a couple of putty knives, square, rule, straight edge, and i am probably forgetting a couple of things so just mark some paper to remind yourself to make a couple more trips to home depot, lowes, or another hardware store.

from what i have read most people say you need to do a proven design as it takes a long time to figure out how to design your own speaker. it doesnt seem as though you can just pick a few drivers and a couple of crossover points. you need measureing equipment and test boxes to find out how the drivers respond to being in the box together and what you will need to correct for ect.

anyway. if you pick the drivers you want and have or buy some test equipment then you could probably buy one each of the drivers and make a test box and take some measurements of the drivers and find out how they perform together. crossovers seem to be very complex. but then you may already know how to make crossovers. i dont know. im sure other will chime in.

you also didnt state what kind of budget your looking at.

ill end it here. talk soon.

oh yeah, im just beginning everything myself so dont take my word as anything. just offering what i think i know.

[jagman]

I would look in the Missions Accomplished forum and go with a proven design. There are several designs there that would meet your requirements and you would know going it that they would sound good. The chances of something you design (as a newb) sounding better are slim. That's not a shot at you... speaker design takes years of practice to get good at.

[peter_m]

If your first priority is price, forget about designing your own speaker. The equipment and software is not inexpensive... and even then it doesn't mean you will know how to reach your goal...

Look for an existing design... By the way you never mentioned the size of your room and if you plan on using a sub or not...

Peter

[mintos]

Room is about 20 ft by 15 feet. Software is free. Hoping to do crossover and cabinet using math only. No measuring equipment. Goal is just to make something cheaper than a Polk monitor 60 per tower and sound better.

I'm not looking to win an award and I don't mind using the EQ. The tower I made in HS sounded better than the retail Sonys and I didn't spend too much time and effort back in that class to make it.

[fbov]

...
Step 2: Cross Over design

This will be a major challenge for me. I need to know where I should start. With those speakers I'm thinking of a 3-way at 500/5k.
...

You're getting consistent advice, 3 for 3 and I'll make it 4 for 4 - pick a proven design and build that unless you have a good electrical engineering background and will treat each configuration as a new project.

To properly design a crossover, you need driver characterization in both frequency response and impedence vs frequency, and it helps to have distortion data available. FR measurement requires ~$100 in mic/mixer and freeware for the PC. Impedence data is an electrical bridge and the knowledge to use it, or another $100 for an automated "woofer tester." I'm not aware of any freeware distortion measurement rigs, but you will find many popular DIY drivers have already been characterized.

Once you have the data, here are a bunch of tools to get your first cut design.

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

Then you listen, tweak, listen, tweak, etc. until it sounds right.

It's so much easier and cheaper to simply find an appropriate design that others ahve successfully built, although I doubt you'll find any surrounds like you've sketched out; there's no point to 3-ways in back unless you have a far larger room (mine's 15x20 and MTM mains are fine) or like to listen at (literally) deafening sound levels.

There's a reason for the uniformity of response...

Have fun,
Frank

[ThomasW]

I'm not looking to win an award and I don't mind using the EQ.

Read this post...

http://www.htguide.com/forum/showpost.php4?p=162857&postcount=1

If you understand what's done in that post, and are able to duplicate it for the drivers you want to use, you can design a good loudspeaker. If reading it makes your eyes blurr with confusion, build a proven design from an known designer...