New DIY Project (1 Viewer)

[PhilMays]

I have watched this forum for several years and have decided to take a Leap Of Faith at building speakers.

I have ordered a speaker building book online to help with the volume issues etc.

In your opinion.

1. Two way two 8" drivers and a tweeter.
2. Two way one 10"-12" driver and a tweeter.
3. Three way one 10-12" driver 5" mid randge and tweeter.
4. Three way two 8"drivers, 5" midrange, and a tweeter.
5. Crossover networks from parts express.
6. Where to buy high end crossovers if not the Parts
Express kind.
7. Should I build the crossover.

I ask this as I cannot find these specific questions in history.

Your help is greatly appreciated.

 

[Arthur_King]

After having been on other forums before, I have finally decided to buy a new car. Ive ordered a subscription to Car & Driver and Road & Track to help with reviews and reliability issues.

In your opinion.

1. A ford thunderbird
2. Chevy corvette
3. Lamborghini Diablo
4. Subaru WRX Sti
5. Honda S2000
6. Nitro Injectors from reputable sellers.
7. Catback Exhaust
8. Should I put in my own stereo system?

Sorry...

I couldnt help it. Phil, I know you're newbie, but you're really jumping in the deep end. You are all over the map with these questions and what they would mean. First, let me recommend that you read the speaker building book. Second, realize that building a set of speakers raw (not kit, not copying someone online) is HARD to do right. It takes a lot of effort, a lot of work and a decent set of "test" equipment (not to mention it helps to have good ears ). Building a sub is about 20 times easier, as there is far less to consider when you're talking 120hz and below than it does when you're talking 100hz 20000hz. Baffle step, speaker issues (peaks, ringing, resonances), crossover issues (slope, frequency of crossing) and 100 other things to consider.

I would recommend, if you're comming at this from point zero, that you seriously consider either buying a kit (speakers and build your own enclosure, or a full kit) or copy a DIY design on the internet.

THEN consider designing your own.

Just a suggestion

Daffy Arthur King
with the brain of a duck

 

[Patrick Sun]

Building a well-regarded kit is the best way for novices to get their ears wet.

I think it was in 2001 that I started from scratch, and it's been going on for 3 years now of tweaks and a recent total overhaul of the crossover design for me, and I've been lucky to know a local friend who's got the measurement gear and the crossover design software to point me in the right direction. If I didn't have access to these resources, I'd be having to rely on using manufacturers graphs (not always going to sync up with your particular set of drivers), and perhaps fiddle around with "Speaker Workshop" (google on it if you're curious) which is an okay way to play around crossover design, but it's limited and won't produce an optimal result.

(I did use measurement data from the manufacturer and imported it into Speaker Workshop to design my crossover for my center channel speaker as an experiment. It took a lot of tweaking, but at least it doesn't stand out noticeably when it's integrated with the rest of my speakers, which is a start.)

It's naive to think you can just build a box, pop in the drivers, and have a competently designed crossover network materialize with very little in experience and simulation/design software. You can stumble onto something that sounds okay, but it might not be optimal, and in the long run, your ears will probably suffer fatigue when listening to the speakers after a long period of listening.

But it's good that you're getting the speaker design book because it will get you up to speed on "talking" speaker design issues and concerns, and what is important to focus on, and what impacts overall performance. Once you get the vocabulary of speaker building down, your next project will come much easier in terms of what to look for to meet your design goals.

 

[PhilMays]

Well I guess I had that coming, or maybe not....

My question to me did not seem that difficult...

Let me reword it.....

What type of succes have others experianced with two way - vs- three way speakers.

Also are the crossovers from Parts Express worth a hoot or are there other sources that can supply high end crossovers.

I used to be a member of this site before the system crashed several years ago with 100's of post's allbeit in other forum sections. When I decided to build a set of speakers I decided to revisit this site to get some good info. I have ordered the book and will read it cover-to-cover to learn more. I was simply hoping someone had some basic opinions to my questions.

Again thanks for your reply.

 

[PhilMays]

Thanks Patrick...that was a little better worded!

I am going to build from scratch as I have the funds, workshop and drive to do just that. I also have a cool looking design in mind. If it works is something else.

I currently have a great home theater in two rooms of our house so I'm not in a hurry to place anything, or finish anything in a hurry.

While my questions seem elementary now, the replies I get will help guide me in direction.

Again thanks and I look forward to any other comments...but please be gentle

 

[ThomasW]

Take Pat's advice.......start with a kit

You shouldn't use anyones 'generic' crossovers. Crossovers should be custom designed after taking measurements of EACH driver operating on the baffle where it will be used. This is a costly and fairly complicated process.

If you want to continue; then buy a test microphone, mic preamp, and loudspeaker measurement software. Also buy box and crossover design software. Note that the XO and box design software doesn't tell or teach you how to design loudspeakers. It does make the design process faster and easier than using formulas and a calculator.

Also buy and read "The Loudspeaker Design Cookbook"

 

[MarkRoberts]

FWIW I built a sub first and now I have built MTM towers with 2 8" and a tweeter. The construction process was about the same. With the exception of the crossover. If I were you I would decide what do you want to have? 3-ways or 2-ways, towers or bookshelves? Then look around and find a project that suits your needs and your buget. If you are comfortable soldering and hooking things up you should have no problem with making a crossover from a parts list. Provided it is a proven design. Madisound, Adire have some nice kits. I happen to come across speakerbuilder.net which had a project that suited my needs. I am pleased with the results. It was a major leap of faith to go with something I had never heard before.

 

[PhilMays]

Thanks for the words of wisdom.

 

[Rory Buszka]

I can tell you don't like hearing what these guys on here have to say about your plans. I didn't like hearing that I couldn't just jump into full-range speaker design without the proper tools, either. So I built my first set of two-way MTM speakers with buyout drivers from Parts Express (actually a buyout of Definitive Technology) about a year ago. The drivers really weren't that bad. I had the Seas H537 tweeter and a pair of proprietary 5.25" cast magnesium frame woofers. How hard could it be, right? I pluged the numbers into a crossover calculator on the internet and bought the components and soldered it all up. I didn't know you actually had to know the response of the drivers. I just thought you could look at the "Frequency Range" specification and assume you'd be within +/- 3dB through there. But a crossover doesn't cut off right at the crossover frequency. Instead, it tapers-off the response of the one driver and rolls-in the other one. So you get a little bit of what is outside your crossover point as well.I was always listening to these speakers off the vertical axis for a while but when I moved everything up to my college dorm room and had a chance to listen to these speakers as they were meant to be listened to (on stands, with the tweeter at ear level), I noticed that it almost sounded as though they were whistling, and they were very harsh. I found Speaker Workshop and made some *.FRD files using software from the FRD Consortium's web page, Microsoft Excel, and the Clio test data from the Parts Express page. When I imported the response data and then put in the crossover components I was using (SpeakerWorkshop has a really good interface for this), I discovered that I was hearing a 5dB peak at 3200 Hz! What I thought would work well enough a year ago turned out to be total crap! Seriously, start out by building a proven kit. Parts Express sells some good ones that include decent-looking enclosures. Then when you build your own speakers you can have some reference of what is the "right" sound for a speaker to have. I know this isn't what you wanted to hear but I strongly recommend that you don't turn off in your head the advice you have been given here today. Anyone can learn from their mistakes. Real wisdom is in hearing guidance from one who knows, and avoiding said mistakes in the first place.

 

[MarkRoberts]

I know it can be a bummer when you get interested in doing a speaker and hear somewhat negative feedback but if you take the advice that is given you will be happier in the long run. If you want to do 3-way towers I say go for it just find a project that has already been proven out. Believe me there is plenty of frustration involved in just building the enclosures. The last thing you want is to hook up your new superswank speaks and then be making this face when you hear them :frowning: followed by
I have about 160.00 inparts invested. I have to remind myself of that every now and again when I get hyper critical of my diy's. When I look at it from that standpoint I dont feel I could do any better without spending at least 2x's what I did. I also still have just as much pride in ownership as I would if I had designed them completely. These aren't words of discouragement that you have been reading more so words of wisdom. Good luck with whatever project you go with and dont be shy about asking ?'s There allot of good people here that will help.

 

[Brian Fellmeth]

The problem with what you want to do is the crossover. As stated, it must be tweaked in real time with the chosen drivers on the chosen baffel using spectum analysing software and a calibrated mic in a suitable (close to anechoic) environment. But do not dispair. The heavy lifting has been done for you by well equiped fanatics who make avaliable free or for a nominal fee a proven design. There are hundreds out there of every size and price range (for the parts). Why reinvent the wheel. Its been said 5 times in this thread already, but its such good advice- follow a recipe.

Going back to your original post, if the application is for HT with a sub, then you want to build 2 ways. When only 80-20,000 Hz needs to be covered, there just isn't enough upside to three ways to compensate for a huge downside- 2 crossover zones instead of one. The crossover zone is where all kinds of problems occur.

Questions 5 and 6- you do not want to buy preassembled crossovers from anybody as has been said many times. You need to duplicate the circuit that has been optimized for a particular driver set and baffel geometry. The raw crossover parts (coils, caps and resisters) are avaliable from many suppliers with a wide range in price/quality.

 

[Hank Frankenberg]

Us older guys give advice at the drop of a hat, but there may be a tad of useful info amongst us. My humble advice is: DO NOT try a 3-way design. 2-way is fine and does not represent a newbie, timid, weak, uneducated approach. 3-ways are orders of magnitude more difficult for several reasons, AND the crossover parts cost can exceeed the driver cost. More importantly, most well-designed two-ways result in superior music reproduction than 99% of all 3-ways.
My 2 Hz.

 

[Christer W.]

More fuel on the fire. If you want to see how complicated crossover design can get, try taking a look at the Proac Response 2.5 Clone project. This legendary (among audiophiles anyway) speaker was successfully reverse-engineered. You can build one for barely 10% of the original cost. Simple construction, but fancy XO. The debate on the crossover is not expressed on the main project site shown below (hadn't started yet!).

http://www.geocities.com/diyproac25/

If you read through that page and find it a worth-while project, take a look at one of the many discussion threads regarding this facinating project. Here, I think that you come to grips with the complexity of getting the XO just right.

http://www.diyaudio.com/forums/showt...0&pagenumber=1

Prices have gone up slightly since the parts list from the first site above was published, but I figure you could build a pair of these for around 5-600 bucks depending on how fancy you wanted the finish. Seems like a lot perhaps, but I heard these in a boutique shop a few years ago. Sounded amazing, but I dismissed them at $4,500 or a pair, but all the work done by these generous folks makes them almost too compelling to leave alone.

 

[PhilMays]

I forgot to thanks those people wh offered some advise to me. Your "words of wisdom" has made me drop a three way project and tackle a smaller two way project.

I'm still going to do this with out a kit and use my own math and design (if I can remember to measure twice cut once). However with the good advise offered I feel I was taking on too much when I was designing a 5 to 6 foot three way tower.

 

[Aaron Gilbert]

Phil,

I applaud you for going ahead and doing this project from scratch. I agree that a two way plus a subwoofer is going to be considerably easier and likely better than a full three-way. So far, textbook formulae plus supplied manufacturer response and parameters, have always come quite close to my expectations.

I am from the school that believes you can build a decent sounding set of speakers without a bunch of test gear or simulation software. Though to friends and family, they are much better than decent, as speaker builders are usually the worst critics of they own creations.

To suggest that you MUST have a bunch of test equipment (other than a stereo of course!) and simulation software to be successful at building speakers, really does a disservice to the hobby and those looking to join it, in my opinion. That's not to say the tools don't improve results, but intuition and your own ears can go a long way.


Aaron Gilbert

 

[PhilMays]

Aaron,

You make a valid point. Last night I was reading an article where a designer uses his equiptment to get close but the final adjustments are performed to meet his "ears" standards.

Heck, if I wern't doing this I would be doing something else so why not this.

Thanks again for the words of encouragement!

 

[TimForman]

The only thing I will add is that you need some sort of reference for your measurement equipment (ears). Unfortunately, if you listen to something long enough it starts to seem "good" even if it's not. If this is your first build and you don't have a reference system to compare against build a proven design before attempting your own.