I have read hundreds of tutorial on speaker building but non of them give a clear description on how to choose the right drivers. I have a feeling that it is possible to make a High Quality with any kind of driver but I want to make sure there isn't a process or anything about how to choose the right driver. I want to make a 3-way loudspeaker, Sub-woofer, Mid-range, Tweeter, for my dorm room and don't want to spend much over $100. I understand frequency response graphs to some extent and I was under the impression that I should be looking for a driver whose frequency response graph had a steady linear increase and as flat as possible at its peak, but there isn't a single driver whose near that perfect. even those $400-500 dollar ones.
I decided to go and look at other peoples builds and look at all there specs they had on their drivers and loudspeaker and I found found guy who make a granite loudspeaker and their frequency response graph for the loudspeaker look very ideal so I went and looked at spec sheet and the graph for the tweeter they used and it looked horrendous. It was wicked jumpy and kinda made a parabola
This is the granite speaker build I looked at.
DIY Granite Speakers with Seas G17REX/P and 27 TFFC Drivers
This is the specs for the tweeter they used in the build.
H0881-06 27TFFC
could someone help me make sense of this so I can have a clear set answer on what drivers I should choose for my build I want to do.
Thank You
I decided to go and look at other peoples builds and look at all there specs they had on their drivers and loudspeaker and I found found guy who make a granite loudspeaker and their frequency response graph for the loudspeaker look very ideal so I went and looked at spec sheet and the graph for the tweeter they used and it looked horrendous. It was wicked jumpy and kinda made a parabola
This is the granite speaker build I looked at.
DIY Granite Speakers with Seas G17REX/P and 27 TFFC Drivers
This is the specs for the tweeter they used in the build.
H0881-06 27TFFC
could someone help me make sense of this so I can have a clear set answer on what drivers I should choose for my build I want to do.
Thank You
I would suggest concentrating on building one or two good subwoofers. If you visit enough thrift stores and yard sales you should be able to pick up a nice pair of name-brand, not-made-in-China speakers for the price of a burger or two. Decent subwoofers are much rarer, so it makes more sense to DIY. Also, subwoofers are much simpler to design since there's no passive crossover, no need to worry about impedance compensation or what happens to the response when the baffle stops being large with respect to the sound wavelength.
As for response plots, manufacturers don't always use the same type of smoothing or measure under the same conditions. Serious speaker designers will order samples of a bunch of candidates, then measure and compare them. Published specifications can't always be trusted, because they were based on previous manufacturing runs or pre-production samples. This is particularly true of house brands like Radio Shack, where the speakers are made by low bidders overseas.
If you want to build from scratch, use a proven design, preferably one that's been built by many people, so you're not trusting just one guy's opinion.
As for response plots, manufacturers don't always use the same type of smoothing or measure under the same conditions. Serious speaker designers will order samples of a bunch of candidates, then measure and compare them. Published specifications can't always be trusted, because they were based on previous manufacturing runs or pre-production samples. This is particularly true of house brands like Radio Shack, where the speakers are made by low bidders overseas.
If you want to build from scratch, use a proven design, preferably one that's been built by many people, so you're not trusting just one guy's opinion.
Ok so I found this site
Zaph|Audio
He has a bunch of tests for drivers. Should I look at his graphs and try to find the cheapest drivers with the most ideal response graphs. Would that be the best way to go about choosing the drivers?
Zaph|Audio
He has a bunch of tests for drivers. Should I look at his graphs and try to find the cheapest drivers with the most ideal response graphs. Would that be the best way to go about choosing the drivers?
Might I be so bold as to suggest you wait until your next pair to worry about a 3 way? With your budget, you are likely to be disappointed. More drivers does not mean better, in fact the idea really is: How to do what I want with the fewest drivers possible. I think that if you were to put those graphs in your back pocket for now and use your ears to judge, you might find that a single driver system will give you more of what you want and less of what you don't. I will further suggest that you consider drivers that give a solid high end and maybe add a separate woofer unit later when the budget allows.
In a word, no. You will not build a pair of three way speakers that will cover 20 to 20kHz for $100, or even a fair bit more. Even reasonable two ways for $100 is pushing it but far more achievable.
I suggest more reading of sites like Zaph's and Troel's and select a two way speaker design. In doing so, you will know what all of the components are, can find sources, and will know what the total cost is before you start. Importantly, you will also know what performance to expect.
Be realistic; you cannot expect very good performance for a pair of speakers for about $100, but you can maximise your return for going for a known design that has been tested.
(BTW, a sub-woofer is something to augment the lower bass response below that of a woofer. I suggest you are looking for a woofer.)
Abs
Almost utterly unrealistic.
Lynn Olson has been designing since the seventies and the discussion on his latest design has been going on here for six years.
Lynn has had a lot to say about selecting drivers.
Path of least pain would probably be a known-good "full ranger" (search for metronome or fonken in the forums) with a matched subwoofer and bi-amping/line-level crossover. Linkwitz should also be on your reading list on crossovers, especially at low frequencies.
And there's a lot of information here abouts.
Last edited:
Cal gave you some excellent advice. For your budget, I too would suggest a single driver speaker. Build a subwoofer when the budget allows. You can build a decent one for less than $100. Also, don't worry about the entire frequency range of 20-20KHz; music doesn't typically play as low as 20Hz nor does it play as high as 20KHz. Furthermore, most people can't hear that high.
The driver to use depends on a few things. One of them is what you plan to do with it; how loud, what type of music, room treatments, to name a few.
Start with someone else's design. A great design is a lot more than just knowing how to measure speakers and read graphs.
Parts Express Tech Talk is almost entirely devoted to speaker building. There's even a DIY show coming up in October in Iowa at which there will be a lot of great designs, great people, and great times.
Mike
The driver to use depends on a few things. One of them is what you plan to do with it; how loud, what type of music, room treatments, to name a few.
Start with someone else's design. A great design is a lot more than just knowing how to measure speakers and read graphs.
Parts Express Tech Talk is almost entirely devoted to speaker building. There's even a DIY show coming up in October in Iowa at which there will be a lot of great designs, great people, and great times.
Mike
I'll definitely take a look at what's he's got to say and done. In the mean time I'm thinking of going with the TriTrix Speaker Kit. Good? It's got good reviews and its 2-way. ( I still can't see why I should just go one driver.)
It's $139 on parts express and will probably cost me in total $170 so unless any thinks its a bad move. Then the TriTrix will probably be my first build.
Last edited:
Going with a kit is Good for speaker set #1. Especially not having to design/tune a crossover
Hard to go too far wrong with paper & silk. At the price point the claimed driver curves look promising. I've never heard the drivers/speakers but, as you say, the reviews are good. And there's lots of tips & tricks to try.
Last edited:
Because you indicated you wanted to diy. If you haven't done an XO before, it can be overwhelming.
This is a great idea as the XO is the toughest part of a speaker. Not quite diy but a very good starting point. You will learn much along the way.
i just take the train now, sorry if OT.
Is the range of sub harmonics important for the choice of a good speaker (mid & tweeter) ? The R2 (pair harmonic) must be much higher than the R3 and so on, or is there a nasty sound then the oposite or R3 level near R2 ?
I heard than much pair harmonic give an pleasant listening but R3 more accurate ? Snake oil ?
Which is most important of the two for a tweeter for example : distorsion measurement or form of the response curve ?
A speaker kit seller in Paris said to me than SN caps are the best because of their selfic characters and help the tweeter crossover to hunt the R3 sub harmonics!
I don't believe him, because I know these SN mkp caps : they sound like shoes-box and second his kits are not so good ! For me a cap is high-pass and attenuate everything : main, R2, R3,...
Is the range of sub harmonics important for the choice of a good speaker (mid & tweeter) ? The R2 (pair harmonic) must be much higher than the R3 and so on, or is there a nasty sound then the oposite or R3 level near R2 ?
I heard than much pair harmonic give an pleasant listening but R3 more accurate ? Snake oil ?
Which is most important of the two for a tweeter for example : distorsion measurement or form of the response curve ?
A speaker kit seller in Paris said to me than SN caps are the best because of their selfic characters and help the tweeter crossover to hunt the R3 sub harmonics!
I don't believe him, because I know these SN mkp caps : they sound like shoes-box and second his kits are not so good ! For me a cap is high-pass and attenuate everything : main, R2, R3,...
Last edited:
Any multi-way speaker will need a crossover, which ideally smooths out the imperfections in a driver that you mentioned. Crossovers can easily run $100/spkr for a three-way system.
What are you looking for in terms of use? Are you hosting late-night, wall-shaking dance parties or just looking to chill out? Dorms are pretty small, you won't need much to fill that space, and honestly, from experience, a loud booming system will get you into trouble in a dorm.
I would get a used set of bookshelf speakers, and build a passive subwoofer.
What are you looking for in terms of use? Are you hosting late-night, wall-shaking dance parties or just looking to chill out? Dorms are pretty small, you won't need much to fill that space, and honestly, from experience, a loud booming system will get you into trouble in a dorm.
I would get a used set of bookshelf speakers, and build a passive subwoofer.
I remember zaph tested a kef coax which was pretty good. maybe its around 100 each.
You can go emince beta with paper tweeter or horn as well for around 70 each.
12 or 10 inch
Behringer makes some pretty good value monitor speakers too.
or go with something like cheap 12 inch, maybe even 15 inch pro woofer. And add a small 'full range' driver from 500 on up. or full range+dayton rs225/270 if you want to go smaller and lower.
even three way is fine with dayton rs270/rs225. vifa tc7/tc9. And tweeter from 5k on up, which you can get for cheap since its easy to be done.
You can go emince beta with paper tweeter or horn as well for around 70 each.
12 or 10 inch
Behringer makes some pretty good value monitor speakers too.
or go with something like cheap 12 inch, maybe even 15 inch pro woofer. And add a small 'full range' driver from 500 on up. or full range+dayton rs225/270 if you want to go smaller and lower.
even three way is fine with dayton rs270/rs225. vifa tc7/tc9. And tweeter from 5k on up, which you can get for cheap since its easy to be done.
May I jump in here? I'd like to reinforce the OP's original question. "How to choose the right driver?" Take the dollars out of the equation and give the principle for looking at drivers , then look at how close your dollar limit lets you get to the ideal. If you mix in the $ right from the start it makes it harder to learn about the ideals for the drivers. IMO
I'm requesting this as much for my own education as for the OP's
Thanks
I'm requesting this as much for my own education as for the OP's
Thanks
Do read Olson and Linkwitz.
Olson on drivers is a good place to start.
I found a copy of a translation by Thomas Dunker of Ragnari Lian's postings on speakers . Worth reading.
Finally, a high Qmms seems to have a correlation with perceived good performance, all other things being equal. (See the Beyond the Ariel thread)
- Status
- This old topic is closed. If you want to reopen this topic, contact a moderator using the "Report Post" button.