A few questions on loudpseakers and DIY

Status
This old topic is closed. If you want to reopen this topic, contact a moderator using the "Report Post" button.
I have a few questions on DIY and loudpseakers which I'm posting together. Hope I haven't erred by using one thread and if I have, please accept my apologies in advance.
What is the cost benefit of making one's own loudspeakers as against buying commercial, all things being equal? (neglecting personal satisfaction etc in 'rolling one's own)
For music (pop to classical) are stand mounted or floor mounted the better way to go - 2 way, 2 1/2 way or 3 way? (I know this is subjective so more or less a 'pole' question).
Are there any web sites around that have comprehensive lists of DIY speaker projects, prefreably with (brief) comments on each design?

As a little background, I have a Marantz 7002 receiver powering Paradigm Studio 40s and whilst fine for HT (With 590 centre, 590 surrounds and a Definitive Prosub 100), I'm not happy with the upper and lower frequencies (mid range is great) with music - either fed from Marantz DV 4001 or CD 6002.
Local speaker choice is limited to Paradigm, Bose, Infinity and Yamaha so I thought I'd have a go at making my own :)
 
Cabinets are the most DIY doable things . They are only passive , so no other thing requested than connect them to an amplifier . Are you sure your amplifier is capable of driving complex loads such three way designs ? Some may have low efficiency ( but large bandwidth :) ) or double woofers ,so you're warned !!
 
Thanks for the comments,
The Marantz amp is a 110W design and the Paradigms are 2 1/2 way but are bi-amped. Not heard that the Paradigms are difficult speakers to drive so on most counts the system, 'as is' should be OK.
No idea what the output topologu of the Marantz amp is but the tweeters are metal dome - not my favourite for a smooth top end:(
 
Is doing DIY cheaper? No.... But you would not believe how cheaply made current speakers are made. Looking up your Paradigm 40's.......they are well into the obligatory 6 1/2" camp..........everyone is making loudspeakers out of 6 1/2" drivers.........it is almost a disease. Look up the model line-up of Infinity loudspeakers from the seventies vs. Now! Even JBL has gone the 6 1/2" route........generations of excellent loudspeakers, now reduced to multiples of 6 1/2" drivers all.
With DIY you can break out of the 6 1/2" routine with excellent results! You can build your loudspeaker with the quality un-heard of in the commercial world.
It is only in the Studio and professional line-ups where you still find quality....and the pricing up to the DIY standards.....assuming you do your part of course.
The range of Pop music to Classical is an extremely wide one.....Pop music is...for lack of a better term, crap.... it is mixed & mastered poorly......tending to appeal to the non-audiphile crowd....The least-common-denominator. Classical is a wide field sound-stage, lots of dynamics from the whisper of a piccallo(sp?) to the crash of the entire symphony.
Just some thoughts & rants.

____________________________________________________Rick........
 
With DIY you can break out of the 6 1/2" routine with excellent results! You can build your loudspeaker with the quality un-heard of in the commercial world.
It is only in the Studio and professional line-ups where you still find quality....

____________________________________________________Rick........
He's cheating !!:cool:
In reality , those will sound DOUBLE with the same power ,'cos they're most High Sensitivity :rolleyes:

But a well made commercial speaker should be purchased to have a reference
:smash:
 
Hi Polsol,
You are in the right place! :)

DIY is something very pleasant and due to some reasons the success is almost inevitable. ;):) For instance the sole fact that you listen to something that came out of your hands is enough for most people.

There are some really pleasant commercial speakers, but they come at a price, some deserve their price more and some less.

You should start your DIY options research with a budget and space constraints if there are any such. Even if you have unlimited budget and no space constraints, it is advisable to limit your first project (if there is a first project, there is a second) both in budget and complexity.

I am somewhere around where you are, only have experimented a little, done one ready published project with some alterations, done some experimental projects with almost no aim, but trying and fun, some modifications such as damping and crossovers to my commercial speakers and etc. But I am more of a listener and music lover. I go to live music clubs and concerts, try to listen to as many systems as possible, listen to recordings critically and etc.

So if you accept an advice from a fellow newbie it would be: Simplicity of design, and two main aims, good bass in all aspects (that's the sugar in your coffee) and the most important aim is the good stereo imaging.
Everything else you can deal with, some uneven SPL across the audible range, some flaws in the high frequencies, some lack of detail and etc. - not a pain to die of. If you have good stereo imaging, you'll be happy - that is the most entertaining feature of a good stereo pair of speakers - the positioning and the good bass.

Since I made my first project, for about a year I read this forum in search of the most forgiving designs of the different component of a speaker, the crossover, the enclosure (box), the number of ways and etc.

I have come to the conclusion that the sealed/aperiodic enclosure would be most forgiving but on the price of big boxes, the first order series crossover would be most forgiving as it is self aligning, so the exact impedance vs frequency of the drivers is not very critical for components values, because you can escape errors with only adjusting the capacitor values. And last, but not least, you get a good imaging from a point source - this would be a full range driver, but a full range can't give you satisfactory bass in simple enclosure, so you add a woofer - the bigger the better.
Full range+woofer is called F.A.S.T.
But the first order series crossover comes at the price of either making the crossover point higher (above 400-500-600 hz) and thus demanding for a woofer that can go higher or a full range that is big enough to work in the lower frequencies.
As you see all is a matter of trade offs.

So the choices for an inexperienced candidate DIY-er are, ready tested, well documented and explained project, or some sort of forum assisted FAST project.

That is about all wisdom that I can share with you, hope it helps.
 
Last edited:
Looking up your Paradigm 40's.......they are well into the obligatory 6 1/2" camp..........everyone is making loudspeakers out of 6 1/2" drivers.........it is almost a disease. Look up the model line-up of Infinity loudspeakers from the seventies vs. Now! Even JBL has gone the 6 1/2" route........generations of excellent loudspeakers, now reduced to multiples of 6 1/2" drivers all.
With DIY you can break out of the 6 1/2" routine with excellent results! You can build your loudspeaker with the quality un-heard of in the commercial world.____________________________________________________Rick........

Yes, my feelings to a degree. It seems that 6 1/2" is the largest one can go without affecting the aesthetics of the design and how it fits into the suburban (or is that appartment?) lifestyle. Then add the exagerated 'throw of the smaller speaker' to reach the lower frequencies plus loss of slam due to lower cone area.
So add multiple 6 1/2 inch units to increase the dynamic range and I'd think that one loses imaging (a guess but correct?) at lower frequencies.

Sound doesn't seem to matter any more.:confused:
 
the most important aim is the good stereo imaging.

Sorry, what?

The most important aim of speakers is realistic in-room tonal accuracy and speech clarity at realistic listening volumes at realistic seating positions.

Imaging is overated to all but "audiophiles"... as long as you can hear the difference between speakers and they can center an image between them, you've got good enough imaging. The above on the other hand should be first priority. I mean, good speakers image well regardless, but you shouldn't say it's the most important aim.

As a little background, I have a Marantz 7002 receiver powering Paradigm Studio 40s

Yikes, boom and tizz speakers like that, now there's your problem. Overresonant high Q bass, and a hot metal dome tweeter that's apparently already in breakup before 10khz.

Now you have many choices, infinitely many, with DIY. For starters you might want to start with someone else's designs, as it may get you a frame of reference for a great speaker. You may also want to consider an active loudspeaker. The most central question is, what is your budget?

For music (pop to classical) are stand mounted or floor mounted the better way to go - 2 way, 2 1/2 way or 3 way? (I know this is subjective so more or less a 'pole' question).

As far as I've seen, the difference between a stand mount and a floor stander is essentially aesthetic, outside of internal volume and that sort of thing. I recommend a floorstander.

2-way is most simple / easiest for a designer to get working. Some of the larger, PA style two-ways are capable of high SPLs (but cannot give deep bass and require subwoofers) while some of the smaller, bookshelf and small tower style two-ways can sometimes extend deep at the cost of SPL, and others still can't really get loud or deep but are cost-effective, high sound quality builds. If you're looking to design your own or save on component costs, 2-way may be the way to go for a first project. MTMs such as the Dennis Murphy ER18 are also going to be 2-way, although they use 3 drivers. Coaxials like the Seas Loki are also two-ways, and make good center channels. And of course TMs like the ER18DXT are 2-ways.

2.5 way is imo a superior method, as it avoids the sheer complexity of a 3-way, but will have notably higher sensitivity (and mid-bass SPL capability handling) compared to the equivalent 2-way. A subwoofer will likely still be necessary as it won't let you extend any deeper than the equivalent 2-way. The Zaph TMM is a well documented 2.5 way

3-way For starters, component cost goes up. Design becomes far more difficult to integrate. IMO center channels should normally be 3-way for optimal performance. If you're gonna go 3-way, you may find yourself well-off with an active 3-way. Still, good passive 3-ways certainly exist. If going 3-way passively, IMO you should really look for ones that use high bandwidth bass drivers that can extend down to 30hz or below (so no subwoofer is vital) to 800hz or so (so a crossover to the midrange driver near 3-400hz is possible).

Also consider some 3.5 ways... the Zaph ZDT3.5 for example :)

the best full range speakers i've heard are 4-ways (although those are the only 4-ways I've heard), and the worst speakers others have heard are 4-ways. So you can imagine the complexity and cost as an extension of the above.
 
To answer your first question, absolutely! I'm fundamentally a cheapskate. It's why I got started building loudspeakers 40 years ago. I have a grand total of about US$200 in my two-way transmission-line speakers and I would put them up against anything in the $1000-$2000/pair range any day of the week.

The key is careful selection of components and coming up with a workable cabinet design. The first point is a subtle one. An awful lot of modern crossover design has to do with working around the limitations of drivers. That's fine. But if you get your driver selection right, crossover design can become embarassingly simple.

In my case, I'm very happy with 6 1/2 woofers. Mine are the venerable Vifa PW-18Js which have a low Fs and a very smooth midrange and upper midrange. Crossed over high (4khz), you have a lot of tweeter options. They actually work very well with no woofer crossover at all. 6 1/2s have a lot of advantages. Good dispersion, low mass and relatively rigid cones make for an exceptional compromise for most applications.
 
I'll answer only on the cost diy-vs-commercial issue.

First, take in mind that a private can buy things (drivers, components, but also wood, MDF, etc), at prices that are far higher than a corporation can: if I can buy for example a Seas tweeter for 40$ I suspect that a large speaker making company can buy the same tweeter for 20-25$.

But the same company would resell that tweeter into a complete speaker at least at 4-5 times the initial cost to cover all their costs (research, manufacture, distribution, marketing and profit). So probably you'll pay that tweeter in the system in the 100-150$ range. The same apply to all components.

So if you compare a diy speaker and a new commercial one built with roughly the same items, you can save some (significant) money doing diy. But if you buy the same speaker used you won't pay far more than the diy one. If you put into the equation also tools that you don't have and need to buy in order to make the speakers, you can find that the diy route can be more expensive than buying used, especially on low cost speakers! That is even more true if you try to design from scratch your own speakers, because you need measurement devices and time to learn how to use them.

If you go higher on value, and chose to build a someone's else proven design there will be an advantage on diy on the cost site, but I think that the most important value on diy it that you can tailor the speakers to your own room and habit, something that a commercial design obviously cannot.

Ralf
 
Many thanks to the contributors thus far (and hopefully more from you :))
The story so far:
Richard recommends larger bass drivers - seems correct IMO
T101 - Full range as point sources - difficult to argue against, just wish there were 3 driver concentrics ;)
RockLeeEV - reproduction quality is paramount. floor standers are a better configuration (from my point, with grandchildren, the more difficult to knock over the better anyway :))
Bunkie - DIY gives one better 'quality' at a similar price point than a commercial design.

On the later point I've always wondered what 'discount' commercial LS manufacturers get as opposed to DIYers when it comes to purchasing drive units.

So, I'm leaning towards a 2 or 2 1/2 floor standing design and a sub (if necessary) using a textile trebble unit at minimum, decent midrange and/or rather large bass driver (> 6 1/2"). Know and accepted designs rather than 'roll my own'.
Speaking of subs, is there any reference to 'spatial imaging' (I suppose there's a better, more eloquent term) at difference frequencies? i.e. I've read that imaging is not a problem at the lower octaves with subs and they can be placed 'anywhere' - put preferably next to a wall corner to boost the response (not altogether sure that's always a good thing - boosting the lower frequencies :()

As a final comment, Subs always have a high frequency cut-off. Wonder why they don't provide a low frequency also (to limit unecessary 'boom/rumble' on hard floors - tiles, concrete etc :confused:)

Sorry for the thousand questions, just trying to get my mind into gear on the whole aspect of Hi-Fi reproduction. :D
 
Sorry, what?

The most important aim of speakers is realistic in-room tonal accuracy and speech clarity at realistic listening volumes at realistic seating positions.

Imaging is overated to all but "audiophiles"... as long as you can hear the difference between speakers and they can center an image between them, you've got good enough imaging. The above on the other hand should be first priority. I mean, good speakers image well regardless, but you shouldn't say it's the most important aim.

Let me explain, even a FAST with first order series crossover would have 270-280 degrees of phase difference, and even a reverse polarity will leave you with around 6 cm (2.5 in) of acoustic centers offset at 1250 hz... and that's not all, he'll still have to experiment in order to get the mounting depth right due to natural phase issues of the drivers themselves. And that's with two ways and first order series...
No imaging means something more though, apart from the phase topic, which I agree is audiophilic to some degree, it means that the speaker has overal acoustic center which is random. That would mean that it sings to the floor, to the ceiling or somewhere which in turn harms the overall tonal balance. It would not be hard for a DIY-er to observe and adress those issues with the help of the people here, so why not mention it in the beginning?

I tried to make it shorter, thus i didn't pointed well enough that a fullrange is a good enough remedy both for imaging and for good tonal accuracy. Since he is asking so kindly for advice and help, I doubt you'll let him buy drivers with bad tonal balance or other significant flaws.

No bad feelings! :)
 
Hi Ralf,

Thanks for your input. On the basis of your figures it would seem that a DIY speaker is around 1/3 of the cost of a (new) commercial 'equivalent' - assuming that one has no extra expenditure on tools.
I think we all have our own 'ideals' for the configuration of a loudspeaker - rightly or wrongly (in some cases).

As an 'exercise of the mind' one could fashion a speaker based on 3 boxes - Tweeter, mid and bass, with each drive unit off centre (vertical plane). thus one could arrange the 'boxes' in relationship of the actual instruments (or artists) as they appear on the stage by placing the boxes in a different order in the vertical plane or upsdie down to emulate the horizontal plane (in other words the 'off-centre drivers' could face inwards or outwards).
Bit of a mission to re-arrange between songs ;) but would work for a single orchestral work/CD (assuming the musicians don't change places :D)
 
No imaging means something more though, apart from the phase topic, which I agree is audiophilic to some degree, it means that the speaker has overal acoustic center which is random. That would mean that it sings to the floor, to the ceiling or somewhere which in turn harms the overall tonal balance. It would not be hard for a DIY-er to observe and adress those issues with the help of the people here, so why not mention it in the beginning?

I tried to make it shorter, thus i didn't pointed well enough that a fullrange is a good enough remedy both for imaging and for good tonal accuracy. Since he is asking so kindly for advice and help, I doubt you'll let him buy drivers with bad tonal balance or other significant flaws.

No bad feelings! :)

You're right, a speaker should sound cohesive, and certainly a full range has that capability by nature, although I feel most better multi ways still sound cohesive to my ears acting as point sources.

I'd stick to my opinion that speech clarity and tonal realism in-room, at seating positions should be the number one goal, with everything else still having its own importance and contribution to the end goal.

No negetivity intended, though. I'm sure your approach is still with the same overall goal in mind, right? I just wanted to clarify that part. Many factors contribute in our quest for that goal.
 
Some factors to consider to determine if building your own speaker is cost effective:
1) Do you own or can borrow a saw, router and clamps, or do you have to buy?
2) If you're going to design your own rather than an established DIY design, do you have design software (free to couple hundred USD/Euro) and a measurement setup (calibrated mic, etc.)?
 
.......Yikes, boom and tizz speakers like that, now there's your problem. Overresonant high Q bass, and a hot metal dome tweeter that's apparently already in breakup before 10khz......

Personally I have always been very partial to fabric (silk) dome tweeters as they usually give a smooth "sweet" sound and and have none of the harshnees of most other tweeter types......IMO.

I also insist on a very "tight" bass. Far too many commercial speaker sys do have a
"overresonant high Q bass" IMO.

....I'd stick to my opinion that speech clarity and tonal realism in-room, at seating positions should be the number one goal, with everything else still having its own importance and contribution to the end goal.......

As a listener, I have to agree. A good, multi band EQ should be a part of every sys. :)
 
Last edited:
Some factors to consider to determine if building your own speaker is cost effective:
1) Do you own or can borrow a saw, router and clamps, or do you have to buy?
2) If you're going to design your own rather than an established DIY design, do you have design software (free to couple hundred USD/Euro) and a measurement setup (calibrated mic, etc.)?

I have a workshop at work that I can use. Unfortunately no signal generator or anechoic chamber though.

As a 'first project' I would be building an established design.
 
Personally I have always been very partial to fabric (silk) dome tweeters as they usually give a smooth "sweet" sound and and have none of the harshnees of most other tweeter types......IMO.
:)

Metal domes are especially probelmatical when used with an FET output stage on the amp.
Silk/Fabric are IMO better but metal domes impress on auditioning but become tiresome on pronlonged listening.
 
Personally I have always been very partial to fabric (silk) dome tweeters as they usually give a smooth "sweet" sound and and have none of the harshnees of most other tweeter types......IMO.

Fabric Dome tweeters are nice... but beryllium dome tweeters can be just amazing albeit pricy...I'd love to have a pair of Scanspeak 6640s to play with on someone else's dime.

As a 'first project' I would be building an established design.

You never did state your budget...

Metal domes are especially probelmatical when used with an FET output stage on the amp.

Hrm? That's kind of silly.

Metal domes are very rigid and capable of low distortion... If the breakup is above the audible range then there's no reason to not use metal - when properly implemented. Implementation is key with any driver. Focal and German Maestro have some quality metal dome tweeters. Again, it's all about individual drivers and the optimal implementation for them. Generalization is a bad idea and leads to audiophilic tendancies of B.S. spewing.

Silk/Fabric are IMO better but metal domes impress on auditioning but become tiresome on pronlonged listening.

You're confusing "commercial speakers that happen to use metal in their dome material" with "metal domes". Poor assumptions are made this way.
 
Last edited:
One doesn't need a full shop to build speakers (although it really helps!)

I used to use Baltic birch for small monitors that built. I designed the cabinets so that the panels were all the same width. The specialty plywood shop that sold me the plywood would, for a fee, cut the sheet into component parts for me. That allowed me to do without a table saw. The only power tools needed were a router with a straight bit, a modest circle gauge and a drill.

Since those days, I've acquired a full set of woodshop tools including a 3HP cabinet saw, 6" jointer, 14" band saw, a ShopSmith (multifunction tool with a lathe, drill press, table saw and various other functions) and a whole lot of other hand and small power tools. I've also acquired a woodworking hobby, something I might not have done had I not had the need to build speaker cabinets. My next skill frontier is slicing veneers to make my MDF boxes look as good as they sound, hence the new bandsaw.

While I certainly don't want to give a wrong impression, it's possible to build very nice-sounding speakers without test equipment. It is possible to get good sound by using subjective by-ear measurement. Back in the golden age, a lot of very good speaker designs were done this way. That's part of the fun. It's easy to breadboard an external crossover and have at it with trial and error. The same goes for driver selection, cabinet stuffing, etc.

We live in a golden age. help and advice is easy to come by, drivers are better than they ever were (I used to make my own woofers from component parts) and computerized testing is within the reach of mere mortals with limited budgets. The whole point is to have fun by learning and doing. I've been at it for 40 years.
 
Status
This old topic is closed. If you want to reopen this topic, contact a moderator using the "Report Post" button.