Build them yourself?

[jliedeka]

Ryan,

Thanks for those links. I'm really interested in the DD8-Mk II from Humble. As good as his cabinet looks, I want to run the numbers for a sealed design. Otherwise I'd use his choice of drivers and crossover.

Jim

 

[wreckingball]

Good discussion

This a very interesting thread, I like zaph's site for the simple fact that he took a bunch of time to test and post measurements on all of those drivers, essentially picking up where Bob Stout's LDSG left off.
In fact, that was one of the daunting tasks when I first decided to DIY back in 99", back when there wasn't a whole lot of proven DIY designs out there, and affordable computer based measurement systems were not exactly plentiful. A ton has changed since then, and thanks to all of the DIY guys out there that are willing to put up web site's and share with someone just starting out, like Mike, I think that's great.
I agree that Troel's site is very informative and it doesn't take to long to realize the guy really pays attention to detail, and knows what he is talking about (and Humble too), but the PE Usher 701 design is by Joesph D, who is not exactly an unknown.
I'll try to boil down my point here, and say that a really excellent choice to build would be Troel's "Amish", *IF* Mike want's to spend that much on parts.
A lot of us that have been around for a while know how many Commercial offerings used the Scan Speak 8545/9500 combo, but you would have around 700-1000 dollars in parts depending how elaborate the x-over components were, and how much goes into the cabinet/accessories. but he could go as far as he wanted in customizing/finishing that cabinet however he wanted. Edit: just found someone on canuck auto mart selling 6 pairs new 150 bucks a pair for the SS8545 not a bad deal
But getting back to kit/DIY discussion, the Usher 701 kit would be a lot less expensive $ 298 - low 300's. that being said the cabinets would be pre- finished and not so customizable, plus rated power is 80 watts, given the music he says he listens to, that probably wouldn't do.
That's a lot to consider in itself, and that's only 2 similar kits, not to mention others out there, much less DIY from scratch. (so much for boiling down my point). A couple more points: Troel states on his site that the 8545/9500 combo got a huge response from the web on that combo, also the Usher kit is all over the web as well, with x-over mods etc.
I think there is a lot to be said that when you are just starting out on the kit/DIY path, the more people out there building the same/similar thing you are, the more potential support you have, and my two choices above are not the only ones out there, I just know from building with the 8545/9500 they are excellent drivers. You would also get a lot of support with the Mini Statement, but that's 4 drivers per side, something else to consider.
Enough rambling on my part, and correct me if I am (factually) wrong on anything.

Just sharing my $.02 worth
Hope this helps Mike get started
Steve

 

[stinkfist]

This has been a great learning experience for me. Researching all of this and reading everything posted here and all the links has been a lot of fun.

I'm certin I'm going to build a set of speakers now. There are lots of good ideas out there and I've got a few designs in mind. One that I keep going back to is the Modulus w w w.humblehomemadehifi. c o m
From my limited knowladge it looks to be a nice speaker. I have to say I really like the style as well. Very simple design. I would love to here any coments you guys have on these speakers.

I figure I'll be spending around a $1,000 on this project and it will probably happen a little at a time over the next couple of monthes.

Thanks for the info guys,
Mike

 

[wreckingball]

Dipole

That is an interesting design, and a great looking speaker Mike, but I rambled on too long the last post, hope others chime in...

Cheers, Steve

 

[WmAx]

This has been a great learning experience for me. Researching all of this and reading everything posted here and all the links has been a lot of fun.

I'm certin I'm going to build a set of speakers now. There are lots of good ideas out there and I've got a few designs in mind. One that I keep going back to is the Modulus w w w.humblehomemadehifi. c o m
From my limited knowladge it looks to be a nice speaker. I have to say I really like the style as well. Very simple design. I would love to here any coments you guys have on these speakers.

I figure I'll be spending around a $1,000 on this project and it will probably happen a little at a time over the next couple of monthes.

Thanks for the info guys,
Mike

I think it should be interesting if you are sure to build the lower woofer module to be low resonance design. Do not use conventional DIY building standards for this module. Also, I would make modifications to the top open baffle piece; to reduce it's resonance. For example, use two thinner pieces of material and sandwich them togehter using a visco-elastic layer core and build a strong steel frame in the back of this open baffle also, or you can build the steel frame using steel square tube into the baffle's core using a router and proper poly construction adhesives.

-Chris

-Chris

 

[Spkr_Bldr]

Ryan, you bring out an interesting dilemma in a speaker design.
A dead flat speaker (granted low distortion, good polar response and friendly impedance) will reproduce all frequencies on the recording equally. There for it's the most honest speaker.
I personally don't find that speakers with even high end mega $$$ ultra low distortion drivers with ruler flat FR always sound musical.
Given the multitude of the listening environments, tremendous differences in person to person hearing and even daily fluctuations as much as 15db in yours or mine hearing, how can one voice the speaker effectively?
Given all this variances, how do you extract the best out of the set of drivers and what do you consider the best?
My personal opinion is that a part of FR of the driver that is flattest and has lowest THD is the one that is the best. Also my personal favorite is 4th order electrical L-R because of how it effects power response and there for the ability of the speaker to portray sound stage.

Well, that's why I say voicing is the true art of speaker design I don't think there's that much fluctuation in hearing, not mine at least. I know not to do any listening when I wake up because I'm usually stuffed up, and no voicing after drinking a couple beers ... but as long as my ears 'feel' right, and I let out a big yawn first to equalize pressure in the ear canal, my hearing seems pretty consistent.

But I do include many sanity checks in any new design voicing. I have a couple reference speakers that I do A/B comparisons with, and I try and let as many other people listen for extended periods as possible ... and I also take the speakers to other peoples houses to listen in other rooms and with other gear. In the development of Cirrus, it's so far been in 7 different systems/rooms, and had close to 50 local friends listen to it. I make sure and ask for what they really think, not just compliments.

All of that is a part of my voicing process.

Something I've found in my experience, drivers all have 'intrinisic characteristics' that are there from the first time you listen to them full range sitting on the floor ... all the way through numerous crossover iterations, and into the finished design. Meaning I can get a pretty good idea of how a system is going to sound when finished by listening to the raw drivers. It sounds weird, but it's true. I think many doubt it simply because they haven't tried, next time you're beginning a new design, first listen to the drivers raw without a filter on them (and a basic cap for tweeter protection). You might just see what I mean.

The way I've been able to apply these basic 'intrinsic characteristics' is with experience in correlation to a final result. For instance, I had in my head that the Peerless Exclusive 7" drivers I had would be easy to work with and implement. But the first time I listened to them raw, it was apparent that they are more unforgiving than I thought. And in system design it was true as well, the final voicing that sounded best with them was about a .5-1db depression from 400hz to about 2000hz, and flat from there. .5 to 1db wideband changes have a large impact, even though it doesn't sound like much. I could come up with numerous examples of how drivers have their own sound that isn't able in ways I know of to be shown in measurements.

Every system is going to determine what electrical filter network works best. Don't think any one is best, it's not. Only a best fit for a particular application. There are lots of ways to manipulate things like polar response in your favor. I honestly only use 4th order electical networks when absolutely necessary, like a center channel I built with a Seas Titanium tweeter and W16NX woofers ... in that application 4th order on both was best. Insertion losses are real though, and if you can hit targets with 2nd order electrical networks, you're going to be better off doing so.

 

[Spkr_Bldr]

This has been a great learning experience for me. Researching all of this and reading everything posted here and all the links has been a lot of fun.

I'm certin I'm going to build a set of speakers now. There are lots of good ideas out there and I've got a few designs in mind. One that I keep going back to is the Modulus w w w.humblehomemadehifi. c o m
From my limited knowladge it looks to be a nice speaker. I have to say I really like the style as well. Very simple design. I would love to here any coments you guys have on these speakers.

I figure I'll be spending around a $1,000 on this project and it will probably happen a little at a time over the next couple of monthes.

Thanks for the info guys,
Mike

That's an excellent design, you'll love it I bet. A couple things to note that you should consider before staring. First, those extra magnets he uses cannot be omitted. You might not be able to find a source for them, write Madisound and Parts Express to see if they can supply them. If not, write Tony himself to see if he can recommend a source. Without them though you can't build the speaker with his network. And second, Tony uses Series networks ... which is fine ... but put extra importance on having accurately measured crossover components. If you want the result to be perfect, you'll need to measure each individual component yourself. Caps and Resistors are typically labeled as + or - 10% tolerance, and sometimes they're not kidding. I see caps off by 10% often, and that much variance would totally ruin the response of any system, but especially one using a Series network.

 

[wreckingball]

Mike's kit/DIY

The way I've been able to apply these basic 'intrinsic characteristics' is with experience in correlation to a final result. For instance, I had in my head that the Peerless Exclusive 7" drivers I had would be easy to work with and implement. But the first time I listened to them raw, it was apparent that they are more unforgiving than I thought. And in system design it was true as well, the final voicing that sounded best with them was about a .5-1db depression from 400hz to about 2000hz, and flat from there. .5 to 1db wideband changes have a large impact, even though it doesn't sound like much. I could come up with numerous examples of how drivers have their own sound that isn't able in ways I know of to be shown in measurements.

Wow, a .5-1 dB depression is one thing, but from 400 to 2500Hz is big correction to make. I Agree with you Ryan, on almost every post you have made though. That being said, the devil is in the details, IE: one should always look for electronics that are within 1% tolerance ALWAYS

 

[Spkr_Bldr]

Wow, a .5-1 dB depression is one thing, but from 400 to 2500Hz is big correction to make. I Agree with you Ryan, on almost every post you have made though. That being said, the devil is in the details, IE: one should always look for electronics that are within 1% tolerance ALWAYS

Well, think about it this way ... all that means is that I built in .5-1db more baffle step then I normally would because otherwise the system sounded forward and could be fatiguing. Honestly that's how I should have explained it in the first place

 

[R-Carpenter]

If I was a beginner DIY, I'd start with Zaph's ZRT revelator towers. AirCirc tweeter and 18W8535G are both exceptional drivers. Cabinets are easy enough to build. It's a safe passage in to audio haven. You'll get an honest, pleasure to listen to, speaker.
Tony's designs IMO are quite a bit finicky to build. I can't say that series crossovers are not as good as parallel (all though it has been a common perception) but as Ryan mention they are definitely more sensitive to the component precision and I should also mention to the drivers precision. Don't forget, Tony can measure the hell out of his stuff and Mike isn't going to go and spend 3k on Clio.
I'd vote for the ZRT for a the simple reason of Scan Speak consistency in the driver manufacturing and Zaph's consideration of any there of.

 

[Spkr_Bldr]

If I was a beginner DIY, I'd start with Zaph's ZRT revelator towers. AirCirc tweeter and 18W8535G are both exceptional drivers. Cabinets are easy enough to build. It's a safe passage in to audio haven. You'll get an honest, pleasure to listen to, speaker.
Tony's designs IMO are quite a bit finicky to build. I can't say that series crossovers are not as good as parallel (all though it has been a common perception) but as Ryan mention they are definitely more sensitive to the component precision and I should also mention to the drivers precision. Don't forget, Tony can measure the hell out of his stuff and Mike isn't going to go and spend 3k on Clio.
I'd vote for the ZRT for a the simple reason of Scan Speak consistency in the driver manufacturing and Zaph's consideration of any there of.

I'd have to agree with that as well. The ZRT would be a great first build, and might end up being your last build. You can get the full kit from Madisound for a good price. Maybe beef up the cabinet a bit, and you'll have a stunning loudspeaker.

BTW, you can put together a full functioning measurment rig for about $150. Behringer ECM8000 mic for $50, m-audio external soundcard with phantom power mic pre-amp for around $100, then a free demo version of ARTA. It's a bit more once you figure in a few cables, and you need a laptop or PC. But point being it can be done cheap, I've designed a lot of systems with that setup. Mic calibration was on $40 too.

 

[Clearwave]

My only problem is I can't find these "kits" anywhere I look.

Hi Mike,

I've designed many free designs that can be found at htguide and my DIY website (Clearwave Loudspeaker Design). I also have kits for sale now with no charge for R&D and includes assembled crossovers. This is an adventure I started in September and it has been going very well.

Let me know if you have any questions or if any of the designs interest you. Since this is my first post I can't link websites to the post yet, but google is your friend. .

Regards,

Jed K.

 

[gus6464]

Hi Mike,

I've designed many free designs that can be found at htguide and my DIY website (Clearwave Loudspeaker Design). I also have kits for sale now with no charge for R&D and includes assembled crossovers. This is an adventure I started in September and it has been going very well.

Let me know if you have any questions or if any of the designs interest you. Since this is my first post I can't link websites to the post yet, but google is your friend. .

Regards,

Jed K.

Are you by any chance the one that designed to Tombstones? If you are I must say I love the design and look of your speakers.

 

[Clearwave]

Are you by any chance the one that designed to Tombstones? If you are I must say I love the design and look of your speakers.

Hi there- yes that is me. And thank you for the compliments- the Tombstones have been a bit of an obsession of mine for the past year or so. It is good to hear word has gotten out about them.

 

[wreckingball]

Well, Stinkfist?

Hey Mike' Just wondering what is going on with your speaker building/choosing
situation, from the looks of this forum, I would say that it's "all over but the shouting".
Anyway, just wondered what you decided on, and what you plan to build.
But...alass, you don't have to listen to me, as someone (not coming forth, and can't imagine why ) has decided that I am not qualified to assist you, even though I have built dozens of speakers an know what I am talking about.
Enough of the bitter, just want to hear some feedback, and congratulate you on deciding to "build them yourself".
Cheers to You, Steve

 

[highfigh]

I read somewhere that the average parts cost for a speaker you buy in a brick and mortar is 15% of MSRP. With ID it's closer to 40%.

I think a kit speaker can sound at least as good as anything you can buy if the design is sound. It needs decent drivers, relatively inert cabinets and a crossover that is suited to the application. If I were building a kit, I'd look for one that comes from a proven designer and/or has gotten favorable reviews.

Jim

How much is your time worth? You have to factor in the labor on any finished product. Also, they add a little to cover warranty claims. The maker of the speaker drivers doesn't usually put much more than a 90 day parts warranty unless the manufacturer is a really good customer.

 

[jliedeka]

It depends on whether or not you put a value on the satisfaction from doing something yourself. If that's not your thing, then there are plenty of respectable speakers available at a reasonable price.

I'm actually wrestling with the issue of whether or not to build or buy. For now I'm leaning towards buy but I have time to keep thinking about it. I'd love to design a set of towers with a matching center and monitor type surrounds. But the more I learn about speaker design, the less sure I am that I could build something satisfying enough to justify the labor.

Another possibility is getting a kit like the Exodus Kepler which runs $250/per without cabinets. Five of those would probably work for me.

Jim

 

[RacineBoxer]

I'd have to agree with that as well. The ZRT would be a great first build, and might end up being your last build. You can get the full kit from Madisound for a good price. Maybe beef up the cabinet a bit, and you'll have a stunning loudspeaker.

BTW, you can put together a full functioning measurment rig for about $150. Behringer ECM8000 mic for $50, m-audio external soundcard with phantom power mic pre-amp for around $100, then a free demo version of ARTA. It's a bit more once you figure in a few cables, and you need a laptop or PC. But point being it can be done cheap, I've designed a lot of systems with that setup. Mic calibration was on $40 too.

What do you think of the ZR71?
I'm sure the ZRT is a great-great speaker, but at $1k for just the parts and no enclosure that is a pretty steep entry fee into DIY. I was eyeballing the ZR71 which appears to be a similar design (2 way 7" x 1") with cheaper components (half the price). For $500-650 I'd still imagine you'd blow away any sub $1k (maybe even sub $2k) B&M designs.

 

[Spkr_Bldr]

What do you think of the ZR71?
I'm sure the ZRT is a great-great speaker, but at $1k for just the parts and no enclosure that is a pretty steep entry fee into DIY. I was eyeballing the ZR71 which appears to be a similar design (2 way 7" x 1") with cheaper components (half the price). For $500-650 I'd still imagine you'd blow away any sub $1k (maybe even sub $2k) B&M designs.

I think that's an excellent design with a simple but effective network and plenty of tuning options. The drivers are top notch, but the TBFCG isn't exactly the most detailed tweeter around. Overall it should be a good balance of detailed and forgiving.

 

[highfigh]

IMO, if someone is expecting a lot out of a kit, $1000 is a lot to pay for something that probably can't be auditioned. A few hundred is a lot easier to swallow as an entry point and at that price, tweaking it and learning on it shouldn't break the bank. If someone just doesn't have any woodworking skills/machinery/inclination, pre-made is a great way to go and if Madisound or Parts Express (or other vendors') cabinets don't do what you want, just about everyone knows someone who does woodworking as a hobby. If they can build a box, they can build a speaker cabinet. A lot of speaker manufacturers have done enough testing and/or analysis to give out a ballpark box size and crossover circuit. Box Builder is relatively inexpensive and should be easy enough to use.