My first time DIY speakers

[aurum_matan]

I'm planning out my first DIY speaker setup and I've got some questions.

The plan is to build a 2.1 system using mostly Dayton Audio drivers (I'm a grad student, trying to keep to a reasonable budget, but will deal with that as I go). I'll build the enclosures and crossovers myself, and plan on testing the drivers' frequency response before buying the crossover components. I'm undertaking the project because I want to upgrade from an old 2-way Sharp system and I thought I might enjoy learning about the building process/art (I've been reading a lot and definitely enjoying myself!). I'll use the speakers mostly for movies/music in a small, carpeted room.

A few questions while I'm still planning out the build:
Do people generally cut off the lowest frequencies (sub 40 Hz or so) of woofers in their crossovers? If I do this, will the speakers sound ok before I add the subwoofer?
I read somewhere that drivers do a poor job producing frequencies with wavelengths larger than the driver diameter (why?). I'm looking at a 10" woofer, which would suggest the crossover should be around 1350 Hz (10" wavelength). I was hoping to place the crossover closer to 1600 Hz or higher (I know a 3-way system would fix this, but that's not a jump in cost I want to take), and the spec sheet for the driver suggests the frequency response up through 1600 Hz would be just fine. Is this okay, or would it make more sense to use an 8" driver instead?
What's the deal with "shielded" tweeters?

Planned components (none bought so far, suggestions welcome)
Subwoofer https://www.parts-express.com/Dayton-Audio-DCS305-4-12-Classic-Subwoofer-4-Ohm-295-204
Woofer https://www.parts-express.com/Dayton-Audio-DC250-8-10-Classic-Woofer-295-315
Tweeter https://www.parts-express.com/Dayton-Audio-DC28FS-8-1-1-8-Silk-Dome-Shielded-Tweeter-275-075
Preliminary crossover design based on Mark Lawrence's design https://www.calsci.com/audio/X-Overs3d.html
Enclosure design: 3/4" MDF bookshelf speakers, drivers vertically aligned/offset from middle of baffle and countersunk, lots of internal bracing

 

[Verdinut]

I'm planning out my first DIY speaker setup and I've got some questions.

The plan is to build a 2.1 system using mostly Dayton Audio drivers (I'm a grad student, trying to keep to a reasonable budget, but will deal with that as I go). I'll build the enclosures and crossovers myself, and plan on testing the drivers' frequency response before buying the crossover components. I'm undertaking the project because I want to upgrade from an old 2-way Sharp system and I thought I might enjoy learning about the building process/art (I've been reading a lot and definitely enjoying myself!). I'll use the speakers mostly for movies/music in a small, carpeted room.

A few questions while I'm still planning out the build:
Do people generally cut off the lowest frequencies (sub 40 Hz or so) of woofers in their crossovers? If I do this, will the speakers sound ok before I add the subwoofer?
I read somewhere that drivers do a poor job producing frequencies with wavelengths larger than the driver diameter (why?). I'm looking at a 10" woofer, which would suggest the crossover should be around 1350 Hz (10" wavelength). I was hoping to place the crossover closer to 1600 Hz or higher (I know a 3-way system would fix this, but that's not a jump in cost I want to take), and the spec sheet for the driver suggests the frequency response up through 1600 Hz would be just fine. Is this okay, or would it make more sense to use an 8" driver instead?
What's the deal with "shielded" tweeters?

Planned components (none bought so far, suggestions welcome)
Subwoofer https://www.parts-express.com/Dayton-Audio-DCS305-4-12-Classic-Subwoofer-4-Ohm-295-204
Woofer https://www.parts-express.com/Dayton-Audio-DC250-8-10-Classic-Woofer-295-315
Tweeter https://www.parts-express.com/Dayton-Audio-DC28FS-8-1-1-8-Silk-Dome-Shielded-Tweeter-275-075
Preliminary crossover design based on Mark Lawrence's design https://www.calsci.com/audio/X-Overs3d.html
Enclosure design: 3/4" MDF bookshelf speakers, drivers vertically aligned/offset from middle of baffle and countersunk, lots of internal bracing

You are rather new to the design and building of speakers.

I perused the above list of your planned purchases and the mention of a passive crossover design.

First of all, values of passive crossover components have to be chosen specifically on the characteristics of each driver. An off-the-shelf crossover will never work.

If you want to start building speakers without the proper knowledge, I suggest that you purchase a completely designed kit from Parts-Express or Madisound. That way, you are assured of a successful result. But if you want to invest time to properly learn how to design and build speakers as a DIY project, I suggest that you get at least one book which will get you more acquainted with the complex speaker building process.

Here is a book with included design tutorials which I strongly recommend:

Home

Speaker Building 201 BookWith this book, anyone can become a speaker builder. You don’t need an elaborate workshop, expensive analytical equipment, or sophisticated software. Learn the concepts you need OR choose any of the eleven tested and proven speaker designs included in the book. Features...

www.parts-express.com

 

[Swerd]

Do people generally cut off the lowest frequencies (sub 40 Hz or so) of woofers in their crossovers? If I do this, will the speakers sound ok before I add the subwoofer?

Usually speakers do not have a high-pass filter to cut off low frequencies, for two reasons: 1) The inductor coils and capacitors required to work at such low frequencies are very large and expensive. 2) Any AV receiver, and some stereo receivers have built-in bass handling features that allow you to do this digitally, without the need for those large & expensive analog filters.

Most woofers naturally roll off their response below a certain frequency. It depends on the driver and the cabinet. Parts Express suggested sealed and vented cabinets for the DC-250 10" woofer:
Sealed Cabinet – volume 1.39 ft³, with an F3 = 55 Hz
Vented Cabinet – volume 4.47 ft³, with an F3 = 29 Hz

The F3 frequency is that frequency where the driver's sound pressure level (SPL) is 3 dB lower than the driver's rated SPL of 88 dB. It's commonly accepted that a change of ±3 dB is about the smallest interval most people can detect. (That topic can be debated, but let's not worry about that now.) Below the F3, the driver's response will naturally decrease, falling off at 6 dB/octave for a sealed cabinet, and 12 dB/octave for a vented cabinet. You have to decide what size cabinet you want to build, and what bass performance you want. (Hint: 4.47 ft³ is a rather large cabinet.)

I read somewhere that drivers do a poor job producing frequencies with wavelengths larger than the driver diameter (why?).

That's not quite correct. Drivers do a poor job producing off-axis sound at wavelengths shorter than the driver's diameter. And that's only a rough rule of thumb. It really should be measured while the driver is mounted in a cabinet with dimensions you intend to use.

Parts Express publishes this frequency response curve for the DC-250. The black curve is when the measuring microphone is on-axis (0° off-axis), red is 15° off-axis, green is 30°, and blue is 45°. Above roughly 1.5 kHz, the green line is about 3 dB below the black line. Above that frequency, this driver's off-axis response falls significantly. That will result in poorly dispersed sound, a poor sounding speaker. The solution is to cross this driver well below that frequency. Moreover, above roughly 2 kHz, this woofer is in break-up mode. Make sure to avoid that by crossing over by at least 1 octave lower if you use 4th order crossover slopes, and 2 octaves lower with 2nd order slopes.

I really don't think a 10" woofer can make a good sounding 2-way speaker. It's meant to be used in a 3-way design.

I'm looking at a 10" woofer, which would suggest the crossover should be around 1350 Hz (10" wavelength). I was hoping to place the crossover closer to 1600 Hz or higher (I know a 3-way system would fix this, but that's not a jump in cost I want to take), and the spec sheet for the driver suggests the frequency response up through 1600 Hz would be just fine. Is this okay, or would it make more sense to use an 8" driver instead?

I'd suggest a 6½" driver if you plan on a 2-way speaker, with that Dayton DC-28FS tweeter. Another rough rule-of-thumb is to use 3× the tweeter's resonance frequency (Fs) of 905 Hz as a minimum crossover frequency for a 2nd order crossover, and 2×Fs for a 4th order crossover. That becomes 1800 to 2700 Hz. Crossing a tweeter too low will lead to audible distortion, or even tweeter failure. As usual, spending more money on a tweeter can get you better low frequency performance.

What's the deal with "shielded" tweeters?

Shielding the speaker motor's magnet prevented interference if they were too close to an old TV with a cathode ray picture tube. Modern flat screen TVs don't require shielded speakers.

 

[-Jim-]

Well done @Swerd and @Verdinut. I'm thinking if he bought a complete package, it would be a better starting place.

 

[aurum_matan]

You are rather new to the design and building of speakers.

I perused the above list of your planned purchases and the mention of a passive crossover design.

First of all, components and values of a passive crossover have to be chosen specifically on the characteristics of each driver. An off-the-shelf crossover will never work.

If you want to start building speakers without the proper knowledge, I suggest that you purchase a completely designed kit from Parts-Express or Madisound. That way, you are assured of a successful result. But if you want to invest time to properly learn how to design and build speakers as a DIY project, I suggest that you get at least one book which will get you more acquainted with the complex speaker building process.

Here is a book with included design tutorials which I strongly recommend:

Home

Speaker Building 201 BookWith this book, anyone can become a speaker builder. You don’t need an elaborate workshop, expensive analytical equipment, or sophisticated software. Learn the concepts you need OR choose any of the eleven tested and proven speaker designs included in the book. Features...

www.parts-express.com

Thanks for the book suggestion! I think I'll feel more fulfilled if I spend a while learning and planning before building rather than jumping into a kit. A good general guide will help keep me from overlooking basic design elements I don't know about.

 

[aurum_matan]

Usually speakers do not have a high-pass filter to cut off low frequencies, for two reasons: 1) The inductor coils and capacitors required to work at such low frequencies are very large and expensive. 2) Any AV receiver, and some stereo receivers have built-in bass handling features that allow you to do this digitally, without the need for those large & expensive analog filters.

Most woofers naturally roll off their response below a certain frequency. It depends on the driver and the cabinet. Parts Express suggested sealed and vented cabinets for the DC-250 10" woofer:
Sealed Cabinet – volume 1.39 ft³, with an F3 = 55 Hz
Vented Cabinet – volume 4.47 ft³, with an F3 = 29 Hz

The F3 frequency is that frequency where the driver's sound pressure level (SPL) is 3 dB lower than the driver's rated SPL of 88 dB. It's commonly accepted that a change of ±3 dB is about the smallest interval most people can detect. (That topic can be debated, but let's not worry about that now.) Below the F3, the driver's response will naturally decrease, falling off at 6 dB/octave for a sealed cabinet, and 12 dB/octave for a vented cabinet. You have to decide what size cabinet you want to build, and what bass performance you want. (Hint: 4.47 ft³ is a rather large cabinet.)
That's not quite correct. Drivers do a poor job producing off-axis sound at wavelengths shorter than the driver's diameter. And that's only a rough rule of thumb. It really should be measured while the driver is mounted in a cabinet with dimensions you intend to use.

Parts Express publishes this frequency response curve for the DC-250. The black curve is when the measuring microphone is on-axis (0° off-axis), red is 15° off-axis, green is 30°, and blue is 45°. Above roughly 1.5 kHz, the green line is about 3 dB below the black line. Above that frequency, this driver's off-axis response falls significantly. That will result in poorly dispersed sound, a poor sounding speaker. The solution is to cross this driver well below that frequency. Moreover, above roughly 2 kHz, this woofer is in break-up mode. Make sure to avoid that by crossing over by at least 1 octave lower if you use 4th order crossover slopes, and 2 octaves lower with 2nd order slopes.
View attachment 53031
I really don't think a 10" woofer can make a good sounding 2-way speaker. It's meant to be used in a 3-way design.
I'd suggest a 6½" driver if you plan on a 2-way speaker, with that Dayton DC-28FS tweeter. Another rough rule-of-thumb is to use 3× the tweeter's resonance frequency (Fs) of 905 Hz as a minimum crossover frequency for a 2nd order crossover, and 2×Fs for a 4th order crossover. That becomes 1800 to 2700 Hz. Crossing a tweeter too low will lead to audible distortion, or even tweeter failure. As usual, spending more money on a tweeter can get you better low frequency performance.
Shielding the speaker motor's magnet prevented interference if they were too close to an old TV with a cathode ray picture tube. Modern flat screen TVs don't require shielded speakers.

Thanks for all the explanation. The woofer high pass filter answer is obvious when thinking about component cost and the problem being solved by the receiver, I've obviously got a lot more reading to do before I know what's going on.

I'll take a look at the 6.5" woofers and adjust accordingly. When I was looking at the frequency response posted by parts express, I didn't have the 3 dB context to evaluate off-axis response. That casts the 10" woofer's response above 1 kHz in a very different light than I'd been considering before. And yes, I did mean to ask about the driver producing sounds poorly at wavelengths shorter than the driver diameter (not larger, got that mixed up) and didn't realize it had to do specifically with directional response.

Thanks as well for the tweeter crossover rule of thumb, I'll make sure to take that into account when pairing the drivers and designing the crossover.

 

[lovinthehd]

Curious, were you thinking of building a single box for the speaker, including the sub? Or separate sub boxes?

 

[j_garcia]

BR-1 kit from PE. About as good as it gets for a first build IMO. It has good bass and is a solid all around speaker for the price.

 

[lovinthehd]

Or the new Philharmonic Affordable Accuracy Monitor kit https://www.philharmonicaudio.com/

 

[j_garcia]

I didn't realize he was selling the AA kit again. I still have that pair of AAs! For sure the tweeter and tweaks by Dennis make that an amazing speaker.

Overnight sensation is another good kit. A friend of mine lent me his to check out and they sound good.

 

[Swerd]

Thanks for all the explanation. The woofer high pass filter answer is obvious when thinking about component cost and the problem being solved by the receiver, I've obviously got a lot more reading to do before I know what's going on.

Glad to offer some general help. Your questions were intelligent, so I answered.

I'll take a look at the 6.5" woofers and adjust accordingly. When I was looking at the frequency response posted by parts express, I didn't have the 3 dB context to evaluate off-axis response. That casts the 10" woofer's response above 1 kHz in a very different light than I'd been considering before. And yes, I did mean to ask about the driver producing sounds poorly at wavelengths shorter than the driver diameter (not larger, got that mixed up) and didn't realize it had to do specifically with directional response.

Be careful about evaluating a potential driver based only on it's posted frequency response range. Parts Express lists a range of 25 – 2500 Hz for it's DC-250 10" driver. It may be true that the driver can make some kind of sound in that range, but you can't expect to actually use it across that range. The off-axis dispersion that I mentioned is an example of a more useful way to look at a woofer's real life upper limit. You must see the manufacturer's frequency response graphs measured on-axis and at several different off-axis angles.

Here's a link to another discussion on how to maximize woofer off-axis dispersion that I took part in.

Controlling speaker directivity

Can someone point me to a good source of information for learning to understand and manipulate this through crossover design? I think this will help broaden my understanding of speaker design, and maybe put a few other things into context.

forums.audioholics.com

There is also a limit to a woofer's lower response. It's based entirely on cabinet dimensions and the woofer's measured electro-mechanical characteristics, the so-called Thiele-Small parameters. The sealed cabinet that PE suggests for that driver, 1.39 ft³ can produce an F3 of 55 Hz, and a much larger 4.47 ft³ (roughly triple the volume) ported cabinet can produce lower sound with an F3 of 29 Hz. The book that Verdinut recommended, Speaker Building 201 by Ray Alden, has good chapters on the details of designing sealed or ported cabinets for a woofer, as well as how to decide if your woofer should go in a sealed or ported cabinet.

Thanks as well for the tweeter crossover rule of thumb, I'll make sure to take that into account when pairing the drivers and designing the crossover.

Those rules-of-thumbs are at best only general rules. There are enough other details that vary widely among tweeters, that you can't generalize about. All the manufacturer's frequency response graphs can only tell you so much. For example, consider the cabinet in which you plan to mount the tweeter. It's width will have a great effect on the tweeter's low end response. You have to measure what the tweeter does when mounted on the cabinet you plan to use. It's essential to know before you design a crossover. This brief link goes through a useful description of this, as well as several other basic issues of designing a 2-way speaker and it's crossover.

Salk Sound

www.salksound.com

 

[Am_P]

Thanks for all the explanation. The woofer high pass filter answer is obvious when thinking about component cost and the problem being solved by the receiver, I've obviously got a lot more reading to do before I know what's going on.

I'll take a look at the 6.5" woofers and adjust accordingly. When I was looking at the frequency response posted by parts express, I didn't have the 3 dB context to evaluate off-axis response. That casts the 10" woofer's response above 1 kHz in a very different light than I'd been considering before. And yes, I did mean to ask about the driver producing sounds poorly at wavelengths shorter than the driver diameter (not larger, got that mixed up) and didn't realize it had to do specifically with directional response.

Thanks as well for the tweeter crossover rule of thumb, I'll make sure to take that into account when pairing the drivers and designing the crossover.

This is a vast topic and catching it all in this thread may not be possible. A couple of things to note...There are no rules of thumb that are really set in stone since the design space is vast. Try not to get bottom of the barrel crap parts in diy to save a few pennies. You will soon regret it and upgraditis will set in quickly. Never cheap out on the tweeter. A capable tweeter crossed over under 2khz will make your design significantly more resolving. ..smaller driver = improved impulse response = improved clarity, if you care anything about detail, clarity, resolution, etc. I have diy'd several speakers over the years from scratch and there are forums dedicated to diy audio. You will need to start reading there. Many knowledgeable guys have already posted a lot over there. So, you can do a lot of research before buying anything.

You will also soon note soon that diy audio can quickly go up in costs as you start caring more about improving sound quality....this comes down to the individual a.k.a how picky he/she might be, of course. GR Research is a company that sells very affordable high quality kits. You may easily end up in the same ballpark price point as when you diy from scratch. So, it also may be a decent option for a first timer. I built/have had this kit of theirs since it was first offered and it sounds exquisite at it's price point.
https://www.gr-research.com/store/p52/X-LS_Encore_Kit.html

I am currently building this bigger/more expensive kit (work in progress). I have heard great things about this kit and pondered it for a while, but, will soon find out for myself
https://www.gr-research.com/store/p12/NX-Otica.html

 

[Noshigami]

DIY speaker setups can be a lot of fun, and it's great that you're taking the time to learn and enjoy the building process. Using Dayton Audio drivers is a smart choice for a budget-friendly option. A 2.1 system should work well for movies and music in your small, carpeted room. As for the klipsch forte compared to cornwall, they are both fantastic options, but it ultimately depends on your personal preferences and the specific characteristics you're looking for in your speakers. Good luck with your DIY journey, and I hope your new setup brings you joy and great sound!

 

[Trell]

Usually speakers do not have a high-pass filter to cut off low frequencies, for two reasons: 1) The inductor coils and capacitors required to work at such low frequencies are very large and expensive. 2) Any AV receiver, and some stereo receivers have built-in bass handling features that allow you to do this digitally, without the need for those large & expensive analog filters.

One speaker manufacturer that does so is the German manufacturer Canton, and they call it "Displacement Control" with a suffix of DC for those speakers that has it.

>>>In addition, like all Reference-series speakers, the 3.2 includes Displacement Control (DC) technology, which implements a custom high-pass filter that extends the linear low-frequency response while simultaneously curtailing out-of-bandwidth cone excursion. <<<

Canton Reference 3.2 DC loudspeaker

Menu

www.stereophile.com

 

[MrBoat]

Some things I have learned after building 7 pairs of speakers and 5 subs is, that something that supposedly "sounds larger than it actually is," doesn't. I know my personal SPL realm, which relies rather heavily on displacement/headroom. In other words, it's either going to be by a 3-way with 8" or larger woofers, and anything smaller than that requiring subwoofers. Somehow, some way, that displacement is going to need to be included.

Driver size needs to be scrutinized. When I see a so-called 5.25" woofer, I look to see just how much of that is frame. Many end up being more like 4". To me, that's a not-so-serious desktop driver or a mid-range.

Thankfully, I built the 12" speakers first and added a pair of subs to go with them for full-range music use. That large displacement makes for life-sized sound that is apparent even when they are idling along for ambient use.

All the smaller speakers I have built since are essentially novelties, comparatively. I use them strictly for point-blank, near field use, and still add subwoofers. In other words, be completely honest with personal expected performance from the start.

 

[Trell]

Driver size needs to be scrutinized. When I see a so-called 5.25" woofer, I look to see just how much of that is frame. Many end up being more like 4". To me, that's a not-so-serious desktop driver or a mid-range.

Yeah, the German speaker manufacturer Canton comes to mind where they overstated the size of their drivers for many years and had to go back to reduce it a few years ago. So 180mm mid-woofers are now 174mm.

I just went back to canton.de and they've removed most speakers except the current or so.

 

[HuenEye]

You are rather new to the design and building of speakers.

I perused the above list of your planned purchases and the mention of a passive crossover design.

First of all, values of passive crossover components have to be chosen specifically on the characteristics of each driver. An off-the-shelf crossover will never work.

UNTRUE!!!!!!!!!!!!!! To select the proper crossover point you find out what the Fs is and then go up at least one octave higher. Any 'off the shelf' crossover will be compatible as long as it compatible with the resistance of the crossover. I make some when the desired crossover point isn't commercially available or for a speaker rated at 6 Ohms, or I want to use something other than 2nd order crossovers which I do frequently to obtain a different lobe pattern or to integrate drivers. I try to keep it simple.

BTW, I use a DATS3 to measure the drivers and all other electrical parts that go into a crossover for accuracy and consistency.

 

[lovinthehd]

UNTRUE!!!!!!!!!!!!!! To select the proper crossover point you find out what the Fs is and then go up at least one octave higher. Any 'off the shelf' crossover will be compatible as long as it compatible with the resistance of the crossover. I make some when the desired crossover point isn't commercially available or for a speaker rated at 6 Ohms, or I want to use something other than 2nd order crossovers which I do frequently to obtain a different lobe pattern or to integrate drivers. I try to keep it simple.

BTW, I use a DATS3 to measure the drivers and all other electrical parts that go into a crossover for accuracy and consistency.

Why would one want to bother with an off-the-shelf passive crossover? One thing to cobble something together, something else to make it worth the effort....

 

[TLS Guy]

UNTRUE!!!!!!!!!!!!!! To select the proper crossover point you find out what the Fs is and then go up at least one octave higher. Any 'off the shelf' crossover will be compatible as long as it compatible with the resistance of the crossover. I make some when the desired crossover point isn't commercially available or for a speaker rated at 6 Ohms, or I want to use something other than 2nd order crossovers which I do frequently to obtain a different lobe pattern or to integrate drivers. I try to keep it simple.

BTW, I use a DATS3 to measure the drivers and all other electrical parts that go into a crossover for accuracy and consistency.

Where ever you got that idea, it is complete nonsense. You need to look at the impedance curve and the acoustic responses of the drivers primarily. In addition you need to pay attention to the sensitivity of the drivers. Crossovers absolutely need to be custom to the driver selected. No exceptions. Not all combinations of drivers have a passive crossover solution.