Questions about speaker building

Status
This old topic is closed. If you want to reopen this topic, contact a moderator using the "Report Post" button.
I would want to keep the whole project under $700 including amps. Right now I don't have an amp, I have my headphones and JBLs hooked up to a mixer. Also I think the only tool I will need to buy is a $30 jigsaw, my dad will have the rest. (Unless and router is necessary because I don't think we have that)
 
Just to be clear, so that's $700 for everything? So for the whole speaker, the amp and any necessary tools? Do you by chance have any access to any supplies at some kind of reduced costs? Like wood or insulation?

You may want to start with finding an amp. You are likely to need about 40-70W into 8ohm to achieve your desired SPL's depending on the woofer choice. But then you'll want headroom of at least 3dB if not 6dB. In terms of amplification, every time you increase the SPL by 3dB, you need to double the power. So if you need say 50W for your max SPL, 3dB of headroom will require 100W and 6dB of headroom will require 200W. The amp should also be capable of driving 4ohm speakers.

For your budget, I'd say under $100 for the amp should be your goal. I haven't really looked at amps for years so I don't know what that kind of money is going to buy, so I can't give you much help there, maybe someone else can do so or perhaps you need to start a new thread in the amp section for that. Perhaps going with something used might be a good choice.
 
Yes $700 for everything, my dad has extra wood but not the right thickness, kind, or size. I will ask about insulation though. I might be able to put some of my own money in the amps if necessary, not all of it needs to be for my bday. I actually think I will buy the wood before hand and build the box earlier but I am not sure. I still need to buy some $70 speaker stands and I want a new keyboard so I will see. Plywood is cheap according to my parents so I would say $600 to $700 for everything besides the wood.

Also, I know this is completely off task and I don't really want to ask this here but I don't know where to go for this rather specific question...

My mixer has 3 frequency knobs, low mid, and high. To get the realistic sound coming from the computer I have these at the minimum which is -15. When I put them all at the middle which is 0, I like the sound, it is less dead. When I try doing this in the parametric EQ in Fl Studio, I can't simulate the sound, it sounds very different...
 
Also, I know this is completely off task and I don't really want to ask this here but I don't know where to go for this rather specific question...

My mixer has 3 frequency knobs, low mid, and high. To get the realistic sound coming from the computer I have these at the minimum which is -15. When I put them all at the middle which is 0, I like the sound, it is less dead. When I try doing this in the parametric EQ in Fl Studio, I can't simulate the sound, it sounds very different...
What's the question ;^)?

Depending on the "Q" (width of boost or cut), cutting the low/mid/high tone controls by -15dB will probably leave peaks of as much as -6 dB between the low/mid and mid/high frequency centers, which would look like a roller coaster track. I can't think of that as a "realistic sound", seems you may be using the mixer's tone controls as an attenuator, not what they are designed to do.

If you adjust the "Q" of a parametric EQ to the same bandwidth as the mixer's EQ it could sound similar. Using a RTA (real time analyzer) would allow you to see what you hear, and start getting a grasp on the decibel scale, EQ function and frequency.
 
What's the question ;^)?

Depending on the "Q" (width of boost or cut), cutting the low/mid/high tone controls by -15dB will probably leave peaks of as much as -6 dB between the low/mid and mid/high frequency centers, which would look like a roller coaster track. I can't think of that as a "realistic sound", seems you may be using the mixer's tone controls as an attenuator, not what they are designed to do.

If you adjust the "Q" of a parametric EQ to the same bandwidth as the mixer's EQ it could sound similar. Using a RTA (real time analyzer) would allow you to see what you hear, and start getting a grasp on the decibel scale, EQ function and frequency.

Still don't want to get too off task but there is no "Q" in the equalizer... Also the -15 on the mixer is just the lowest the knobs go and is the same sound that the computer is sending to it. putting everything on 0 sounds very much more alive and that is what I'm trying to simulate in FL studio. Maybe I should open a new thread for this? or is there a simple answer?
 
Speaker Budgeting

So what you need to do is a cost analysis of basically everything except the drivers. Get out a piece of paper (or 2) and start making a list of everything you are going to need to buy for the project. And there shouldn't be any "well, this is going to be cheap so I won't include it" type of thing - every little expense adds up and in fact it's often smart when budgeting to build in a 5 or 10% leeway for unforeseen costs.

Maybe break it down something like this.

Tools: You should have a table saw, a router and a drill. However, a circular saw used with a long and true straight edge can work in conjunction with getting your wood supplier to make a few major cuts for you. It is possible to substitute a jigsaw for a router but it is trickier to flush mount the drivers that way and to do 45 degree chamfers on the rear of the driver cutouts. You can cheat in terms of flush mounting by glueing a 1/4" piece of board on top of the front baffle, but you have to be very exact with the cutouts. Best to finish them up with sandpaper I think. For your router, you'll need a 1/4" or 1/2" straight bit, a 3/4" 45 degree chamfer bit and a 3/4" roundover bit, the latter if you want to manually flare your port ends on the baffle. And fair warning, router bits aren't cheap. You'll also need a soldering gun and solder. And probably at least 4 decent size clamps although using heavy weights can work as well.

The Cabinets (Again to see how it can be done right, have a look at some of Troels' big speaker designs)

Wood: You will want to use 3/4" material. One of the best is very good quality Baltic Birch plywood that has no voids and many thin layers but it is rather expensive so forget about that. Most people on a budget will use MDF which makes good cuts and can be painted to an excellent finish but the edges are notorious for splitting if you screw into them. Plan on glueing MDF together only. There are other materials you can use (for eg) but it is often how you are going to finish them that makes a difference. Plywood is stiffer than MDF so it's better for bracing, but on a budget, MDF will have to do. Count on bracing all the cabinet walls with something like 1" x 3/4" strips. To be really picky, you'd place them about every 4" or so, but on a budget, about every 6 or 7" might do. And you'll want to place them side to side, front to back and top to bottom. Helps if they are tied into each other too to form a sort of matrix. There are other ways of doing it but this works well and uses the least wood so it's also the cheapest. The mid (and usually the tweeter is included as well) needs to be in its own separate closed chamber. Depends on the driver but about 5-8L of space should suffice.

So draw up a rough cabinet diagram with a total internal volume of somewhere around 160-170L. That should work for the woofers you've been looking at. Figure out how many sheets of 4' x 8' material you'll need for it and any other kind as well and cost it out. Expect to be in the $100 ballpark.

Damping and Insulation: Cabinet walls should be damped with something like bitumen pads, like these for eg, but on a budget you might be forgiven for skipping them. But you should use some kind of insulation to absorb at least some of the drivers' rear sound energy. You can buy acoustical foam made for the purpose but, again expensive. Many people opt for memory foam mattress toppers, say like this from Walmart. Others use 1/4" or 1/2" heavy felt, or I've seen foam rug underlayment pads (pic below) used as well. The key thing with any type of foam is it has to be open cell or breathable in other words. For the mid chamber, you should also stuff it with a little polyfill type light stuffing. So you need enough to cover all the interior surfaces which means just about as much square footage as the wood you need. Expect to spend up to about $75 unless you happen to score some kind of awesome deal.

Ports: You have the choice to buy plumbing pipe (ABS) or something like a Precision flared port from PE. Price each 1 out to see which way you want to go. Expect to need 4" x maybe 8"-14", Unibox will tell you. Maybe $15 to $30?

Hardware: You'll need cabinet feet, speaker wire, rear terminals and driver fasteners. When using wood screws to fasten drivers, the wood will often get stripped if you have to remove and insert them a number of times. For that reason, many people use machine type screws with what are called hurricane nuts. Price these out. Might be up to $30.

The Crossovers: You might try pricing out the xo at Parts Express for a speaker something like you want, say the Tarkus to give you an idea of costs. For the Tarkus xo, the 6mH inductor should have as low resistance as possible, (so a solid core, which is cheaper too) and the 125uF cap can be a non-polarized electrolytic to save you a lot of money. Otherwise, these for caps, these for inductors and these for resistors. Expect to probably pay somewhere around $150-200 if you opt for the least expensive parts and depending on how nicely the drivers you end up choosing want to play together.

Finishing: Sandpaper, Bondo, primer, paint (rollers or rattle cans), or veneer (and glue?) and clear coat? Costs can add up really quickly here. Expect $30-100?

And adding things up, it could look like this:
Amp = $150
Tools = $50?
Wood = $100
Insulation = $75
Hardware = $30
XO = $175
Finishing = $50

For a total of about $630. Just a guess at this point. Now maybe you don't worry about the finishing right now. Maybe you need no new tools. Maybe you don't include the amp. The total then will be about $380. On a $700 budget, that will leave you with $320 for the drivers which isn't bad. With a total of 8 drivers, that will average out to $40 per driver. Not exactly what you've had in mind up to this point though.

And right about now, you might want to re-think your design and acquaint yourself with the word 'compromise' Or 'patience' perhaps if you decide to wait longer to save up more money. Going with 1 woofer instead of 2 will save you in terms of driver, wood and insulation costs. Maybe finishing as well. You can still expect around 96 or 98dB at the listening position with a single with most of the woofers we've been looking at so far. Just something to think about anyways......

And hopefully, I didn't forget anything....... :eek:
 

Attachments

  • carpet pad.jpg
    carpet pad.jpg
    33.9 KB · Views: 68
Last edited:
hollowboy's linked paper may be a little more complex than you need right now. Or maybe not. But it is looking at the effects of more than just room pressurization gain.

The simple version is just to look at the 2nd picture. The smooth line is response outside (close to an ideal simulated response), spiky blue line is response inside (what you actually hear in most systems).

room gain

Below 200 or 300Hz, real results are not like the simulation*, and the results vary when you move the speakers around, or install them in different rooms.
That's got nothing to do with how good your speakers are - it will be true for your build, and also true for a commercial equivalent that cost $50,000 for a pair.

*the exceptions being 'perfect' rooms, where the user has used some combination of:

-planning
-multiple bass sources
-room treatment
-good layout
-equalisation
-trial and error

...to get the low frequency response to be more even.
 

Attachments

  • room effect at listening position on raw sub response.JPG
    room effect at listening position on raw sub response.JPG
    17.4 KB · Views: 62
the jigsaw on amazon was $30 but routers were all in the $200 to $300 range... Also my dad has a power sander so finishing won't be a problem. Might just add $10 for a thing of black spray paint. He also has a solder.

This is a very long term project so I could easily do some work and wait till I can afford the next thing If I can't get everything for my Bday.

I also think a new keyboard can wait since it is mainly for looks, by October I should have like $150 to $200 of my money to add if needed. Of course I don't need to spend every dime I can but this could be insurance because I don't know what everything will be in the end.

Hopefully if you think it wouldn't be stupid, I would want to get to work on the box asap. If you think I should get the crossover design down before I'll do that but after the drivers are confirmed I should be able to confirm a bunch of other things in the budget.
 
The simple version is just to look at the 2nd picture. The smooth line is response outside (close to an ideal simulated response), spiky blue line is response inside (what you actually hear in most systems).

room gain

Below 200 or 300Hz, real results are not like the simulation*, and the results vary when you move the speakers around, or install them in different rooms.
That's got nothing to do with how good your speakers are - it will be true for your build, and also true for a commercial equivalent that cost $50,000 for a pair.

*the exceptions being 'perfect' rooms, where the user has used some

combination of:

-planning
-multiple bass sources
-room treatment
-good layout
-equalization
-trial and error

...to get the low frequency response to be more even.

Not sure how I'm supposed to implement this into my build. I can't change the room...:confused:
 
Not sure how I'm supposed to implement this into my build. I can't change the room...:confused:

You don't :)

My point is that bass response will be messy in real life, no matter how much time you spend obsessing about the length of the port you use.

Measuring the finished box, and spending 20 minutes finding the best position + eq settings for it will probably give more benefit than getting the LF alignment absolutely perfect on paper.
 
You don't :)

My point is that bass response will be messy in real life, no matter how much time you spend obsessing about the length of the port you use.

Measuring the finished box, and spending 20 minutes finding the best position + eq settings for it will probably give more benefit than getting the LF alignment absolutely perfect on paper.

Well unfortunately I can't move them around in the room, I can attach an image of approx where they will go but when I get a 31 band equalizer finding the best settings is something I can do.

I can't move them because there is weight equipment and I already spent hours designing and planning how the stuff in the room is setup and there is only one logical choice on where to put the speakers. In the pic, the speakers are in red.

27.PNG
 
Still don't want to get too off task but there is no "Q" in the equalizer... Also the -15 on the mixer is just the lowest the knobs go and is the same sound that the computer is sending to it. putting everything on 0 sounds very much more alive and that is what I'm trying to simulate in FL studio. Maybe I should open a new thread for this? or is there a simple answer?
Will,

If you want to design a three-way speaker, it would help if you understand filters, so no need to open another thread.

Simple answer is every equalizer or tone control ever made has a "Q", the width of the filter. The filter could be very narrow, a fraction of an octave, or very wide, several octaves.

By definition, a parametric equalizer can adjust the width of the Q, the frequency, and the amplitude (amount of boost or cut in dB). An EQ that adjusts frequency and amplitude but uses a fixed "Q" is semi-parametric.

If the filter frequency and "Q" are fixed, normally the circuit is called a "tone control", though a tone control can achieve some limited degree of "EQ".

If "FL studio" has a 3 (or more) band parametric EQ, it could reproduce the sound of a 3 band fixed tone control. That said, depending on the tone control circuit placement in your mixer's signal path, by running the input hot and cutting the tone controls the input stage may be clipping, that distortion could also be part your perception of an "alive" or "dead" (or "realistic"...) sound.

When you reduce each of three bands of a tone control by -15dB, the overall level is cut by a similar amount, other than between center frequencies. It only takes a 10dB reduction to sound half as loud, so obviously "putting everything on 0 sounds very much more alive".

As to your recent post, there is nothing "logical" from a sound standpoint of an asymmetrical placement of your studio monitors on the wide wall of your room.
Exercise is good, but your speaker placement is severely compromised, and no amount of EQ will fix it :Ouch:.

Time to get used to spending hours designing and planning things you will later find are completely hosed- you can't design for things you have not yet learned about.
Get used to letting go of things that don't work, it saves a lot of time in the long run.

Have fun with exercise, both your mind and your body!

Art
 
Last edited:
"We live in the midrange"... Since the TM covers 120-20,000Hz, I always start by designing an excellent TM like the 12" Eminence Delta Pro 12 + Selenium CD, and then select woofers with required F3, SPL curve and cabinet shape. After an education on ported woofers, your study of midrange and tweeter options will likely move you to a "top-down" design philosophy.

1) You really want a great midrange like the Eminence Delta Pro 12A
2) You really want a proven passive crossover for the T-M
3) You really want to bi-amp...with adjustable LR2/LR4 active crossover
--passive woofer crossover parts costs ~$30

Still my recommendations:
$13 B-52 PHRN-1014 1" Horn 10" x 14" Bolt-On waveguide
Model: PHRN-1014|Part # 299-2303
$34 Selenium 1" compression driver 2/3 Bolt . ( Proven Xover ckt available)
Selenium 8 Ohm Model: D220Ti-OMF-8 |Part # 299-2321
$130 Eminence Delta Pro 12A midrange 12"
$76 Dayton DAS315-8 (12" woofer need 2/speaker)
-------
$319 = $329 -$10 coupon + free shipping
$25 passives for each TM

$130 Adcom GFA-2535 4 Channel Power Amplifier USED on ebay
$40 stereo 2/3-way active LR crossover ebay (analog opamps, just requires +/- 15V supply)
Linkwitz-Riley Electronic Crossover
----$860 total
 

Attachments

  • 320_Bucks.jpg
    320_Bucks.jpg
    158.6 KB · Views: 87
Well I can't change the way the weights are setup, by logical I meant only place besides like behind me in random places... Right now the JBLs are playing fine in terms of how they are positioned.

It is also more than a few weights, it is an entire mini gym, the hours I spent were making sure it was setup so it could be used with easy access to the weights and no risk of damaging the electronic equipment on the studio side of the room. The main reason I am building the speakers (Besides the proccess) is for doing weights anyway!

Anyway, by "alive" I didn't mean louder, I was just describing how I thought it changed it.
 
when I get a 31 band equalizer finding the best settings is something I can do.

That's going to be a huge help. If you play all your music through a PC*, you won't need a physical unit - you can eq for free in software.

I can't move them because
OK. But quite small differences in location, particularly near room corners, can sometimes make a big difference to the sound.

Regarding room interaction, using 2 woofers is better than using 1. It means you'll have a more complex floor bounce pattern (which usually means smaller nulls).

*how are you making music, and do you have any samples available? Just curious.
 

Attachments

  • bounce.jpg
    bounce.jpg
    19.2 KB · Views: 79
Most of what jReave said about budgeting is perfectly workable. I'm giving no budget versions of some of it.

It is possible to substitute a jigsaw for a router but it is trickier to flush mount the drivers that way
Or simply don't bother with flush mounting. Many big PA speakers don't flush mount their drivers.

and to do 45 degree chamfers on the rear of the driver cutouts.
Again, you can skip this - many big PA speakers don't have chamfers. It is more important on boutique home speakers (small cutouts and massively thick panels), as shown here:
chamfering driver holes

...whereas big PA speakers usually have huge cutouts, relative to the panel thickness.

If you do chamfer the holes, they don't have to be perfect 45 degree angles. Cut chamfers with whatever hand tools you have. One of Troels' examples was done with a chisel and hammer :)

You can cheat in terms of flush mounting by glueing a 1/4" piece of board on top of the front baffle, but you have to be very exact with the cutouts.
Before I had a router, I scored the cutout circles by making a 'compass' from a scrap of wood. I put a bolt through the centre, and a slightly protruding screw to scratch a circle to mark the cutout line. I'd scratch the circle progressively deeper, then follow that line with a keyhole saw (a hand tool; I didn't own a jigsaw until recently).

Wood I'd go for cheap ply over cheap MDF. Local price + quality variation means I can't say much more.

Cabinet walls should be damped with something like bitumen pads, like
Or: simply don't bother. Many big PA speakers don't.

Cheap options include: almost anything that's dense but not stiff. e.g. I've used a slurry of very cheap gap filler glue + kitty litter (clay pellets).

You can buy acoustical foam made for the purpose but, again expensive.
Cheap options include: almost anything. Many big PA speakers use fibreglass. Used pillows and clothes (such as old moth-eaten wool garments) work too.

Ports: You have the choice to buy plumbing pipe (ABS) or something like a Precision flared port from PE. Price each 1 out to see which way you want to go. Expect to need 4" x maybe 8"-14", Unibox will tell you. Maybe $15 to $30?
You can build square ports for free, with offcuts of the plywood you'll be using. You can use found / re-purposed objects, like poster tubes: the big JBL boxes I salvaged used stiff cardboard tubes, painted black.

Hardware: You'll need cabinet feet, speaker wire, rear terminals and driver fasteners.
Skip the feet. If you have a couple of months waiting time, keep your eyes open. You can tear trashed speakers apart to get rear terminals and miscellaneous bits for $0

When using wood screws to fasten drivers, the wood will often get stripped if you have to remove and insert them a number of times.
Possible, but hasn't happened to me. When I've drilled a hole too big, I've repacked it by jamming in some PVA glue + small bits of wood (matches or toothpicks). Once it dries, you can re-drill a smaller hole. That fixes your worst-case scenario.

Similar stuff is covered in this page:
Tips and ideas Copyright 2012-14

For that reason, many people use machine type screws with what are called hurricane nuts. Price these out. Might be up to $30.
Isn't that incompatible with chamfering? Troels advises against them in his 'tips and ideas' page.

I found, when dismantling a couple of old PA speakers, that many of the threaded inserts had fallen out, and others had been over-tightened, causing damage.

The Crossovers: You might try pricing out the xo at [ ...] Expect to probably pay somewhere around $150-200 if you opt for the least expensive parts

A miniDSP replaces this (or most of it if you make a hybrid). It can also replace an equaliser.

And adding things up, it could look like this:
Amp = $150
Tools = $50?
Wood = $100
Insulation = $75
Hardware = $30

Confused: $15-$30 ports, $30 hurricane nuts and several other things (feet, wire etc) shouldn't come to $30 total.

XO = $175
Finishing = $50

On top of my no budget substitutions, I'd go with:

Amp = $20. Salvage a free laptop charger or similar power supply. If you don't have one at home, ask around e.g. from a hotel - they often have a box of them, left behind by guests.

Use this to power two cheap ~50 watt amplifiers. Parts Express has caseless amps for $10 each.

XO = $125. Use a miniDSP for the bass-to-mid crossover, and build a simple textbook passive XO for the mid-to-tweeter. A miniDSP is cheaper (and sometimes better) than a passive crossover, it lets you use 2 small amps rather than 1 big one, and also lets you skip buying that 32-band equaliser.

And hopefully, I didn't forget anything....... :eek:

Just optional stuff, depending on how moved / abused the boxes will be:

Handles. Simple recessed steel ones are ~$5 each, so $20 for a pair of woofer boxes. I've kept the old handles from PA cabinets I've scrapped: they seem to be indestructible.

Corner guards. Only a few $ and handy if the boxes will get dinged a lot.

Grilles. I don't use them, but essential for some situations.
 
If you actually had the budget for it, I quite like LineSource's suggestions. But when you include all the extras, that's going to be a fair bit over $700.

And hollowboy shows you that there is more than 1 way to skin a cat.

And really, I have to agree. That looks like a very poor way to set up your room if you have music as a priority.

Also, weltersys has offered some sage advice.

For myself at this point, having now spent over a week working with you on vented box modeling and getting a better sense of your budget, priorities and capabilities and realizing that getting your cabinets designed and built right now while you are off for the summer is probably what you should be doing, here is my recommendation:

You should build the Tarkus. It won't meet some of your preconceived notions about what you wanted but it has these factors going for it:

- it's a more realistic project for you
- you can re-design the cabinets to your liking and get started right away on them. I've got the sense that that may end up being the fun part for you.
- you do not have to learn all about xo's and measurements and a slew of other speaker design topics which look to require a lot of time in front of a computer screen. And I mean a lot of time. I'm not sure but at 14, I don't think I would have considered that the best way to spend my summer off from school.
- you will learn all about cabinet construction, soldering, driver and xo wiring and maybe finishing, which is a great start for your 1st project
- you will be guaranteed a set of very good sounding speakers that will play very loud, I'm guessing louder than you think they will, which is not necessarily the case when you are designing such a complex project your 1st time around
- and you should be able to stay well within your budget.

Anyways, maybe not what you wanted to hear, but that's the way I am seeing things at the moment.
 
Hollowboy said I could use a miniDSP and build a simple tesxtbook crossover. Anyway my patience level is more than 5 minutes, I do not have a problem learning things!! I don't care how long this project takes, I just want to build speakers that I want. Maybe I don't get everything for my bday, I still have plenty of weekends to build!

There is nothing wrong with how the JBLs sound in their current position, I also thought I made it clear that I can't change the room. Weights is the reason I want the loud speakers. It is also the only room where I can play loud music... If you have recommendations for more woofers that I should put in Unibox I will do that but I still think the DSA315 are what I want.

I would go with the DC300s but they are meant for like replacing old speakers and are probably not the best quality for sound.
 
Status
This old topic is closed. If you want to reopen this topic, contact a moderator using the "Report Post" button.