$500 DIY Subwoofer Challenge!

[turbodawg]

Originally Posted by KaffiMann

Yes, but the kit is 2 x LM4780, so that is 2x120W. At any rate, I do not see the need for more power than 120W, 3db increase from 120W is 240W, ~10db increase would be around 1000W total, why on earth would you ever want to use a design that requires more than 120W is beyond me, thermal losses and issues start building up and distortion figures when the voice coil heats up will probably be something to think about. If you make a design that requires 300-1000W to perform that is just wasteful. All the good magic happens at lower wattage.

Huh. When testing my personal 95 db/w/m/2pi Dayton 15" HO project with a high quality stereo receiver that does a bit over 100w into 4 ohms, I was easily able to hard clip the amp when running test tones. Good thing I have an amp that will do 1500w RMS both channels driven into 4 ohms.

IMHO, digital power is cheap and of extremely high quality these days. I have friends running an NU3000DSP in a rec-room setup with installed wiring and some vintage klipsch's, it's amazing how well it drives the load and how clean and musical it sounds. You would never think this was a digital amp. For subs it is extremely accurate.

These days there is no good reason to under power subs.. PA guys recommend an amp with an RMS rating double that of the speaker being used. For the project I presented here, the sub is well over x-max at clipping when used in a big, low tuned box.

[KaffiMann]

turbodawg:

I guess we have a difference of opinion then.
Completely agree with those PA guys you mention, I normally add +3db to the max power rating as well, to better handle impedance variations + a tiny bit extra. So I never expect to run the subs i make at more than 60W peak pr channel (run 2 subs). My subs get very favourable evaluations from audio interested visitors, no problem making hair, couch and clothing vibrate at listening position if that is the goal.

Edit:
If you have unlimited power you do not have to list the minimum required db spec, but you will eventually run out of cooling or begin having issues with heat/air speed/compression/turbulence when using small boxes. It's a free world, we all make different sacrifices.

[Ile]

Hope you accept also small IB design, there is no physical box so it will go to specs.

Needed back space is just 2-5m2.

two
Dayton Audio IB385-8 15" Infinite Baffle Subwoofer 295-455

and
Dayton Audio SA100 100W Subwoofer Amplifier 300-802

$349 total without shipping, so there should be enough left for shipping and even to manifold material.

Not us in Europe though, maybe $700-1000 would be total after shipping, few % customs and 23% taxes.

[Juhazi]

I have real life project going on and it can be made to match the competition criteria.

PE sells Dayton DCS380-4 at discount 89,97$ each, take 2 - 180$
PE sells Yung 200W plateamp with boost - 89$
2 sheets of 21mm spruce plywood, glue, saw, screws est. 230$
Total material cost 500$

in Europe it's more than that in € (+cargo always)

The box - appr. 470 liter closed, inner bracing. Yes, you can call that big, but I made 2x235L boxes
Drivers in serial connection
Q= 0,7
Xmax reached at 120W power at 20Hz
spl 103dB@20Hz or 112dB@50Hz according to Unibox

my project A bench style cabinet subwoofer 2x15 closed
My costs for the bench sub is roughly 1000€

In the attached graph there is a smaller box, but it measures the same as 470L

[weltersys]

Originally Posted by turbodawg

Huh. When testing my personal 95 db/w/m/2pi Dayton 15" HO project with a high quality stereo receiver that does a bit over 100w into 4 ohms, I was easily able to hard clip the amp when running test tones.

Turbodawg,

The Dayton 15" HO is only 87 dB one watt one meter at around 100 Hz in half space, and with a 35 Hz tuning in a small box drops in level at lower frequencies from there.

Where are you getting the extra 8 dB sensitivity from?

Art

[turbodawg]

Originally Posted by weltersys

Turbodawg,

The Dayton 15" HO is only 87 dB one watt one meter at around 100 Hz in half space, and with a 35 Hz tuning in a small box drops in level at lower frequencies from there.

Where are you getting the extra 8 dB sensitivity from?

Art

The very agressively tuned port and the quoted parameters siming different than the quoted sensitivity.

I am quoting the sensitivity peak at 40-50hz, it's at about -3db @ 34hz and -4 db @ 100hz, off the top of my head.

[weltersys]

Originally Posted by turbodawg

The very agressively tuned port and the quoted parameters siming different than the quoted sensitivity.

I am quoting the sensitivity peak at 40-50hz, it's at about -3db @ 34hz and -4 db @ 100hz, off the top of my head.

Using the Dayton Reference 15 in a simulation of about 90L net (20x22x16 using 3/4") 2 volt with a 35 Hz Fb does show 95 dB at 60Hz, 91.96 at 40Hz, but only 89.34 at 35 Hz. At 30 Hz it shows 85 dB.
Hornresp is very useful, but it often shows peaks larger than reality, and drooping upper response (or more LF response) than real measurements.

Comparing the real measurement of my Lab 2x12" to the sim, the sim shows the response at 40 Hz to be some 10 dB higher than 500 Hz, while it is actually 2 dB less.
I built the Lab 2x12" before using Hornresp, that peaky port "boom" Hornresp predicts would have scared me off.

If the same holds true for your Dayton Reference 15 in a 20x22x16 35Fb box, it may be closer to 87 dB than 95 dB sensitivity.

Without actual measurement, we won't know what the reality is, but having done the sims myself I know why you think your cabinet may be 95 dB one watt one meter.

Art

[turbodawg]

Originally Posted by weltersys

Without actual measurement, we won't know what the reality is, but having done the sims myself I know why you think your cabinet may be 95 dB one watt one meter.

All I'm doing is stating my horn resp sim figures, as many other people have done for other projects, as a point of comparison. I have also simmed in WinISD and response is much less peaky. But the peak is right where you want it for a house curve and overall grunt.

I stand by my figures. In WinISD the peak is about 45hz with -3db @ 35hz and 100hz. Hornresp 95 db/2v/1m/2pi @ 40-50hz.

[weltersys]

Originally Posted by turbodawg

All I'm doing is stating my horn resp sim figures, as many other people have done for other projects, as a point of comparison. But the peak is right where you want it for a house curve and overall grunt.

I stand by my figures.

You can stand by your simulations, I'll stand by my measurements.

[turbodawg]

Originally Posted by weltersys

You can stand by your simulations, I'll stand by my measurements.

There are plenty of designs on here that have been simmed but never formally measured. If I ever buy the proper equipment I'll be sure to do it.