DIY bluetooth speaker

SentinelAeon said:

It's not as simple as that. My first goal was to make any bluetooth speaker and was very impressed that i could do it. And my first 2 speakers were very very loud. I was satisfied wih the speakers but my friends were very obsessed with bass. So because of them, my next goal was to make speaker with bass, remain loud and be very cheap. And it seems with last 2 speakers i accomplished loud and good bass.
Now the speaker sounds good to me. But since audioguru noticed something that i couldnt, since he obviously know a lot more bout audio than me, i was wondering what he noticed. So logical next step is to improve sound quality.

I am learning as i go. And while attitude of some (not you gophert) in audio comunity was in lines that i barely have a RIGHT to make my own speakers until i know all the theory, i belive i have a right to make fun things, do mistakes, learn as i go (afterall its my time and my money i spend).

So to sum it up, i would love to know what intermodulation distortion and doppler frequency shift is and how it sounds, so i can correct it, as audioguru mentioned i can do that with cutting woofer from playing high frequencies.

Click to expand...

From my point of view, iterative development is the best development. You get to learn how the various versions of bad sounds from first-hand experience. Everyone else will have to read their test books, ompare speaker dimensions to your dimensions and diagnose your audio issues based on on-line calculators before it has even been built. They have no chance of knowing what the specific distortion sounds like since any audio will only be as good as the online calculator and the class D amplifier & speaker in their iPad or phone.

Good luck, enjoy learning from real experiences (no sarcasm intended)

You have two problems that are causing distortion:
1) The woofer is too small.
2) The power is too high.
Did you see the video? The tiny woofer is moving way beyond it being linear.
Because the amplifier input is Bluetooth stereo, you cannot easily add the 2 channels to make mono so you have the weird arrangement of the woofer on one channel and the tweeter on the other channel.

Will i solve anything adding a coil to cut it at about 2800Hz same as tweeter, just the other way around ?

As the size goes, cant go any bigger if i wanna remain portable. And as the power goes ... the song in video is special bass compilation. With normal techno music i have to put it to maximum to see any movement. Oh, is it the wrong time ti mention that right now i use amp that gives 20W per channel to this speakers and i soon plan to test it with the one that does 50W (when i get 4 ohm version) ? Dont freak out, its all good fun

ps: i just watched some Sony and JBL disasembly speaker. They are small as it is, but when u see them opened, they are basicaly full lol, there is barely any free space there with all the electronics. I am amazed that they get such bass from them. I mean obviously, i blow them away with my big speaker that is 5 times cheaper but still, what an acomplishment on their side. If i did a speaker of such small size, i would prolly only hear 1000Hz and up xD

An inductor in series with the woofer cuts high frequencies gradually because the woofer is also an inductor and the two inductors in series form only a simple first-order lowpass filter. At 2800Hz the datasheet shows that the impedance of the woofer might be 12 ohms and a calculation for the total inductance would cause 2800Hz to be -3dB, 5600Hz to be -6dB but still pretty loud, 11,200Hz to be -12dB (very audible) etc.

Well i will postpone this woofer capping for a while.

I just wanted to let you know i tested the wooden speaker again today, the one with dayton woofer, tweeter and a calculated enclosure and port. First of all, i MIGHT have gotten the enclosure a tiny bit to small, see i cut 2 side pieces correctly and 2 of them not correctly (this always happens when i am trying to make a square out of wood and i just cut the length i want, not subtracting double the thickness of sides. ). So next time i will be more careful to draw it all onto a paper, double check, etc. BUT. BUT.

I played a song from the prodigy and wanted to see the difference between normal usage and how it sounds if i cover the port with my hand. And it makes a HUGE difference. Really really really big difference. Before i was just testing this ultra bass song on the video and didnt notice much difference since bass was super deep anyway. But on prodigy song, the difference was huge. U might find some other reason for this but for now i am sticking to the version that it IS like this because this time i properly calculated the enclosure. And it feels really really good that it works. So dont go 1 note bass on me this time

I think this will make a nice speaker and in future i can work in improving the sound quality, etc.

It is good to hear that your port works well.
A speaker sounds better if it is lightly stuffed with sound-absorbent material. I use fiberglass home insulation. The stuffing also covers up the hollow sounds if the enclosure not calculated with The Golden Rule ratio for its dimensions.
Don't let the stuffing or its fibers get blown out of the port.

Oh 1 more thing. 1 of my goals in the future is to build a more portable speaker consisting of dayton TCP115. The speaker from wood i made is about 2.5L altogether, but i choose dimensions poorly so its very high, probably to high to comfortably put into backpack. It should be a lot wider. I will check the dimensions of those JBL extreme and sony speakers and try to be within the limits of them, see what bass that gives me. That is also the reason i am prefering thin materials - speaker is less bulky in the end. Sony SRS-XB41 is 291 x 104 x 105mm. If i used really thin material, that would give me 3.1L of volume, which is more than i have now, quite a lot more. So maybe in this case, a metal frame would be good ? I am not sure what contributes more to vibrating, the weight of the material or how stiff it is. But maybe with braces .. Then to further optimise it, i would use wires less long. Oh and the thing u prolly will not approve of ... shorter port with smaller diameter (all the calculators and guides i checked, the bigger the diameter, the longer the tube). Anyway this is my goal. But, what if we want to push the physics to the limits ? Look at the graph i attached of a stupidly small enclosure. 1.5L !! Granted not very low but stil, 70Hz, isnt that what JBL charge 4 has ? I would love to hear how it sounds like this.

And 1 last thing. Though i prefer port with daytons since i can actualy simply calculate it. But i am still, as an engineer by degree, very interested how it would sound with passive radiator, not tuned or calculated, just picked by stroke of luck. I cant wait to get parts i ordered to play around, learn, test.

Again, I played 69Hz and 99Hz on my cheap 3" ported computer speakers and both frequencies played at the same loudness and were not deep bass sounds.
I am not a singer but I can hum as low as 75Hz. The cheap computer speakers play well down to 48Hz and sound pretty good for most music.
I can play musical and generator sounds in my brain by thinking about them but I cannot think what one octave below 48Hz sounds like and I don't have any speakers that play 24Hz.

I just read that Meyer Sound has a column speaker that produces frequencies down to 13Hz for movies with cannons. Worshippers can "experience" the 3Hz vibrations of a few church pipe organs. That is what is real BASS.

I just put my old Visaton FRS 8 m into the QSpeaker and it seems there is an error. I imported the data from that hifi site so i dont manualy have to set it. And no matter what resonant frequency i set, the port diameter stays at 0.5cm and port length stays at 0 lol. And even though port and volume stays the same, when i change resonant frequency the graph changes. Which is well, impossible since i didnt change any parameter lol. I think i will download another program called WinISD, just to double check my data everytime im making calculations. I should get used to WinISD anyway since it has soooooo many other data like how port diameter affects that airflow hiss, etc. I just choose QSpeaker because its really simple and i needed an easy start. But god damnit, if i can program complicated apps and control embeded systems, i should be able to copy paste a few damn parameters into a program

Btw this site is pretty cool. Gives you all the data u need, u can export data into well known programs like QSpeaker and WinISD and also gives u some similar speakers to compare:

It seems that your enclosure software is all screwed up about seeing your tiny hole with no length. It is a sealed enclosure with a tiny hole, not a port. A port needs length plus area to make a new resonant frequency.
I did not calculate it but the large 7l sealed enclosure probably resonates with the speaker in it at around 141Hz then its response drops below 141Hz but nowhere near 62Hz or 66Hz.
The dropoff rate on your graph is -44dB per decade of frequencies but a ported enclosure is at -80dB per decade. The dropoff rate for a sealed enclosure is -40dB per decade.

My guess is that something got messed up at speaker data import, cause it calculated correctly for daytons, i verified that with the other software. Well, a good thing i wont be using visatons for bass.

Just for fun i tried WinISD and when i import speaker data it tells me speaker data integrity didnt pass the test or something, but from there on it gives me a green light. And this is what i get at everything default, only changed port diameter. I heard elsewhere u can actualy get some bass from those visatons but in a big enclosure

I just verified data for dayton and it gives me same graph as QSpeaker. Oh and just for fun, i checked the graph for air velocity, it should be under 20 and when i input 30W and my dimension, i get waaaay over 20. Good thing i am it seems deaf to both this air hiss and to the high frequency modulation. It seems doctors are correct that i am 2% deaf. Only in 15.000Hz and up so far but yeah

Your new port is enormous. If the response is -3dB at 75.5Hz then my cheap little computer speakers make much better bass.

Its not my port, its only a program I am not defending the visatons, just wanted to see how low u can go with them. I still think them nice speakers instead of tweeters. Have a bunch of visatons i have to use, i am just trying to get rid of things i dont need

HOOLLLYYY what is holy and what is not. First of all, i sold 1 of my smaller speakers and i was sure its not very loud and has bad bass. I cranked it to 74% and played prodigy - smack my ***** up. And its reaaaaalllyyy loud and the bass is good considering speaker isnt some very low bass, it has air leaks, etc. The person who was buying it, the moment the bass hit was like wwwwoooww, soooldd and kept laughing. Comparing to what u get for 40 euros in store, u cant hear it over my speaker, i guarantee that.

Then just for fun, i decided to do the same thing with my bigger speaker and crank it to 100%, play smack my ***** up from prodigy and well, the beginning of the song with no drums i thought to myself, good thing i only charge 60 euros for the big one as opposed to 40 euros for the little one, since this big one is only like what, 30% more loud ? But then the bass hit, i swear to god i never felt anything like this. It was like hell was unleashed in a second and the whole workshop shook and me and buyer were like wooooooooowwwwww. I cant belive i managed to make something like this for such cheap price.

I am really happy, i have found the perfect enclosure size, i have found the speakers, actualy im looking forward to seeing how much i will sense the extra 5dB from 4 ohm version. As for tweeter, i just picked the cheapest one i could find, if i find the quality to not be good i will obviously switch. Now i have to work on my woodworking skills, its hard to cut straight with jigsaw and have to make speaker pretty to sell it

Many of the spec's for the 8 ohms and 4 ohm speakers are different so the sizes of the enclosure and port will be different, if you want the best results.
The 4 ohm speaker will be 4.9dB louder than the 8 ohm speaker at the same amplifier volume control position. If the amplifier can produce its full power into 4 ohms the maximum output will be almost 7.9dB louder than the 8 ohm speaker. A TPA3116D2 amplifier can produce 40W per channel into 4 ohms at fairly low distortion with a 19V supply.

Here are the spec's of each speaker from Parts Express (don't look at the simple copy of the frequency response graph that seems to be only for the 4 ohms speaker):

Very similar graph even though data isnt the same. The current amp is 2x25W, the one from the video and the one i tested, though i think it peaks at 20W per channel, most i measured. The amp i want to use in future is 8 ohm 2x38W and 4 ohm 2x50W. So we will see

Btw i was checking for some ultra cheap speakers yesterday and found something amazingly cheap, small that still produces bass. Now granted, the graph is a very bad one, we would want a linear graph. But look at how deep it goes in only a 1L enclosure. Am i missing something ? Because with this, i can make ultra ultra cheap speakers. This speaker is 6.45€, can you belive it ?

Parts Express shows that the Dayton CE65W speaker is tiny (65mm which is only 2.5" in diameter), has a max continuous power rating of only 7W and has a low sensitivity of only 83dB at 1W/1m. Its fairly high frequency resonance has a high Q causing the 6dB peak at about 150Hz. One Note mid-Bass. No deep bass.
Parts Express says -3dB at 60Hz in a 0.1 cubic feet (2.83l) ported enclosure. Maybe the 1l rating you found is wrong.
It looks like a cheap Chinese speaker sold by AliExpress and Banggood.

-3dB at 66Hz. I tested it in both winISD and QSpeaker, same result. Do you trust me or parts express ??

I know it is tiny, has bad sensitivity but please, keep in mind other speakers of this size have -3dB at 120Hz. I put so many different speakers and simulated ported enclosure. Tried even speakers that cost 15-20 euros. This speaker is 6.45€. Yes, you read that right, 6.45€. And its not cheap aliexpress or banggood, its the mighty dayton. I will test them when i get them. Today i got a bunch of nice mono 5W amps and tested them, work great. That speaker would be perfect for a small speaker and still produce bass. Sometimes you are just to harsh.6.45€, 2.5" and F3 at 66Hz. If that doesn't impress you, you have no emotions !