Portable Bluetooth Speaker Build With Passive Radiators Build

Status
This old topic is closed. If you want to reopen this topic, contact a moderator using the "Report Post" button.
Hello everyone

I've been here a long time but more on the reader side, today a decided to share the new portable speaker I've been working.

First, this probably the 5th portable speaker I've built, so its a culmination of everything I've learned on previous build plus a few bookshelf and sub-woofers.

The drivers will be the same as my previous box, two 4,5" Eastech mid bass and two Peerless DQ25SC16. No wait, 4 of each, since I will build two almost identical boxes, one for me and one for my father.

This is what I wanted:
-Smaller than the previous one (10l max vs 21l)
-Play way lower

This mid bass was something golden that I found on a electronic stores, bought it what parameters whatsoever. After measuring I knew it would perform well

Bellow is an simulation in Hornresp of what I came up. Extension to almost 40 Hz in a 3,5 l enclosure. The way to do that was tunning the passive radiator really low as a mean to keep excursion under control, and them just brute force trough equalization.

Response @ 2pi
An externally hosted image should be here but it was not working when we last tested it.


An externally hosted image should be here but it was not working when we last tested it.


In theory it would work, but I wasn't so sure, so I build a test box to measure and more important, listen. My biggest worry is that the high excursion even at low volumes would distort the rest of the spectrum

An externally hosted image should be here but it was not working when we last tested it.



It worked perfectly, I used my Mini DSP to apply the required equalization and took a quick measurement to see what the response looked like, and after that hooked up a tweeter just to listen for a few hours and see if the sound was pleasing.

Near field response of each driver and summed response

An externally hosted image should be here but it was not working when we last tested it.


After that what was left to do was finish the box design and design a filter to go with it

For the box I went with the smallest size that could fit the drivers, the final dimensions is 401x148x183mm, all made to be CNC cut

An externally hosted image should be here but it was not working when we last tested it.


An externally hosted image should be here but it was not working when we last tested it.


An externally hosted image should be here but it was not working when we last tested it.


For the filter I have a PEQ, a 4th order butterworth low pass and a BSC, all adjustable from outside the box by just removing the back portion of the cloth. The PCB should arrive soon

An externally hosted image should be here but it was not working when we last tested it.


An externally hosted image should be here but it was not working when we last tested it.



Today I received all CNC cut wood, my plan is to assemble the box this week and take some measurements to start working on the crossover

An externally hosted image should be here but it was not working when we last tested it.
 
Very impressive build!
Do you have any information about the midbasses? Part number and so on?
I assume you know from experience what is required from your filter PCB if you dare to order it before measuring the box :)

What kind of amp are you going to use?
 
Thanks

The part number is FSB527550-5B24. Unfortunately the 8 ohms versions is long gone and the 12 ohms was available until recently but I went to check and it's also gone. I bought them 5 years ago

About the filter, after I modeled the required filter and tested on the test box with the MiniDSP I am pretty confident that it will work, bellow is the expected filter response. Red with just the PEQ, blue PEQ + High pass and purple PEQ + High Pass + BSC

An externally hosted image should be here but it was not working when we last tested it.
 
Hopefully it will, the drawback is that the overall level is reduced a lot in order to keep under the excursion limits, but I don't need it to be super loud. The previous version used 4 of these drivers, but I can count on one hand how many times I used them to the full potential
 
Today I printed and started assembling the passive radiators, still need two order 4 more surrounds for the other box since the initial plan was only one PR per chamber

2q3yk37.jpg


Also put the first speaker together and put on some threaded inserts for mounting the drivers

30kycux.jpg


3321c37.jpg
 
Last edited:
A few updates:

Tested the acoustic cloth and how to attach it using a notch on the back, I didn't like the looks and will use a plain black cloth, unfortunately it was not cheap

An externally hosted image should be here but it was not working when we last tested it.



Today was measurement day, in the pic the speaker on the jig I made to accurately set the angles

An externally hosted image should be here but it was not working when we last tested it.



The grill is causing some serious problems with the tweeter response, I will raise it from the baffle so it become flush with the grill, but them the woofer whole can also cause some problems

Also my impedance measurement jig is somehow measuring about half of the actual impedance, I don't know if I can trust the data by just doubling it

An externally hosted image should be here but it was not working when we last tested it.
 
After another round of measurements I have a crossover, i may tweak it a bit

I'm not sure about the Dz of the tweeter, following Troels methodology (Time-Alignment) I arrived at 33mm and adjusted the crossover accordingly, but them I can't get a deep notch at the response anymore, I guess that's the price for not having the drivers acoustically aligned

Any sugestions?

Normal

An externally hosted image should be here but it was not working when we last tested it.


Out of phase

An externally hosted image should be here but it was not working when we last tested it.
 
Subscribed.

I am onto this as well, see if you can source Monacor SP-60 - these have very low Fs, suitable for sealed and 90db/w (and are cheap :D). They require a large box, so small sealed with active help should work. Just an idea.

Why would you go for passive crossovers and not active ones (smaller, cheaper, can boost bass output, lighter)? Something like a TPA3255 I believe can do the job and run quite long on batteries?
 
Just looked into it, they look nice, a bit pricey though

At this point I kinda wished that I went full active, the PCB would be a bit more complex but nothing I can't manage. The amp for the tweeter can be cheap, 10W would be more than enough since it's way more sensitive

I will simulated an active crossover and see what it looks like

Also the PCB's arrived a few days ago, they look beautiful and very high quality, for the price it's a steal

An externally hosted image should be here but it was not working when we last tested it.
 
With help from you guys (Crossover Design - When to use driver offset? (Dz)) I've made a few modifications and optimization to improve the power response and also manage to remove one inductor while keeping the response as close as possible to a LR 4th order acoustic

Inductors are ordered, should be a few days to arrive since they are custom made. For capacitors and resistors I will see if I can source them locally

2vtabk5.jpg
 
Looks very good!
How much EQ are you going to add in the bass frequencies?

Expected is 10dB at 40Hz with a Q of 1,12 if I remember correctly, but that should be to achieve a perfectly flat response, let’s hope that I can get a full sound with less because the more boost I add the less maximum output I can get before running into power and excursion limits

What would be great for this box is a multiband compressor, this way I can limit the bass without compromising the maximum output. It’s what almost all portable Bluetooth speakers do

Unfortunately the cheapest way I’ve found to that is using two Mini DSP 2x4 per speaker, and that still too expensive
 
That's exactly what I wanted, you can easily do that with the multi-band compressor

Unfortunately I didn't find any stand alone board for that, only using multiple mini DSP's or integrated in pro amps like the iNUKE

If anyone know how to do that let me know.

Only on the finished box the max volume can be determined, the plan is to adjust the gain using lots of songs, mainly bass heavy ones so it never runs into clipping no matter what is fed

Also finished gluing the wood, after sanding the corners it looks amazing, honestly I didn't thought it would end up this good

Now only need to sand a little more and apply varnish

An externally hosted image should be here but it was not working when we last tested it.


An externally hosted image should be here but it was not working when we last tested it.
 
It works!!!

At first I tough the filter wasn't working, but the trimpot is a 30 turn one and I had only turned 5 or so

The amp needs more voltage as it is distorting a bit, not sure exactly what I will do, for the stationary one is easy, will order a 28V power supply and should be good

the tweeter is not wired yet

The audio recording quality is not great, when the crossover is ready I will use the UMIK

YouTube

2lnfod2.jpg
 
Status
This old topic is closed. If you want to reopen this topic, contact a moderator using the "Report Post" button.