DIY Class D "portable" speakers

Status
Not open for further replies.
Been looking at various threads here for a while now (read all of the boominator posts) and I think it is time to start putting a plan together for a build of my own.

Years ago a did multiple builds for home, car and my first home theater. But a lot has changed since then also because of the class D amps.

I did like most of the ideas of the Boominator design and choice of hardware. But at the same time the boominator was first and foremost a portable party boombox chasing party sound level over refined top end and deep low end.

From other projects I have too much left overs in form of resin and carbon fiber. I also have access to vacuum infusion so I think I will try to mould a cabinet of some sort of carbon fiber sandwich. CF is tough and has little to no flex giving speakers optimum work conditions.

I have previous used Seas, and Scanspeak elements with great results. But as I have been out of the game for a decade or more I am not up to date. I am not set on final config. 2 way system or 3 way? Both have benefits and drawbacks.

For speakers I picture using some long throw speakers that sounds good in small space cabinets. Not sure if I will use master/slave bass or harvest the benefits of woofers playing back to back.

Regarding class D amps I am green and totalt noobie. I have no idea what to look for or who has good reputations. What I have seen is the wealth of different class d boards and at great prices. Anyone feeling for giving advices is welcome.

As the plan matures and make the choices for hardware I will make sure to update thead. For now I hope to get feedback for usable hardware for my project.

Sound quality over quantity. Not looking to do a boominator although I was inspired by that thread.
 
Im looking forward to hear your plans and execution as im looking into doing something semilar

Im saing
compact transportable
including battery power (and out let for phone charge)
Bluetooth
mini jack input

I got this tip from another forum on speker design that I like

http://www.lencoheaven.net/forum/index.php?topic=19778.0

Build into one case with a center of batteries, dock , amplifier all build in wood (im actually thinking MDF or ply)

And I thought about changing the speakers to a set of Fostex FE 83 en that I know sounds great (in my opinion)

Class D Im thinking TPA3118D2 based thing

Im thinking about hos to make the bluetooth part but I know that SURE has some.

how does that sound
 
I'd be tempted to go 4-way fully-active just to make it that much easier for each driver and more controllable. 4 amps per full-range signal with the very bottom one possibly summed to mono to make a 3-way mains + sub arrangement. That's pretty much the practical limit in terms of quality vs. complexity.

Short of finding an anechoic chamber to calibrate them in (probably not going to happen), the next best thing would be outside on a calm day away from traffic, or a very carefully defined procedure to ensure that you get a good measurement before the first reflection arrives. The whole point is to prevent reflected sound (room reverb) from affecting the test. Make the speakers sound as good as they will on their own, and then use them anywhere you need without modifying the calibration. Any new problems at that point are from the environment or the placement, not the speaker itself, so don't try to "fix" the wrong thing. Ears and brains are very good at separating direct sound from reverb; microphones and software are not yet.

I want to build an outdoor live concert set on a 50 pound budget per stack, so I'm interested to see how the CF thing works. Never used it myself.
 
Producing ample quantities of deep bass in a small box without drawing too much power would be my main concern. I'd choose a FAST system(fullrange with a woofer). I chose mono sound in my last boomblaster build and that helped with keeping the budget small. Also the filter cost in a FAST setup would be really low. Just a single cap for the fullrange and some filtering for the woofer(or a bandpass box without filters->really cheap). Funny that you mentioned a master/slave cause I used that principle in my setup. I used cardboard concrete tubing, tube/5mm rubber sheet inbetween/tube and active/passive speakers in each end. Plays pretty loud and low enough for bass guitar🙂 The active part is a high end car component 2 way setup with tweeter and a 5.25" speaker.
The slave was made from an old bass speaker. Sawed off the magnet and reinforced/added weight to the cone-->free long throw slave 🙂
I have heard that this board has decent sound; TPA3116 2 50W Class D Digital Amplifier Board | eBay. If you bridge this amp and build a low ohm speaker setup that will give you plenty of power( handles 2-4ohms in bridged mode).
Maybe you could build a slightly hornloaded/recessed fullrange in that carbon enclosure you mentioned. That would give you cleaner midrange at high spl and you could choose a smaller cheaper fullrange.
Most of the bluetooth cards I've tried doesn't cut it! I've tried 2 diy bt 4.0 aptx cards and both had big problems. This one is the best I've found so far: Bluetooth V4 0 NFC Aptx Music Receiver Adapter Handsfree Car Aux for iPhone 6 PC | eBay. Plays loud and has a full rich sound which is what you need match the acoustics outdoor. Easy to pair and gives about 3 hours on the inboard battery. Runs off a 250mAh battery which easily can be soldered to something way bigger..
I hope this gives you some ideas and I tried to utilize the fact that you can create an advanced speaker box but with simple low cost electronics which would give really good sound 🙂
 
Last edited:
The tpa3116 can also be driven from a 12V car battery or even lower voltage 🙂 It's just less powerful at lower voltage.. To produce deep bass from small speakers you need all the power you can get from higher voltage batteries. Size of the inductors also matters when pushing large currents trough them. They distort the sound pretty bad when overdriven. Chokes are expensive so I like to buy class D amps with decent size inductors from the start.
 
I have yet to get started with the project. Still doing CF work trying to find the best sandwich material. I get some sheets of balsa woods tomorrow and that might work.

Has anyone any idea of the bandwidth usage for flac files? Someone told me blue tooth might not work well with flac files. What other wireless standards are possible to use? Would prefer not to use wifi. But for all I know maybe there is no problem at all transferring music in flac format via blue tooth. What do people think?
 
The tpa3116 can also be driven from a 12V car battery or even lower voltage 🙂 It's just less powerful at lower voltage.. To produce deep bass from small speakers you need all the power you can get from higher voltage batteries. Size of the inductors also matters when pushing large currents trough them. They distort the sound pretty bad when overdriven. Chokes are expensive so I like to buy class D amps with decent size inductors from the start.


I really liked this - have you tested the tpa 3116? Wonder if one uses one of those L speaker and one for Right speaker and transferred via blue tooth would one experience any out of sync or phase errors?

That amp was so nimble it seemed like a good choice if someone here have tested it.
 
4 ways ? or 3 ways with sub and multible amps ?? for a simpel 50 pound budget ??

Sorry I didn't get to this sooner. Somehow I missed the e-mail about a new post.

Anyway, for a mono system, a single 4-way active crossover should do the trick. Separate amps for sub, low, mid, and high. Possibly two amps in bridge mode for the sub and maybe the low also. So a 4-way mono system with a minimum of 4 amps, 1 for each frequency range, and no passive crossovers at all. All 4 drivers and their amps, corrective EQ's, the crossover, and the input buffer/gain control (in that order) can all be in the same box for external simplicity.

For stereo, just duplicate the mono system so you have one on each side. Then realize that the subs are hard to tell apart even if they do get different signals, so they can be combined. The result is now 3-way stereo mains + mono sub, but it may be simpler for setup/teardown to have stereo subs anyway because it allows a single audio cord to each box. For bluetooth, you'll probably have a third box anyway for a stereo receiver, to allow some decent separation between speakers, so you might put a mono sub in there and send a highpassed signal to the stereo 3-way mains.

I doubt that this could all be done in 50 pounds using more traditional materials like plywood or MDF, but since the OP was thinking about carbon fiber, it might just work. I'd like to see what he comes up with too, because I want to do something similar. I'm thinking horns for mine as much as possible, which should boost the efficiency and get the drivers out of the rain at the expense of some more design challenges. And this design for amps, also to keep the efficiency up while being able to mount them almost anywhere.

Given the OP's desire (and mine too, to an extent) for quality over quantity of sound, I think the 4-way idea is worth the apparent complexity of construction, depending on the OP's application. (apparent because anything becomes simple if you break it into modules that can be understood on their own and then assembled) If it's designed for outdoor use or for a large room, I would definitely go 4-way active for two reasons:

  1. Sound doesn't come back outdoors like it does indoors (reverb), so you need a lot more power to get the same volume, and you can only get so much from each driver. So you'll end up arraying them anyway, which gives you the opportunity to optimize them for different frequency ranges.
  2. No one driver can cover a wide range in wave-mode by itself with decent quality. In pressure-mode (low frequencies in small rooms, or almost anything in headphones), they're more forgiving, but in wave-mode (high frequencies in big rooms, or anything outdoors), you're more dependent on the drivers' inherent properties to do a good job. Since you're going to array them anyway for power reasons, you might as well get a variety of drivers that are optimized for specific, overlapping ranges and set the crossover frequencies and rates so that the signal to each driver goes away before the driver becomes "bad". Then tweak the individual corrective EQ's so that each driver's in-range response matches up with its part of the desired overall response*, then match the levels to each other to finish the overall response and seal the settings.
* I'd suggest no EQ at first, but expect to add one between the crossover and each amp. Measure the raw response, then design a fixed EQ to drop into each slot. No need to ever change it unless the driver changes, and then only to compensate for the driver. Don't include room EQ here, or anywhere really unless you're having trouble with feedback squeal and you're okay to sacrifice a bit of quality for quantity.
 
Last edited:
1 amp per speaker seems smart.

Btw I just heard that blue tooth is not that great for transmitting hi quality audio signals. I was told that BT should be like my third option if I for some reason could not do one of the other two.

I do remember one option was regular wifi but I can't remember what was "the best way" to transfer wireless signals from ie phone, pad or whatever to the active speakers.

It seems BT does not have the bandwidth to transfer full cd quality and even less chances to work with high resolution audio.

So what is the best way to send wireless signals to the speakers?

Or should I always use wired connection for listening to high resolution audio or whenever I play lossless music?

I could let the BT be a fallback option for special occasions - like if someone visits and they would like to show off a certain playlist or whatever.

What you think, are BT so terrible as the audiophile people claims?
 
There are some pitfalls but generally modern bluetooth devices deliver a high enough sound quality that practically no audiophile can tell the difference from the original. Let's assume we're talking bluetooth 4.0 or later on both the sender and receiver end otherwise there can be other problems.

A2DP. This protocol can send any file coded in a supported compressed format to the receiver in a bit-for-bit state. Remember both must support the compressed format for this to work. Up to 800kbps bitrates are supported (256kbps with 4.0, 320kbps with 4.1 and 800kbps with 4.2) and the usual formats supported are mp3 and aac (flac and alac with 4.2).

ATPX. This is a standard to replace the mandatory SBC audio encoding layer in bluetooth when the music is sent in a format that is not supported for bit-for-bit transfer by both sender and receiver. It's developed by the broadcasting industry so audio quality is supposedly very good. Bitrate is 352kbps in a format similar to mp4 or aac.

Remember that you can't get better quality than what you have already. If your files are in 192kpbs mp3 then it doesn't matter how good a bluetooth you have the output quality will obviously be that (or lower if unsupported and it has to be transcoded to SBC).

You and your friends might say they can always tell the difference between lossless and encoded. That's probably just not true though. Less than 0.1% actually can and then they have to listen of high quality headphones, on speakers, I'd say it's impossible. But you and anyone can test themselves in a blind ab test if they like or dare. Follow this link: foobar2000: Components Repository - ABX Comparator and have them post their log files so you can see they're not cheating.

EDIT: To clarify before you ask. Most modern phones, laptops, etc supports aptx and various encoded formats except iphones and other apple devices. They only support their own aac format and transcode everything else to low quality SBC. They will not support alac either even when bluetooth 4.2 is enabled due to them only supporting lossless audio over their own proprietary Air formats.
 
Last edited:
I agree with saturnus I personally will not be able to hear the difference especially on a potable system as this threat is about,

I must say that my own ambition in the transportability range is more in direction of "boominator mini or micro" but purpose build for indoor use playing in one direction (am I right that the boominator project is build for outdoor and projection in both direction , front and back).

I know that people have done something similar and I'm looking to learn from this I hope before making my system and when I start I think I will make 4 or 5 to use in different places.

To go 4 way stereo and 10 amps is not an option for me at all.

If you know of a project done I will appreciate a link.

N
 
I haven't tested it but I think a lot of people in the tpa3116 thread have used it. It is a design from DUG here on the forum I believe. The tpa3116 is the most attractive class D amp at the moment for 50-100W (2ohm). I have several tpa3116 and its the way they are built that decides if they are good or bad. The chip itself is superb. It has analogue in (no digital) so you need something with a dac to make it work. There's some tda chips you can consider aswell but I think the tpa chip sounds the most lifelike.
The BT version I mentioned has a bit to much bass, so I think I could separate it from a cable in a test. I agree with keeping it your 3. option. There are some wifi alternatives that is lossless(airplay) and (dlna?). But to me those alternatives aren't as elegant in use as BT.
 
Any tips to what is good for sandwich cf bits? I got a partial! advise to look for some honey comb material but the guy could not remember the brand name nor the material it was made from. According to him honey comb was almost a universal super structure that would be perfect for such use. Any idea what I should look for?
 
I was looking for various TPA3116 amps, then I came across this TDA7498:
Yuan Jing Audio - 24Bit / 96KHz USB DAC Power Amplifier [80w x2] TDA7498 + TE7022 + CS8416 + CS4398 + SGM8054 - USD $128.90 - Finished Units


Seems to have it all incorporated. DAC, preamp etc. How is the sound quality compared to the popular TPA3116?

Think it will do well in 4 Ohm setup?

This ready made also seems like a bargain. 2 x50 w + 100w for a sub.
http://www.ebay.ca/itm/Finished-TPA...r_Parts_Components&hash=item19e9fa47b0&_uhb=1

Any one tried these head to head for sound quality?
 
Last edited:
Finally got a solution for additional material to use with the carbon fiber. I came across a large slab of aluminum for next to nothing. So I ordered it right away. I figure I will use aluminum in the front and rear, with one layer of fiber glass (insulator to shield the aluminum from carbon fiber as they don't mix well) and then a few layers of carbon fiber. If there is enough alu material left I will make an inner skeleton to make sure no flexing is possible. If I run short on alu I will make skeleton from carbon fiber.

I am not set on the layout/design. I've been thinking I would do 2 "bookshelf" active speakers. But that might not be very portable? What if I make some quick connects so I can easily clamp the two together before going to the beach for a BBQ, or to chill on an island? Maybe the two speakers could clamp to a third mid section that I can fill with 18650's li-ions? Any ideas for a quick connect that might work? At home I could then remove the battery housing making the speakers with higher WAF.
 
Last edited:
Status
Not open for further replies.