Mobile Mini-speaker project questions

Thread Starter

sHanQ

Joined Oct 2, 2017
7
Good day ladies and gentlemen o/

I have recently started to plan my new project for a mobile speaker i want to build from scratch and i could really need your help regarding OP-AMPs and detailed speaker information/clarification.

I do not want to buy a finished Amplifier board. I'm interested in learning that stuff and building something i can be proud of.

What i have planned and other information you possibly need:
- Size should be roughly 15x8x8(cm) / ~6x3x3(inch) (so roughly like the Bose Soundlink Mini II)
- 3.5mm audio signal Input(phone, mp3 player)
- 2 way Stereo
- Should run on rechargable Li-Ion Battery pack(s) with a minimum playtime of 4-6h on full volume.
- highest possible audio quality and power efficiency
- price can be up to 200 USD as i prefer having a good result i can be proud of.


What i need/know so far(correct me if im wrong):
- 2x 5V Batteries for +Vs and -Vs with ground in between for OP-AMP supply
- 1-2 OP AMPS(depends what type)
- Low/High-pass filters
- 3.5mm Socket
- PCB + passive components
- Drivers(4 ohms) if i'm going with +-5v

Whats still unclear or needs confirmation:
- What type of OP-AMP should i use. by going through this forums i found out the OPA2134 seems to be good. As far as i know this one cannot be used for stereo(unless i use 2) or rather a OPA1679 for stereo use?
- What supply voltage should i use for the OP-AMPs? Will +5V and -5V be enough?
- I have created a small circuit just for testing the OP-AMP and the voltage/current i need. With a +5V/-5V OP-AMP and a 4 Ohms driver i only end up with 6.25 Watts on the driver. Will that be enough? almost seems like it for me..
- Just to clarify: If i supply my OP-AMP with the required power source, it will also forward enough current to the driver, correct? or do i need to support the driver seperately?

This is my first - very simple - test for the 4 ohms driver:

with a max Voltage of 500mv given by the audio signal source and an OP-AMP Supply of +-5V i reach 6.25W with a Current of 1.25A at 5V The driver im currently thinking of is capable of handling up to 25W.

If this project works out - i will post some pictures and details

I'm very grateful for all the help i'm getting from you as this is my first bigger project
 

philba

Joined Aug 17, 2017
960
Sounds like a fun project. A few things pop out. In no particular order
  • No such thing as a 5V battery. If you want to use LiPos (great power to weight ratio), then you get multiples of 3.7V (nominal, it ranges from 4.1 down to 2.7ish). You'll need to implement battery management to make sure you don't kill the LiPos. I'd build in USB charging. [edit] lipos go from 4.1 down to 3.0[/edit].
  • I would not run the project straight off of batteries.
  • 5V seems really low for an audio amp. Take a look at designs for autos that run off of a 12V single rail supply. Build/buy a power supply that gets you at least 12V out of your batteries. You'll be able to get the same power out of your amps but with lower current. Maybe even consider a higher voltage. There are dual rail PS designs out there.
  • For stereo, just design/build two identical amps/pre-amps.
  • Think about your box early so you can make sure all the pieces fit together at the end. You'll tear your hair out over shoving it all together at the end if you don't. This especially true if you want good acoustics (though it's all going to be a compromise for portability).
  • As a corollary to the above, think about your controls/connections layout for ease of use, accessibility (and appearance). It sucks to have a great project with hard to access controls or inputs.
  • Do some usage scenarios. Think through how you would use your "product" when completed. Write them down. Maybe even ask a friend or two what they think. That will guide you in a number of areas. I often get my wife to listen to my ideas and she really helps me understand the good (and just plain wrong) of the design. You'll need to check your ego at the door!
  • Document with pictures, schematics, a journal, design docs, ... This is helpful for things like decisions you make about the design. If you need to revisit, you don't have to reconstruct your rational if it's written down. If you do V2 of the project (like adding bluetooth), it's invaluable. and, there are delays built into the process - PCB fab time is a week or more. I find it easier to get back up to speed when the PCB comes back if I've got good documentation.
 
Last edited:

GopherT

Joined Nov 23, 2012
8,012
The output of an OPA2134 is very nice and clean sound but, with +/-5V supply, the output can only range from +/-4V on that OpAmp.

What driver do you plan to use? Note that the supply voltage of a lithium battery pack drops as it is discharged. Typically from 4.6V (fully charged) to 3.0V discharged. If you have some type of a 5V cell-phone charging pack, then some audible noise will be transferred to your circuit because there is internal circuitry in those charge packs to step up the voltage.

Seems like a waste of time and money if you are trying to do something better than what is on the market. The speakers on the market are mostly wonderful because they use classD amplifiers (Very efficient) and they are well designed to get full range audio (difficult for a diy technician) with carefully selected speakers and carefully designed enclosure. The enclosure means everything to the bass.
 

philba

Joined Aug 17, 2017
960
Seems like a waste of time and money if you are trying to do something better than what is on the market. The speakers on the market are mostly wonderful because they use classD amplifiers (Very efficient) and they are well designed to get full range audio (difficult for a diy technician) with carefully selected speakers and carefully designed enclosure. The enclosure means everything to the bass.
I agree that there are packages that solve all the OP's problems. I personally would just find a decent board but he wants to design it himself and I celebrate that spirit! Also, the commercial designs often make compromises just to hit a particular price point. This is amazingly true with portable audio. Certain companies often put out crap and coast on their name with the cheapo Chinese knockoffs being just as good, er, bad. It works fine for them since most customer don't really care about acoustic quality. (B**e springs to mind).
 

Thread Starter

sHanQ

Joined Oct 2, 2017
7
Sounds like a fun project. A few things pop out. In no particular order
  • No such thing as a 5V battery. If you want to use LiPos (great power to weight ratio), then you get multiples of 3.7V (nominal, it ranges from 4.1 down to 2.7ish). You'll need to implement battery management to make sure you don't kill the LiPos. I'd build in USB charging.
  • I would not run the project straight off of batteries.
  • 5V seems really low for an audio amp. Take a look at designs for autos that run off of a 12V single rail supply. Build/buy a power supply that gets you at least 12V out of your batteries. You'll be able to get the same power out of your amps but with lower current. Maybe even consider a higher voltage. There are dual rail PS designs out there.
  • For stereo, just design/build two identical amps/pre-amps.
  • Think about your box early so you can make sure all the pieces fit together at the end. You'll tear your hair out over shoving it all together at the end if you don't. This especially true if you want good acoustics (though it's all going to be a compromise for portability).
  • As a corollary to the above, think about your controls/connections layout for ease of use, accessibility (and appearance). It sucks to have a great project with hard to access controls or inputs.
  • Do some usage scenarios. Think through how you would use your "product" when completed. Write them down. Maybe even ask a friend or two what they think. That will guide you in a number of areas. I often get my wife to listen to my ideas and she really helps me understand the good (and just plain wrong) of the design. You'll need to check your ego at the door!
  • Document with pictures, schematics, a journal, design docs, ... This is helpful for things like decisions you make about the design. If you need to revisit, you don't have to reconstruct your rational if it's written down. If you do V2 of the project (like adding bluetooth), it's invaluable. and, there are delays built into the process - PCB fab time is a week or more. I find it easier to get back up to speed when the PCB comes back if I've got good documentation.
Thanks for your fast response and helpful post.

Yeah i forgot to mention it should be chargable with a standard 5v phone charger. I will be using it as a portable mini speaker for work. It must be battery powered!
PCB construction, the part placement and soldering won't be an issue as i work as a microtechnologist.. its what i do every day ;)

keeping in mind the loss of voltage due to the batterie's nature should i also leave some voltage space for the OP-AMP? For example use 4x 3.7V LiPos, but only use a signal amplification that brings me up to roughly 10V(almost dry) and up to ~16V(full).. that would mean even if the batterypack is almost dry it would still make sure the sound doesnt get distorted.

I think your idea of starting off with V1 and upgrade it to a V2 after i gained a little more knowledge and fixed the process is great.. so i probably just start off with a standard Mono Full Range driver, so the whole low/highpass filter stuff is out
 

MrChips

Joined Oct 2, 2009
23,847
As I see it, you have two critical decisions you have to make before you get started.

1) What type of battery do you plan to use?
2) What class amplifier do you want to design, Class-B push-pull or Class-D PWM (pulse-width modulation).

For maximum efficiency, Class-D is the way to go.
For choice of battery, you need to do your due diligence and do the math.

For first iteration, let us assume 100% efficiency for the moment.
Assume you want 25W RMS per channel = 50W RMS.
Calculate the battery requirement in Ah for running at 50W for 5 hours.

Compare that with the energy available in a 12V SLAB (sealed lead acid battery) versus that in a popular 18650 Li-ion battery.
Determine the size of the battery in Ah.
 

philba

Joined Aug 17, 2017
960
I really think you will be much happier if you run off of a single lipo and use a boost converter to get to 12V. Then there is no worry about changing voltage rails as the battery discharges. You will absolutely want a battery manager (fuel gauge, undervoltage protect) and to charge 4 cells you'll need a fairly complex charger scheme.I'm not familiar with any charger chips that do 4 cells, you may wind up having to implement 4 chargers (I would not charge 4 in series). Once you've committed to the charger and manager, it's a very small step to add a boost converter. Take a look at the various products sparkfun has. They give you the schematics.
 

Thread Starter

sHanQ

Joined Oct 2, 2017
7
Thanks, You guys helped me out a lot! I guess i have to start off by solving my issues with the power supply(voltage drop). Once i'm done with that i can go forward to the amplifier..

I will look into the boost converter you mentioned philba :)
 

GopherT

Joined Nov 23, 2012
8,012
I agree that there are packages that solve all the OP's problems. I personally would just find a decent board but he wants to design it himself and I celebrate that spirit! Also, the commercial designs often make compromises just to hit a particular price point. This is amazingly true with portable audio. Certain companies often put out crap and coast on their name with the cheapo Chinese knockoffs being just as good, er, bad. It works fine for them since most customer don't really care about acoustic quality. (B**e springs to mind).
It actually works for them for a lot of reasons. Name any other US-based consumer electronics company that has been around longer (and is still around) and not focusing on computers.
 

philba

Joined Aug 17, 2017
960
If it didn't work, they wouldn't keep doing it. But my point wasn't that they are doomed but rather using them as a standard doesn't get you a high quality solution. With a little effort, you can build your own speakers that produce much better sound quality than them.
 

GopherT

Joined Nov 23, 2012
8,012
If it didn't work, they wouldn't keep doing it. But my point wasn't that they are doomed but rather using them as a standard doesn't get you a high quality solution. With a little effort, you can build your own speakers that produce much better sound quality than them.
I still doubt a newbie can pull off anything with the volume, audio quality and battery life that will satisfy his nebulous target. Additionally, it will take longer, cost more and require significant support from others (you) to achieve is disappointing result.
 

philba

Joined Aug 17, 2017
960
Other than that, Mrs Lincoln, how was the play?

I think it's great he wants to do this. Sure, it may not turn out well but I think it great he's trying. I'm always building things that I could probably buy better and cheaper off ebay. But there is a huge amount of satisfaction in creating and building your own design. It's not for everyone. Heck, if I could make my own semiconductors, I'd give it a shot. There's a lot of fun in the doing.
 

GopherT

Joined Nov 23, 2012
8,012
Other than that, Mrs Lincoln, how was the play?

I think it's great he wants to do this. Sure, it may not turn out well but I think it great he's trying. I'm always building things that I could probably buy better and cheaper off ebay. But there is a huge amount of satisfaction in creating and building your own design. It's not for everyone. Heck, if I could make my own semiconductors, I'd give it a shot. There's a lot of fun in the doing.
Now you’re singing the right tune. Your earlier claim that the OP can achieve better results than mid-range commercial products made me laugh. Thanks for returning to reality.
 

philba

Joined Aug 17, 2017
960
No, I don't agree with that and am not sure why you are being argumentative about it. If you look at most minispeakers out there, they are all really cheap using crappy speakers. Just selecting a good speaker will up the quality level. Speaker cabinet design is covered by numerous web sites out there and it's fairly easy for even a total newbie to build a decent one. Won't be cheaper, will take time, has potential for error and failure. But I never said it wouldn't turn out well. It could turn out great. Also, I never said mid-range. I sure don't understand the need to deny the OP any hope.
 

MrChips

Joined Oct 2, 2009
23,847
Ok, knock it off fellas.

This is a great project for anyone who wants to learn more about the basics of audio electronics and power amplifiers.
The TS has to do his due diligence and learn about battery capacity and power output.

Here is an example.
Suppose the output is 4V RMS into a 4Ω speaker. Current demand is 1A. Power output is 4W.
Battery requirement to run for 5 hours is 5Ah.

At the very least, TS will have to learn to reduce his expectations and make compromises.
 

GopherT

Joined Nov 23, 2012
8,012
No, I don't agree with that and am not sure why you are being argumentative about it. If you look at most minispeakers out there, they are all really cheap using crappy speakers. Just selecting a good speaker will up the quality level. Speaker cabinet design is covered by numerous web sites out there and it's fairly easy for even a total newbie to build a decent one. Won't be cheaper, will take time, has potential for error and failure. But I never said it wouldn't turn out well. It could turn out great. Also, I never said mid-range. I sure don't understand the need to deny the OP any hope.
I’ll sit back and watch. My money is on an incomplete project in a shoe box in the bottom of a closet within a month.
 

GopherT

Joined Nov 23, 2012
8,012
sorry for taking the discussion off topic. I'm maybe a little too excited to see someone stretching to learn new things by doing.
I agree. You are too excited. On the other hand, I do not apologize for discussing the probability of success. I just get tired of newbies getting pushed beyond their skill level for a first project and the corresponding failure and the newbie dropping their interest in Electronics.

Management is a skill, not just a job title. Know your people and their capabilities.
 

Audioguru

Joined Dec 20, 2007
11,249
You mentioned low current opamps but never said anything about high current power amplifiers. Your schematic showed a low current opamp trying and failing to drive a high current 4 ohm speaker. Why did you say "driver"? Is it the missing power amplifier or is it the speaker?
You should use bridged class-D power amplifiers. A bridged amplifier uses 2 amplifiers, then both wires of the speaker is driven with anti-phase producing almost double the voltage swing and almost 4 times the output power. Class-D is efficient.
A small speaker does not produce low frequencies unless it is near your ears (headphones) and/or has equalization to boost the bass.
A low power supply voltage for the power amplifier produces low output power.

My old clock radio has my power amplifier designed with plenty of bass boost for its 3" diameter pretty good replacement speaker.
My new clock radio has a speaker in a properly designed enclosure with a 6.5" woofer and a 1" dome tweeter.
 

AnalogKid

Joined Aug 1, 2013
9,283
First, there is no explicit need for bipolar power sources. National Semiconductor was making high quality preamp and medium quality power amp parts back in the 70's, all designed from the ground up to run on a single power supply. Two lipos in series would work just fine.

Note - no small signal opamp such as the OPA2134 can drive a speaker directly. You need separate preamp/control and power amp stages.

For better efficiency, the power amp section can be a pair of BTL (bridge-tied load) linear amps per speaker. In round numbers, 2 times the parts for 4 times the output power for the same battery voltage.

For the best efficiency, the power section should be class D. Not as much design fun, but better output quality and power.

ak
 
Top