Hi,
Been reading and reading about speakers and amps for the last 2 months and i only get more puzzled with all the possibilities. I don't have an electronics background but have been reading some online courses on the subject. I would like to get hands on now and i have set myself this goal: build an active mono speaker. In more detail i want to to the following:
build a speaker that consist of
I have read about chipamps and gainclones and the different mythical creations (brianGT, ...), different chipsets, power supplies, but it is all so much to choose from.
I have not found an exact project like the one i would like to build so i don't know where to start. I would like to use good quality components to get a killer sound without spending over hundreds of $ or €.
Anybody attempted such a project?
Got some good tutorials or reading on this subject?
If possible I would like to do it step by step with the help of you guys. I will post my progress on each fase and will be happy to read your comments and tips.
Thanx and let me know i you need some more info.
Been reading and reading about speakers and amps for the last 2 months and i only get more puzzled with all the possibilities. I don't have an electronics background but have been reading some online courses on the subject. I would like to get hands on now and i have set myself this goal: build an active mono speaker. In more detail i want to to the following:
build a speaker that consist of
- a woofer and tweater
- an amp of about 50 to 100W
- a passive crossover (if possible)
- a raspberry pi with a DAC as audio streamer (logitec media server & squeezelite)
- a power supply that is sufficient for the RPi and the amp
I have read about chipamps and gainclones and the different mythical creations (brianGT, ...), different chipsets, power supplies, but it is all so much to choose from.
I have not found an exact project like the one i would like to build so i don't know where to start. I would like to use good quality components to get a killer sound without spending over hundreds of $ or €.
Anybody attempted such a project?
Got some good tutorials or reading on this subject?
If possible I would like to do it step by step with the help of you guys. I will post my progress on each fase and will be happy to read your comments and tips.
Thanx and let me know i you need some more info.
Hello Cypress.
some thoughts: Amplifier power won't mean anything by itself, it's the combination of amplifier power and efficiency of speaker which give you a range of usable SPL (sound pressure level) at a given distance.
Maybe you should clarify the use you want to make of finished speaker (mono is unusual this days) it could be a lot easier for comments or link you to some existing design.
If using a Pi with DAC maybe you could skip passive cross over and go active biamp if dac as 4 outputs* ( one amp for each tweeter and woofer, filtering done by software).
For amplifier chipamps are 'easy' to tailor to your needs and regularly used in nearfield monitors.
*:four outputs if you want stereo, for mono only two output is needed obviously... 🙂
some thoughts: Amplifier power won't mean anything by itself, it's the combination of amplifier power and efficiency of speaker which give you a range of usable SPL (sound pressure level) at a given distance.
Maybe you should clarify the use you want to make of finished speaker (mono is unusual this days) it could be a lot easier for comments or link you to some existing design.
If using a Pi with DAC maybe you could skip passive cross over and go active biamp if dac as 4 outputs* ( one amp for each tweeter and woofer, filtering done by software).
For amplifier chipamps are 'easy' to tailor to your needs and regularly used in nearfield monitors.
*:four outputs if you want stereo, for mono only two output is needed obviously... 🙂
Last edited:
Hi Krivium.
The goal of my project is to get some experience in working with amps, pcb soldering (if required) and crossovers. Even some simulation software to determine the crossover frequency.
I choose mono since i will be building some kind of docking speaker. The RPi will enable wifi streaming anywhere in my home. I can control is using my phone.
In the end i would like to build one for most rooms and use it as a cheap Sonos. From what i have read you can build way better sounding systems then the commercially available ones (sony, bose, ...).
The DAC that i use is a hifiberry card that has only 2 outputs (RCA) but any recommendations on other/better DAC boards are also very welcome.
The goal of my project is to get some experience in working with amps, pcb soldering (if required) and crossovers. Even some simulation software to determine the crossover frequency.
I choose mono since i will be building some kind of docking speaker. The RPi will enable wifi streaming anywhere in my home. I can control is using my phone.
In the end i would like to build one for most rooms and use it as a cheap Sonos. From what i have read you can build way better sounding systems then the commercially available ones (sony, bose, ...).
The DAC that i use is a hifiberry card that has only 2 outputs (RCA) but any recommendations on other/better DAC boards are also very welcome.
Cypress, ok so you're looking for some kind of background music system.
May i ask why mono and not stereo?
Which kind of music do you usually listen to? Classical,jazz,techno,metal...?
May i ask why mono and not stereo?
Which kind of music do you usually listen to? Classical,jazz,techno,metal...?
Ok it won't probably help you about choice but here is idea about some kind of 'docking' speakers but in stereo:
http://www.diyaudio.com/forums/full...orations-matrixed-single-stereo-speakers.html
Note that this kind of speaker needs a specific placement in room (as you can see in pictures in linked thread). But as a background music system it could be a good solution in my opinion (given you're not trying to get party level and are not addicted to loud heavy bass). Maybe it won't satisfy what you have in mind as it's not really a classical MT (woofer/medium + tweeter design).
http://www.diyaudio.com/forums/full...orations-matrixed-single-stereo-speakers.html
Note that this kind of speaker needs a specific placement in room (as you can see in pictures in linked thread). But as a background music system it could be a good solution in my opinion (given you're not trying to get party level and are not addicted to loud heavy bass). Maybe it won't satisfy what you have in mind as it's not really a classical MT (woofer/medium + tweeter design).
i mainly play all kinds of rock music. I choose mono for simplicity. It just needs one amp and 1 crossover. or am i missing something. I just want to build a simple speaker that gets its input from a streamer and that i can put anywhere to play some music.
Am I wrong to say that if i have the basic composition of my system, i can always create improved versions of it by using bigger and better components?
My only problem is that with all the info i have read over the last few months, i have no idea on how to start building it. So many chips, so many crossover-configurations.
What is a good and simple composition for an active speaker?
Thanx
Am I wrong to say that if i have the basic composition of my system, i can always create improved versions of it by using bigger and better components?
My only problem is that with all the info i have read over the last few months, i have no idea on how to start building it. So many chips, so many crossover-configurations.
What is a good and simple composition for an active speaker?
Thanx
I've been working towards the goal of an active speaker (actually I call it a 'speaker activator') for a number of years now. The reason its a 'speaker activator' rather than an 'active speaker' is because I see fitting all the kit inside the speaker's cabinet as too much of a limitation. Also I want a piece of kit which I can make many of fairly cheaply and use it in conjunction with all the ripped-apart active speakers I've got lying around my apartment. I ripped many actives apart as a learning process to understand what sounds good and what doesn't - their drive units are intact but the electronics has been butchered to varying degrees.
First a bit of terminology - an 'active speaker' normally means one which has a line-level crossover (which is normally active, could in theory be passive) and separate amps for each drive unit. So in your case it'd have two amps inside. What you've described so far (single amp, passive XO) I'd call a 'powered speaker'.
My advice would be to go active with separate amps and an active XO. That's the easiest way to top-notch sound in my experience. I'd omit all the digital stuff which looks to me like biting off more than you can usefully chew at the outset and give your active speaker a balanced analog input.
First a bit of terminology - an 'active speaker' normally means one which has a line-level crossover (which is normally active, could in theory be passive) and separate amps for each drive unit. So in your case it'd have two amps inside. What you've described so far (single amp, passive XO) I'd call a 'powered speaker'.
My advice would be to go active with separate amps and an active XO. That's the easiest way to top-notch sound in my experience. I'd omit all the digital stuff which looks to me like biting off more than you can usefully chew at the outset and give your active speaker a balanced analog input.
Thanx. I have indeed wondered why some call it a powered speaker and others an active speaker. In the end i would like to build one of both. Just for practice 😉
I have heard that active XOs are very expensive. That is the main reason i would first start with the passive XO that sits after the amp.
Any tips on affordable active XO solutions are welcome.
Got any experience with DSP?
I have heard that active XOs are very expensive. That is the main reason i would first start with the passive XO that sits after the amp.
Any tips on affordable active XO solutions are welcome.
Got any experience with DSP?
Last edited:
Funny you should think active XOs are expensive. I've just finished building a couple where I reckon the BOM cost is under $6. Mind you that is buying parts in China which is probably the cheapest place on earth for electronics parts. Quite probably cheaper than a couple of chokes and foil caps you might want to use for a passive XO. Of course active costs you more in amp channels but you get away with lower power for each than when you have a passive XO. So I reckon overall for the same SPL active might work out cheaper when done the way I've just done mine.
Yes I do have a small amount of experience with DSP - so far I haven't found a killer app for DSP in an active speaker but I'm still looking 🙂
Yes I do have a small amount of experience with DSP - so far I haven't found a killer app for DSP in an active speaker but I'm still looking 🙂
do you buy kits or you pick your components yourself? And if you say that you have built one for under 6$, than you can build a killer XO for 30$? Or does that math not apply 🙂 ?
if heard some good things about miniDSP. Have you worked with that? I might get me one for this project.
if heard some good things about miniDSP. Have you worked with that? I might get me one for this project.
I fully agree with abraxalito comments about line level crossover vs passive one.
Do you have a 'minimum' background in electronic Cypress (by minimum i mean able to read schematic and basics operation of opamp circuits)?
If Yes i could share some schematics of common nearfield studio monitors to give you inspiration if needed of analog crossover/driver behavior correction/amplifier .
About dsp, i'm really into this. I've used some dedicated hardware and finally find my 'dream machine' at an 'affordable' price (which is really expensive in the end as you stated).
Most entry level (Behringer, Dbx,...) are ok for most 'home' purpose, but as stated by abraxalito, nearly anything they do can be built line level analog (the exception is for compensation delay).
Some software solutions are really great but all in all are not cheap once you take into account computer and soundcards needed (i'm thinking about Accourate).
I would say that if you're not looking for really specific needs (FIR) neither dsp software nor high end hardware are needed and you could go with analog crossover (especially if your source output is analog, if it is digital a dedicated hardware is worth the money).
You are wrong and right at the same time: once you choose drivers you could always go higher in range afterwards but it is not as easy as changing drivers on box: you have to take into account individual behavior of each drivers and 'compensate' for them but it can be real problematic for woofers (you'll have to build new box at extreme case if different alignement need to be used to take the most adavntage of new better driver by example...).
For what i've heard, more than the absolute quality of driver it's more a game to find combo which works together and are in same 'quality' range: a system which is coherent.
That's a reason why you find many projects using 'same' drivers...
If your are listening to rock music you are not looking for some real loud and deep bass so you can refine your choice about woofer diameter. I would say that up to 8" is ok. Now you have to find a tweeter which is able to reach down low enough to be crossed with the woofer.
Or... go with a coaxial! Many advantages (from my point of view). Or... whith a small full range driver in a Fast (acronym for Fullrange Assisted by Subwoofer Technic ... or something like that). Same as with a tweeter except you cross down low (150/300hz) to full range... Many options as you said! 🙂
Do you have a 'minimum' background in electronic Cypress (by minimum i mean able to read schematic and basics operation of opamp circuits)?
If Yes i could share some schematics of common nearfield studio monitors to give you inspiration if needed of analog crossover/driver behavior correction/amplifier .
About dsp, i'm really into this. I've used some dedicated hardware and finally find my 'dream machine' at an 'affordable' price (which is really expensive in the end as you stated).
Most entry level (Behringer, Dbx,...) are ok for most 'home' purpose, but as stated by abraxalito, nearly anything they do can be built line level analog (the exception is for compensation delay).
Some software solutions are really great but all in all are not cheap once you take into account computer and soundcards needed (i'm thinking about Accourate).
I would say that if you're not looking for really specific needs (FIR) neither dsp software nor high end hardware are needed and you could go with analog crossover (especially if your source output is analog, if it is digital a dedicated hardware is worth the money).
You are wrong and right at the same time: once you choose drivers you could always go higher in range afterwards but it is not as easy as changing drivers on box: you have to take into account individual behavior of each drivers and 'compensate' for them but it can be real problematic for woofers (you'll have to build new box at extreme case if different alignement need to be used to take the most adavntage of new better driver by example...).
For what i've heard, more than the absolute quality of driver it's more a game to find combo which works together and are in same 'quality' range: a system which is coherent.
That's a reason why you find many projects using 'same' drivers...
If your are listening to rock music you are not looking for some real loud and deep bass so you can refine your choice about woofer diameter. I would say that up to 8" is ok. Now you have to find a tweeter which is able to reach down low enough to be crossed with the woofer.
Or... go with a coaxial! Many advantages (from my point of view). Or... whith a small full range driver in a Fast (acronym for Fullrange Assisted by Subwoofer Technic ... or something like that). Same as with a tweeter except you cross down low (150/300hz) to full range... Many options as you said! 🙂
I'm going to toss in another possible direction. rPi has some issues with USB connections, you may want to look at some other possibilities. I found an old PC laptop that nobody was using (free). Loaded Ubuntu Linux (free), MPD, Ecasound (all free), and purchased a cheap 7.1 USB sound card. Crossover and DSP are all in the laptop. Inexpensive amp cards and kits are quite available, browse the "solid state" forum for tips to which ones are good...
Oh, yeah, my old laptop? Pentium III. That's old.
Oh, yeah, my old laptop? Pentium III. That's old.
I'd suggest a minidsp 2x4. Just one will do. The flexibility it has will allow you to design a crossover that can be the basis of something simpler, like a line level passive if need be. This will cut costs in the long run and allow for much experimentation eg learning curve without additional cost.
Also invest in a calibrated mic. Can be had for under $100 and is absolutely essential. REW measurement software is free.
Also invest in a calibrated mic. Can be had for under $100 and is absolutely essential. REW measurement software is free.
I buy the bits and wire them up on double-sided plated through perfboard.
Been listening to the $6 XOs and haven't noticed anything that I'd like to improve so far. But I should really build one with much more expensive opamps and use it as a reference. I'm thinking maybe OPA551s because they'll run at insanely high supplies (60V) - meaning dynamics could improve. To my ears, at present the sub $6 XO is already killer.
Seems to me its most probably limited by its on-board DACs. I've not got one, this is just based on what I've read about the chip from the DS and other critical listeners' evaluations of how it sounds. Its certainly very cheap for the flexibility it offers.
I've several minidsp products, they work very well regardless of what some claim. For their flexibility alone its worth alot more. Today when we talk about consumer grade electronics we're not much past what was available three decades ago "quality wise". Have an old Onkyo TX-DS787(b) a friend had me check out after sitting in his garage for two decades. A lowly minidsp 2x4 still has a better s/n and distortion specs. If we are talking the ultimate speakers or a high effeciency design I'd be looking for better source.
My nearly 4 decades of designing audio (everything) tells me I can do better and that experience asks at what cost?... For the money you can't make a dsp based hardware better. Analogue active is fine if you do not require eq or time delay and are fine with a given filter slope and type. Personally prefer to know the what ifs than to decide good enough because the hardware is simply not flexible enough to really let it shine, or tweak a design compromise on the fly, in real time.
Minidsp's usb streamer combined with a minisharc and the new addon 8channel dac would be a very attractive solution. If you got the urge to burn fingers rolling your own dac's etc is certainly possible. 😉
I have a feeling you'll want to dive into acoustic theory and what makes a good speaker before all of this ancillary electronics bumbo jumbo. 🙂
My nearly 4 decades of designing audio (everything) tells me I can do better and that experience asks at what cost?... For the money you can't make a dsp based hardware better. Analogue active is fine if you do not require eq or time delay and are fine with a given filter slope and type. Personally prefer to know the what ifs than to decide good enough because the hardware is simply not flexible enough to really let it shine, or tweak a design compromise on the fly, in real time.
Minidsp's usb streamer combined with a minisharc and the new addon 8channel dac would be a very attractive solution. If you got the urge to burn fingers rolling your own dac's etc is certainly possible. 😉
I have a feeling you'll want to dive into acoustic theory and what makes a good speaker before all of this ancillary electronics bumbo jumbo. 🙂
A lot of great info, thanx!
My background is 'Electronics all in one for dummies' 🙂 I can read a scheme and i know what all the components do. it ends there 🙂
this results in a lot of "Why's" when i read about chipamp configurations.
I'm thinking i should start out with a mini dsp. I assume it can be used as a line level XO?
Next i will need two chipamps. I have found a lot of those on ebay at around 20$ but they all have a THD of around 10%!!! My research tells me to steer clear of those. So i'm willing to build one form a kit or some proven schematics i can find online.
I recently came across this post: Op Amp Shootout | Cycfi Research
It compares 10s of opamps and rates them on various criteria. If i choose one of those highly rated ones, can i fit it in an existing scheme? i will probably need to change all the capacitors and resistor values?
Another question; do i have to know what drivers i will pick before i choose my amps? i know the amps should be powerfull enough for the speakers.
My background is 'Electronics all in one for dummies' 🙂 I can read a scheme and i know what all the components do. it ends there 🙂
this results in a lot of "Why's" when i read about chipamp configurations.
I'm thinking i should start out with a mini dsp. I assume it can be used as a line level XO?
Next i will need two chipamps. I have found a lot of those on ebay at around 20$ but they all have a THD of around 10%!!! My research tells me to steer clear of those. So i'm willing to build one form a kit or some proven schematics i can find online.
I recently came across this post: Op Amp Shootout | Cycfi Research
It compares 10s of opamps and rates them on various criteria. If i choose one of those highly rated ones, can i fit it in an existing scheme? i will probably need to change all the capacitors and resistor values?
Do you have the schematics of those? or can i buy them as a kit?abraxalito said:
Why a mic? to analyse different XO frequencies?greebster said:
Another question; do i have to know what drivers i will pick before i choose my amps? i know the amps should be powerfull enough for the speakers.
The reason for actual measurements is that the drivers you have in hand (or not) quite likely have specs different to those published by the manufacture and once mounted in an enclosure will change the response (think baffle step etc). Having a jig to measure T/S parameters with REW is a must. There are instructions on how to make this simple jig in the REW documentation. Once you have your driver mounted and measured the response can be imported into PCD or similar and proper corrections can be made. Without is a lost cause. You (your ears) cannot define amplitude peaks, nulls, phase or frequency.
As far as amps go use one that is rated for the resultant speaker impedence. EG if 4ohm speaker for amp rated into such loads. I use pairs of drivers which results in a nominal 4ohm impedence, the amps on the otherhand are rated for 2ohms.
As far as amps go use one that is rated for the resultant speaker impedence. EG if 4ohm speaker for amp rated into such loads. I use pairs of drivers which results in a nominal 4ohm impedence, the amps on the otherhand are rated for 2ohms.
Last edited:
Wise choice imho. It is in fact the same as a line level XO except all processing is done by software. It's great because you end up with this configuration:
D/A converter-> amplifier->driver. Nothing more than cables in signal path. Make a huge improvement versus passive design in my point of view. If you decide to go analog you insert a crossover between converter and amplifier. Nothing serious if circuit is well designed but can be a potential source of signal degradation. Nice to skip it.
Last edited:
i've been looking at several DSP solutions and i have my eye on the Hypex AS2.100 230Vac. Does anyoune have any experience with that. Its much more expensive than i initially planned but it has power supply, amps and dsp in one device.
I was also thinking about maybe a miniDSP 2x4. Its way cheaper but i need to take care of power and amps.
And where to buy these parts from in europe? anything above 50€ from outside the european union is charged extra tax by customs.
I was also thinking about maybe a miniDSP 2x4. Its way cheaper but i need to take care of power and amps.
And where to buy these parts from in europe? anything above 50€ from outside the european union is charged extra tax by customs.
Have heard good things from those whom own Hypex amps. Advantage they have over the 2x4 other than being an all inclusive solution that it is independantly network addressable. The 2x4 cannot do this via usb. A problem my system has, to me it's a minor inconvienance.
I modified a pair of amps, each has 4/ 100 watt channels. They were cheap enough, the power side is about twice the cost and that'll make you make you wonder if this is worth DYIing. 🙂
I modified a pair of amps, each has 4/ 100 watt channels. They were cheap enough, the power side is about twice the cost and that'll make you make you wonder if this is worth DYIing. 🙂
- Status
- Not open for further replies.