a search would eventually turn up this link:
Switchmode Power Supply For Car Audio
some useful info but not quite a complete design ....
mlloyd1
Switchmode Power Supply For Car Audio
some useful info but not quite a complete design ....
mlloyd1
I can't read the schematic, it's blurry with small text. But look like a couple zeniers down at the bottn used as a voltage reference. Change their value as required.
Here is the schematic in pdf format.
what about changing secondary winding to change voltage to 32-0-32v ...?
if i have changed what will happen to output current..?
is there any modification need for control section..?
anybody pls HELP...
The each secondary has 9 turns. Needs to be reduced to 8 or 7 turns. The zeners in the opto circuit have to be reduced to ??. YGM. But, I've seen this PS schematic before in power supply forum. Perhaps the designer can help you out. Other than that, it may work.
Personally, I'd buy Meanwell SMPS off Ebay.
connexelectronic.com has a 1KW in a kit form and a 350-450W smps for 50 USD. See if they might do the job as well. Yes, you can save money by buying a blank PCB and chasing all the parts, winding the transformer(s) but it takes a bunch of time and work.
Good luck.
BTW, what's wrong with a 50/60Hz transformer supply??
Personally, I'd buy Meanwell SMPS off Ebay.
connexelectronic.com has a 1KW in a kit form and a 350-450W smps for 50 USD. See if they might do the job as well. Yes, you can save money by buying a blank PCB and chasing all the parts, winding the transformer(s) but it takes a bunch of time and work.
Good luck.
BTW, what's wrong with a 50/60Hz transformer supply??
I can understand why you wanna use a switch-mode pwr-supply if u have "space" issue or you're to cheap to by a linear solution. I've build a surround system with LM3886, just 'cuz they are well-designed and very few external parts is needed. So the Amp becomes very compact, I've also use the same IC in active speakers. Using SMPS with audio is not recommended, though you will get ripples how well designed SMPS you'll use. Those who builds class-D amp's, knows that converting the pulses into clean sinus is almost impossible, SMPS are o the march, especially in hi-watt PA-systems, were the sound quality isn't the deciding factor, you want people to hear you, and if it sounds good is just a bonus 
. It's not like leaning back in your favorite chair or sofa and put on Gary Moore, on vinyl...
So why you want to "distort" a good audio-ic with a smps is beyond me... LM3886 is a B-Amp, so you have a quite high efficiency - 85% - 90% - so you wanna use a transformer listed for 150VA, and 25A rectifier and two 10mF(or one 22mF)/channel, and maybe two three 0R15/10W resistors in parallel, in +V and -V, to further suppress eventual ripples, like;
----C(10mF) --- R (0R075)----C(10mF)---> + 35V(-35V)
as simple as they come... But if you up for a challenge and loooong nights, go ahead w/ your SMPS(U).... I'd never use SMPSU in a audio application, RF, pulses, digital designs etc. then they make life easier, but try to build a 20W rms A-amp with a switch-mode supply..... I, like my 'guru' John L Hood(who sadly died a few years ago) am very fond of the class-A amplifier, you have none of the drawback that B, D, T etc has, the major drawback though is the low efficiency, only 20 - 25%, so you'll need 100W of power to get 20W from the amplifier, but you don't have to consider the phase, and you can achieve a well-sounding amp with very few components.... but opinions are like 'behinds'; everybody has one....
. It's not like leaning back in your favorite chair or sofa and put on Gary Moore, on vinyl... So why you want to "distort" a good audio-ic with a smps is beyond me... LM3886 is a B-Amp, so you have a quite high efficiency - 85% - 90% - so you wanna use a transformer listed for 150VA, and 25A rectifier and two 10mF(or one 22mF)/channel, and maybe two three 0R15/10W resistors in parallel, in +V and -V, to further suppress eventual ripples, like;
----C(10mF) --- R (0R075)----C(10mF)---> + 35V(-35V)
as simple as they come... But if you up for a challenge and loooong nights, go ahead w/ your SMPS(U).... I'd never use SMPSU in a audio application, RF, pulses, digital designs etc. then they make life easier, but try to build a 20W rms A-amp with a switch-mode supply..... I, like my 'guru' John L Hood(who sadly died a few years ago) am very fond of the class-A amplifier, you have none of the drawback that B, D, T etc has, the major drawback though is the low efficiency, only 20 - 25%, so you'll need 100W of power to get 20W from the amplifier, but you don't have to consider the phase, and you can achieve a well-sounding amp with very few components.... but opinions are like 'behinds'; everybody has one....
Actually, using SMPS with audio can be great. There are threads here, about it, saying it is better than a linear supply. And if you think about it, it is much much easier to filter out a few hundred kilohertz, which is not in the audio band, than it is to filter out 100 or 120 Hz, which are in the audio band.
With an SMPS, a simple one-stage LC low-pass post-filter can get rid of basically all of the high-frequency stuff. And if you want perfection, you could add a three-terminal post-regulator circuit.
Regarding finding an SMPS circuit, there are zillions of them on the internet. But you should probably check at national.com and linear.com. They both have fully-automatic SMPS design tools, for free. National's is an "on line" type, aka Webench.
With an SMPS, a simple one-stage LC low-pass post-filter can get rid of basically all of the high-frequency stuff. And if you want perfection, you could add a three-terminal post-regulator circuit.
Regarding finding an SMPS circuit, there are zillions of them on the internet. But you should probably check at national.com and linear.com. They both have fully-automatic SMPS design tools, for free. National's is an "on line" type, aka Webench.
Last edited:
are you sure you know what you need?
A ClassAB amplifier running on +-35Vdc supply connected to 8ohm speakers can be asked to supply 11Apk on fast transients.
Where will this transient current come from?
Capacitors or the SMPS or shared between both?
If you want to drive 6ohm speakers the potential current demand will be even higher.
I still don't agree w/ good audio+smps... it's a none workable solution. If you have a B-stage and want 250W rms?(~700Wpp), like the former reply, were do you get the current....!
I'm very "old-school" when I design and build audio amps, even though I sometimes design a class-D(mostly for battery op. amps, you have nearly 100% efficiency, vs. class-A with about 20 - 25% - from the PSU, for 10W rms you'll need a 50W PSU - so therefore I use class-D in small active-speaker systems, for your MP3-player etc. when the sound is "impaired" to begin with).
I design my pwr supplies as capacitance multiplier, as the class - A amp will draw max current with no input signal. So can you get 1F(!)/ch. with with "small" capacitors - 4700µF+470µ/line, with a voltage drop of 7 volts. This can be used with a AB-amp as well, but you'll need bigger cap's after the "multiplier".
I say it again, for PA and guitar amp's and likewise, you can feed them with a SMPS, but you need very fast and big cap's to hold the current needed for high-peak outputs. And they are not cheap! It's probably much cheaper(and you'll get better results) to buy a 500VA transformer, some 10 - 33mF, 100VDC capacitors, and if you real want a regulated power supply, go old-school !!! And build a linear PSU, with BJT's(or FET's). In my opinion SMPS are good for TTL, MCU, PIC, digital applications, and has no place in audio-amp's, purhaps if you do as I, use an ATmega for monitoring- a watch dog - or sound-effects, some use full info, like how "strong" is the input signal, temp., output current, Z, SIM's, fan ctrl etc.
So if you're bent on using a SMPS, you'll not only need a good filter, you're gonna spend hours, wounding your own coils, cuz the value you need almost never exists, and if it does, it can't handle the current. You wanna use military quality capacitors - fast, low leakage, high values - and you still need a big transformer, where else are you going to get the current to the switching FET's :?
You can use the mains power - 230/115AC - and and do a simple 1:1 ratio, transformer and feed your FET's with. And a sugar-cube-transformer to take down the 280VDC(rectified mains, of 230AC, squrt2 times 230 minus 1,5V for the full wave rectifier) to +/-35V=
If you are interested in how SMPS works, and want to "diddle" around with 'em, do a variable lab-SMPS or some other application...
The only thing I can see, by using SMPS in an audio-amp is to make it lighter. If you have 50% volume and it jumps up to 90% for a ms. for example you are going to need a very very very very well designed SMPS with parts that costs many times more than a linear supply would.
And in Public Access applications and instrument amp's they can be useful, when the sound quality isn't the goal.... but who am I to tell you what is and what is not, you do as you please, if you're happy then I'm happy!
Good luck, you're gonna need it
I'm very "old-school" when I design and build audio amps, even though I sometimes design a class-D(mostly for battery op. amps, you have nearly 100% efficiency, vs. class-A with about 20 - 25% - from the PSU, for 10W rms you'll need a 50W PSU - so therefore I use class-D in small active-speaker systems, for your MP3-player etc. when the sound is "impaired" to begin with).
I design my pwr supplies as capacitance multiplier, as the class - A amp will draw max current with no input signal. So can you get 1F(!)/ch. with with "small" capacitors - 4700µF+470µ/line, with a voltage drop of 7 volts. This can be used with a AB-amp as well, but you'll need bigger cap's after the "multiplier".
I say it again, for PA and guitar amp's and likewise, you can feed them with a SMPS, but you need very fast and big cap's to hold the current needed for high-peak outputs. And they are not cheap! It's probably much cheaper(and you'll get better results) to buy a 500VA transformer, some 10 - 33mF, 100VDC capacitors, and if you real want a regulated power supply, go old-school
!!! And build a linear PSU, with BJT's(or FET's). In my opinion SMPS are good for TTL, MCU, PIC, digital applications, and has no place in audio-amp's, purhaps if you do as I, use an ATmega for monitoring- a watch dog - or sound-effects, some use full info, like how "strong" is the input signal, temp., output current, Z, SIM's, fan ctrl etc.So if you're bent on using a SMPS, you'll not only need a good filter, you're gonna spend hours, wounding your own coils, cuz the value you need almost never exists, and if it does, it can't handle the current. You wanna use military quality capacitors - fast, low leakage, high values - and you still need a big transformer, where else are you going to get the current to the switching FET's :?
You can use the mains power - 230/115AC - and and do a simple 1:1 ratio, transformer and feed your FET's with. And a sugar-cube-transformer to take down the 280VDC(rectified mains, of 230AC, squrt2 times 230 minus 1,5V for the full wave rectifier) to +/-35V=
If you are interested in how SMPS works, and want to "diddle" around with 'em, do a variable lab-SMPS or some other application...
The only thing I can see, by using SMPS in an audio-amp is to make it lighter. If you have 50% volume and it jumps up to 90% for a ms. for example you are going to need a very very very very well designed SMPS with parts that costs many times more than a linear supply would.
And in Public Access applications and instrument amp's they can be useful, when the sound quality isn't the goal.... but who am I to tell you what is and what is not, you do as you please, if you're happy then I'm happy!
Good luck, you're gonna need it
As I wrote in my previous post,I booth build and design class-D amps, and hi-power PA-systems, and I also use SMPS in my designs, especially in Amps intended for public access. To use an A or AB amp today in PA-systems and music-instrument amp's is waste of energy, since everybody thinks man is destroying earth and so on...
But what I'm trying to say is, for a person who enjoys listening to music, you can't match the sound quality from a completely analogue tube amp.
And of course there's no rule saying that is wrong to you SMPS, but as a hobbyist, that don't work with audio amplifiers and power supply, it is very easy to go wrong with a SMPS, with a linear supply, you can't fail, and suppress ripples from a non-regulated can everybody build, just use bigger cap's....
And if you have the money, the know how, the patience, then go ahead and build a class-D w/ SMPS !
But you will never match the sound-quality w/ a D-amp from an A-amp, how well designed your filters are(to convert the pulses to sinus/audio fq) and your SMPS!
The world is analogue, all in-/output is analogue.... MCUs and "digital-tech" is good for monitoring and sample signals, so you can play around with them, especial for music-effects and amp's! And a bunch of non-audio related applications!
But what I'm trying to say is, for a person who enjoys listening to music, you can't match the sound quality from a completely analogue tube amp.
And of course there's no rule saying that is wrong to you SMPS, but as a hobbyist, that don't work with audio amplifiers and power supply, it is very easy to go wrong with a SMPS, with a linear supply, you can't fail, and suppress ripples from a non-regulated can everybody build, just use bigger cap's....
And if you have the money, the know how, the patience, then go ahead and build a class-D w/ SMPS !
But you will never match the sound-quality w/ a D-amp from an A-amp, how well designed your filters are(to convert the pulses to sinus/audio fq) and your SMPS!
The world is analogue, all in-/output is analogue.... MCUs and "digital-tech" is good for monitoring and sample signals, so you can play around with them, especial for music-effects and amp's! And a bunch of non-audio related applications!
Pardon my bad spelling, I'll try to type slower, so it don't make so many mistakes..
It should say, "no rule saying that it is wrong to use SMPS", and if you enjoy your class-D designs with SMPS, then use it!! I have my believes, and you yours! It all comes down to the speakers at the end anyway..... but I've listened to some very expensive D-amp's, rated for 300W rms, and they sounds like some one's cat being tortured, well maybe not that bad, but a pocket-amp for your MP3-player sound better... long live the LP's! Analogue all the way....
It should say, "no rule saying that it is wrong to use SMPS", and if you enjoy your class-D designs with SMPS, then use it!! I have my believes, and you yours! It all comes down to the speakers at the end anyway..... but I've listened to some very expensive D-amp's, rated for 300W rms, and they sounds like some one's cat being tortured, well maybe not that bad, but a pocket-amp for your MP3-player sound better... long live the LP's! Analogue all the way....
Of course you can change your transformer, just don't use one that have to big difference. Power handling, output voltage, and remember that Ac vs. DC are not equal to each other! A AC voltage of 12V, will become about 15 - 16 VDC, AC voltage times 1,414 gives you a pointer to what you'll expect, and this is w/o load,. So if you have 12 VAC@10A --> 120VA, then the 16VDC will drop to 12VDC if you take out 10 amps.... if you only use 2 - 3 amps the DC voltage will be higher then 12V=.... here's a quote from an Audio-design guidebook, High Performance Audio Power Amplifiers, B.Duncan.
QUOTE;
4.9 Class summary
Type Key Features
A
Pros: Ease of low distortion, both static and dynamic
Cons: High hardware cost, high weight, high electricity cost, high heat output,
power capability practically limited to about 100 watts.
B, A-B, A-B1, A-B2, QPP.
Pros: Simple circuitry, low hardware cost, medium electricity cost. With good
design and application to ameliorative tools, almost as low a distortion as
class A, is possible.
Cons: The reference condition.
Super A, ð-mode, Dynamic and Plateau biased-A.
Pros: Potentially lowered dynamic distortion over ordinary B and A-B types. Much
of the class A advantages at some fraction of the cost.
Cons: Slightly more complex circuitry than A-B. Topologically more limited.
Higher electricity cost and heat output if quiescent periods dominate, but
still much lower than class A types.
C Radio Frequency amplifiers only, not for audio.
4.9 Class summary
Topologies, classes and modes
154
D
Pros: High efficiency, lightweight, potentially low hardware cost,
Cons: Complex circuitry, sonic performance may be poor, particularly above 5kHz,
EMI emission likely.
E, F: Category undocumented, not for audio.
G, H
Pros: Both permit high transient power capability. usefully higher efficiency than
class A-B under medium to hard-drive conditions.
Cons: Slightly more complex circuitry than A-B. More PSU parts and hardware
and higher build cost. Added dynamic distortions possible (but well containable
in practice) due to multiple-supply switching and finite part matching.
In inexpert designs, multiple power supplies might be improperly apportioned,
leading to wasted hardware and unsuitability with some
programme.
S See section 4.9.3. Not really a class.
per watt output: ⇐Lowest – non-linear scale– Highest⇒
Waste heat D G, H B, A-B SA A
Weight D G, H B, A-B SA A
Parts Cost D G, H B, A-B, SA A
Distortion A A/H, SA B, A-B G, H D
Notes: A-B includes NSB (non-switching class B).
‘SA ‘ (special A) includes Super class A and sliding A and ð-mode A-B.
The following table concludes, with a listing of power amplifiers established in
professional sound, that go beyond ordinary class B and A-B operation.
...hmmm
QUOTE;
4.9 Class summary
Type Key Features
A
Pros: Ease of low distortion, both static and dynamic
Cons: High hardware cost, high weight, high electricity cost, high heat output,
power capability practically limited to about 100 watts.
B, A-B, A-B1, A-B2, QPP.
Pros: Simple circuitry, low hardware cost, medium electricity cost. With good
design and application to ameliorative tools, almost as low a distortion as
class A, is possible.
Cons: The reference condition.
Super A, ð-mode, Dynamic and Plateau biased-A.
Pros: Potentially lowered dynamic distortion over ordinary B and A-B types. Much
of the class A advantages at some fraction of the cost.
Cons: Slightly more complex circuitry than A-B. Topologically more limited.
Higher electricity cost and heat output if quiescent periods dominate, but
still much lower than class A types.
C Radio Frequency amplifiers only, not for audio.
4.9 Class summary
Topologies, classes and modes
154
D
Pros: High efficiency, lightweight, potentially low hardware cost,
Cons: Complex circuitry, sonic performance may be poor, particularly above 5kHz,
EMI emission likely.
E, F: Category undocumented, not for audio.
G, H
Pros: Both permit high transient power capability. usefully higher efficiency than
class A-B under medium to hard-drive conditions.
Cons: Slightly more complex circuitry than A-B. More PSU parts and hardware
and higher build cost. Added dynamic distortions possible (but well containable
in practice) due to multiple-supply switching and finite part matching.
In inexpert designs, multiple power supplies might be improperly apportioned,
leading to wasted hardware and unsuitability with some
programme.
S See section 4.9.3. Not really a class.
per watt output: ⇐Lowest – non-linear scale– Highest⇒
Waste heat D G, H B, A-B SA A
Weight D G, H B, A-B SA A
Parts Cost D G, H B, A-B, SA A
Distortion A A/H, SA B, A-B G, H D
Notes: A-B includes NSB (non-switching class B).
‘SA ‘ (special A) includes Super class A and sliding A and ð-mode A-B.
The following table concludes, with a listing of power amplifiers established in
professional sound, that go beyond ordinary class B and A-B operation.
...hmmm
- Status
- This old topic is closed. If you want to reopen this topic, contact a moderator using the "Report Post" button.