for example: www.qrp4u.de/docs/en/smps_new/ ,
i made my own smps for tda7294 with ir2153 + pair irf740 + ferrite toroid 45x28x12 mm without regulation , output +- 35v without load.
in google look for sg3525 , tl494, sg3825.
IR and other IC and mosfet manufacturers have a lot of application notes on smps.
Good luck.
i made my own smps for tda7294 with ir2153 + pair irf740 + ferrite toroid 45x28x12 mm without regulation , output +- 35v without load.
in google look for sg3525 , tl494, sg3825.
IR and other IC and mosfet manufacturers have a lot of application notes on smps.
Good luck.
a good friend of mine has ingeniously worked out how to derive +/-32 volts from a standard PC powersupply
benefits of this are: PC supplies are cheap or free, can provide a couple of hundred watts, and all the SMPS circuitry is already done, it just need a little tweaking
i'll provide the link to my site detailing how as soon as he finishes the writeup
benefits of this are: PC supplies are cheap or free, can provide a couple of hundred watts, and all the SMPS circuitry is already done, it just need a little tweaking
i'll provide the link to my site detailing how as soon as he finishes the writeup
i should add, the SMPS sounds brilliant.
infact, i gave him an LM4766 stereo 40W chip and he assembled the whole amp on a 1.5inch sqaure prototype board and then fit the entire thing inside the PSU. granted it has a very small heatink (maybe 2x3 inches by 1/10 thick) however thats offset by the fact it has a fan - the sound of which is easily overcome by the power of the amp
but i should also warn you, modifying a PC PSU is not entirely simple, and definitely not for beginners. it requires a lot of thought, some planning and the removal of quite a number of parts (ALL the caps on the old 3.3, 5 and 12V rails will otherwise explode for example)
but yeah, the writeup shouldn't be too far off
infact, i gave him an LM4766 stereo 40W chip and he assembled the whole amp on a 1.5inch sqaure prototype board and then fit the entire thing inside the PSU. granted it has a very small heatink (maybe 2x3 inches by 1/10 thick) however thats offset by the fact it has a fan - the sound of which is easily overcome by the power of the amp
but i should also warn you, modifying a PC PSU is not entirely simple, and definitely not for beginners. it requires a lot of thought, some planning and the removal of quite a number of parts (ALL the caps on the old 3.3, 5 and 12V rails will otherwise explode for example)
but yeah, the writeup shouldn't be too far off
Results with an SMPS may cause problems with some audio kit.
I use a Sony car radio/cd in my dining room, it's the place we live most of the time, and a car radio doesn't take up much space. This unit is the third I've used over the years, powered by a 13.5v linear supply mounted in the airing cupboard behind the kit.
However, the linear supply is getting long in the tooth, and I thought I'd replace it with a natty SMPS supply when I put in the Sony. So I used an SMPS to do the first out-of-the-box check on the new unit. ........... Oooo dear! ............ but did I get a lot of funny whistles and swishes. Fortunately, I had the sense to try it with a linear supply before rushing it back to the shop, and that showed no trouble at all.
The SMPS supply I used is the well established L30-10, a variable output professional/industrial unit, so I had a look at what it was producing on my 'scope. To be fair, it didn't look too bad. I didn't keep a record of what I saw, but I seem to remember there were switching spikes of around 100mV at 25-30KHz, and I was probably test loading it at a minimum of 2 or 3 amps. It was pretty obvious that the switching frequency was low enough too produce audible beats, although I was surprised that a unit which is expected to perform in the 'dirty' environment of a car didn't do better.
It needs to be borne in mind that switchmode supplies do not regulate to the actual output point, but sample the feed to an output LC filter, and that's all the smoothing you get. And even with a true, additional, linear output regulator, any switching spikes would probably pass straight through.
Being more optimistic, and having several SMPS around to check, it seems that current designs are more likely to have switching frequencies upwards of 50KHz - I measured one at 270KHz, so maybe the problem I encountered would be less likely if a recent unit was to be used. But it does suggest that you should try and rig up some sort of prior test, approaching what you have in mind, before commiting much time/money/effort to running an audio rig from an SMPS.
JohnT
I use a Sony car radio/cd in my dining room, it's the place we live most of the time, and a car radio doesn't take up much space. This unit is the third I've used over the years, powered by a 13.5v linear supply mounted in the airing cupboard behind the kit.
However, the linear supply is getting long in the tooth, and I thought I'd replace it with a natty SMPS supply when I put in the Sony. So I used an SMPS to do the first out-of-the-box check on the new unit. ........... Oooo dear! ............ but did I get a lot of funny whistles and swishes. Fortunately, I had the sense to try it with a linear supply before rushing it back to the shop, and that showed no trouble at all.
The SMPS supply I used is the well established L30-10, a variable output professional/industrial unit, so I had a look at what it was producing on my 'scope. To be fair, it didn't look too bad. I didn't keep a record of what I saw, but I seem to remember there were switching spikes of around 100mV at 25-30KHz, and I was probably test loading it at a minimum of 2 or 3 amps. It was pretty obvious that the switching frequency was low enough too produce audible beats, although I was surprised that a unit which is expected to perform in the 'dirty' environment of a car didn't do better.
It needs to be borne in mind that switchmode supplies do not regulate to the actual output point, but sample the feed to an output LC filter, and that's all the smoothing you get. And even with a true, additional, linear output regulator, any switching spikes would probably pass straight through.
Being more optimistic, and having several SMPS around to check, it seems that current designs are more likely to have switching frequencies upwards of 50KHz - I measured one at 270KHz, so maybe the problem I encountered would be less likely if a recent unit was to be used. But it does suggest that you should try and rig up some sort of prior test, approaching what you have in mind, before commiting much time/money/effort to running an audio rig from an SMPS.
JohnT
SMPS
Hi Francis
Hier some notes.
Oscilator of UCC3844 run on 200KHz and transistor
IRF810 and ferrite core(Siemens RM10) on 100 KHz.
Switch Mode Power Supply can handle up to 30W power.
I am veri suprised how this power supply influenced
on sound.At this moment i work on revision of circuit and PCB layout.AC voltage will be universal (88VAC-250VAC).At this moment it is for 220VAC.Do you have any expereince with SMPS.
You must take precautions when dealig with.PCB is very critical.
In my preamplifier i have no hume or buzzzzz.
In the preamplifier I have used the tube 6N1P one halfe
per chanel.In anode I have used CCS(10mA) an in
cathode NiCdm battery 1,2V.The coupling condenser
is WIMA MKP-10(1uF/400V).Volume control is
ALPS 100K/log.Signal is coupled with copper solide
core 0.1mm insulated with silk.
Regards
Hi Francis
Hier some notes.
Oscilator of UCC3844 run on 200KHz and transistor
IRF810 and ferrite core(Siemens RM10) on 100 KHz.
Switch Mode Power Supply can handle up to 30W power.
I am veri suprised how this power supply influenced
on sound.At this moment i work on revision of circuit and PCB layout.AC voltage will be universal (88VAC-250VAC).At this moment it is for 220VAC.Do you have any expereince with SMPS.
You must take precautions when dealig with.PCB is very critical.
In my preamplifier i have no hume or buzzzzz.
In the preamplifier I have used the tube 6N1P one halfe
per chanel.In anode I have used CCS(10mA) an in
cathode NiCdm battery 1,2V.The coupling condenser
is WIMA MKP-10(1uF/400V).Volume control is
ALPS 100K/log.Signal is coupled with copper solide
core 0.1mm insulated with silk.
Regards
SMPS
HI Francis
I am sorry for diley.I was busy last few days.I dont know where
you can bay exectly what you need.You probably need +/- 32V
to supplay LM3886.My plan is to make SMPS and power amplifier
with LM3886.It will be this summer.If yuo wish bee pationt meybe
we can make a deal.I am sorry for my English.I hope you understunde me.
Best regards
HI Francis
I am sorry for diley.I was busy last few days.I dont know where
you can bay exectly what you need.You probably need +/- 32V
to supplay LM3886.My plan is to make SMPS and power amplifier
with LM3886.It will be this summer.If yuo wish bee pationt meybe
we can make a deal.I am sorry for my English.I hope you understunde me.
Best regards
- Status
- This old topic is closed. If you want to reopen this topic, contact a moderator using the "Report Post" button.