Is that true? All of the guitar effects units we have here use cheap, chinese SMPS adaptors/wall plugs types thingies. Even consoles with 4-6 stomp buttons and pedals. For the latter I build a linear supply as the chinese SMPS blew up quickly.
Ah and the polarity is the opposite of what you'd expect from those adapters/chargers, centre pin is negative.
Thanks jcx for this lead, I will check it out. Good to know.
Are these all analog units? I thought that only on digital & sampling effects these SMPS were used? Then again I could be dead wrong!
JCX is it this one?
http://www.diyaudio.com/forums/diyaudio-com-articles/214808-amp-camp-amp-1-a.html
Are these all analog units? I thought that only on digital & sampling effects these SMPS were used? Then again I could be dead wrong!
JCX is it this one?
http://www.diyaudio.com/forums/diyaudio-com-articles/214808-amp-camp-amp-1-a.html
Last edited:
to look for comment I'd try: diyAudio - Search Results
many smaller "wall warts" with single DC V out, a lot are going sw mode to meet energy saving regs
you can buy split supply wall warts rather than make an active supply splitter gnd
many smaller "wall warts" with single DC V out, a lot are going sw mode to meet energy saving regs
you can buy split supply wall warts rather than make an active supply splitter gnd
It's a bit hard to find split-supply SMPS wall-warts - they're not commodity items, at least. I use 2x 19v, 2A netbook AC adapters to obtain -19v/0/+19v for the MiniRef 1875 nested chipamp, and it works well:
http://www.diyaudio.com/forums/chip-amps/184165-miniref-schematic-pcb-layout-10.html#post3922917
I did modify the adapters by replacing some electrolytics with 5000- to 10000-hour 105c Japanese low-ESR electrolytics, however. These are flyback converters, so in principle they can modified for -15v/0/+15v also, with a simple change of a resistive divider network.
Edit: They can also be modified for Spread-Spectrum clocking, at least in the case of those with UC3842B primary-side current-mode controllers. A 3 Mohm resistor from the unrectified AC to the switching clock RC network is sometimes found on some of these adapters for frequency-modulating the switching frequency.
Last edited:
I did try sourcing dual/split supply SMPS, but these are quite pricey! Nice work linuxguru! way above my head though!Perhaps in my next life!Lol 
Single supply SMPS are more readily available, hence thought I could perhaps use this split supply board to it to get +/- supply. The BUF634T too is a viable option. There's already one on my bench for load testing!
I found this 12v+/- module which looks just might do it. If the SMPS offers more advantages & less noise/problems,then I will pursue that path, otherwise, I'd stick to my trusted linear supply! It's just that I want to experiment to see if SMPS are workable in analog circuits!
Single supply SMPS are more readily available, hence thought I could perhaps use this split supply board to it to get +/- supply. The BUF634T too is a viable option. There's already one on my bench for load testing!I found this 12v+/- module which looks just might do it. If the SMPS offers more advantages & less noise/problems,then I will pursue that path, otherwise, I'd stick to my trusted linear supply! It's just that I want to experiment to see if SMPS are workable in analog circuits!
Attachments
Last edited:
Cool - the LM2575 is limited to about 1A, and the flywheel diodes look to be 1A rated. The inductors are fairly small, too. It should be fine for small-signal analog circuits, maybe with some additional LP filtering.
Hi,
I'd nowadays use isolated DC-DC converters, where isolated stands for galvanically isolated.
3W modules may deliver up to 125mA at +-12V, come in casings as small as a DIP28, and may have very low ripple figures of <20mV (found even some with <8mV).
The module in the thread before looks very basic and non-professional, but cheap.
I'd certainly worry about its EMI compliance and performance in general.
Have a look at the listed manufacturer's websites (add Murata) at the usual distributor's websites and You find thousands of modules to choose from.
jauu
Calvin
I'd nowadays use isolated DC-DC converters, where isolated stands for galvanically isolated.
3W modules may deliver up to 125mA at +-12V, come in casings as small as a DIP28, and may have very low ripple figures of <20mV (found even some with <8mV).
The module in the thread before looks very basic and non-professional, but cheap.
I'd certainly worry about its EMI compliance and performance in general.
Have a look at the listed manufacturer's websites (add Murata) at the usual distributor's websites and You find thousands of modules to choose from.
jauu
Calvin
Yes,It's more than adequate for my needs, as I need only a little over 100mA/rail However, I find the circuit below to be one of the most quiet & flexible psu I've worked with! I will just omit section A when connecting to a SMPS.
Attachments
Like the MuRata NDTD2415C -- Pete Millett used in his sound-card interface project -- they come in all sorts of flavors -- but remember to put a small inductor on the power-in:
NDTD2415C | Bipolar / Dual Output Isolated | DC-DC Converters | Products | Murata Power Solutions
I have missed the point of this thread, why replace the transformer + rectifiers with an SMPS unit? For 100mA at 12V we are talking a very small transformer, such as VB3.2/2/9 - BLOCK - TRANSFORMER, 3.2VA, 2 X 9V | Farnell UK or perhaps a size larger. The transformer solution is virtually noiseless compared to the SNMP and much safer in terms of large voltages appearing at the output. Unless you want to power LEDs or something.
Hi,
is there any proof for Your claims??
What do You calll noiseless?
IIrc 50Hz/60Hz and its rectified ripple is right within the audio range (not accounted for the rectifier diode switching noise), while the switching frequency of an SMPS or switching converter will be in the ultrasonic range from a couple of dozens of kHz up to the low MHzs.
Apart from always possible ground loops the switcher is much less noisy (virtually noisefree) over the audio range.
True it gets noisy in the ultrasonic range, but we can deal with that easier and more effective as with noise created right within the audio band.
The high switching frequencies allow for much smaller higher quality post filtering and regulation may be much better too.
Quite often no additional post regulating stage is required -which would be ´too slow´ in many cases anyway, for examle the LM317/LM337 regs.
A small Gyrator and rather tiny high quality caps are highly effective in that case.
A good powersupply design blocks the power line against the supplied circuit over at least that´s complete bandwidth, which may reach into the high MHz range with modern OPamps.
A classic 3-Pin regulator design alone using the known candidates is certainly not sota any more.
How, why and when could large voltages apear at the output???
Maybe You´re talking of SMPS AC-DC-Converters where rectified powerlinelevel voltages may appear?
Almost all switching converters feature a soft startup, some even allow for setting the startup ramp individually and externally.
In case of overvoltage/overcurrent, most converters feature protection mechanisms.
Due to the very short ´reaction time´ and much smaller associated capacitances a switcher will shut down much faster than a conventional transformer supply.
In general there are more and fast safety mechanisms at work in a switcher than in a conventional supply.
I can´t find one single claim of Yours to be true.
jauu
Calvin
is there any proof for Your claims??
What do You calll noiseless?
IIrc 50Hz/60Hz and its rectified ripple is right within the audio range (not accounted for the rectifier diode switching noise), while the switching frequency of an SMPS or switching converter will be in the ultrasonic range from a couple of dozens of kHz up to the low MHzs.
Apart from always possible ground loops the switcher is much less noisy (virtually noisefree) over the audio range.
True it gets noisy in the ultrasonic range, but we can deal with that easier and more effective as with noise created right within the audio band.
The high switching frequencies allow for much smaller higher quality post filtering and regulation may be much better too.
Quite often no additional post regulating stage is required -which would be ´too slow´ in many cases anyway, for examle the LM317/LM337 regs.
A small Gyrator and rather tiny high quality caps are highly effective in that case.
A good powersupply design blocks the power line against the supplied circuit over at least that´s complete bandwidth, which may reach into the high MHz range with modern OPamps.
A classic 3-Pin regulator design alone using the known candidates is certainly not sota any more.
How, why and when could large voltages apear at the output???
Maybe You´re talking of SMPS AC-DC-Converters where rectified powerlinelevel voltages may appear?
Almost all switching converters feature a soft startup, some even allow for setting the startup ramp individually and externally.
In case of overvoltage/overcurrent, most converters feature protection mechanisms.
Due to the very short ´reaction time´ and much smaller associated capacitances a switcher will shut down much faster than a conventional transformer supply.
In general there are more and fast safety mechanisms at work in a switcher than in a conventional supply.
I can´t find one single claim of Yours to be true.
jauu
Calvin
Well, I hope you read my 1st post? It was just I was thinking aloud about SMPS not being so common in analog circuits & hence thought perhaps some one might be able to educate me on this!
I don't have much experience with SMPS other than using these in lap-tops, peripherals & in digital musical equipment. These are always slim, light weight & very quiet! On the other hand, I found linear transformer circuits relatively easy & fun to DIY & to work with for semi-noobs & kitchen table wizards alike! However, few things I've come to know with these are that;
1. Some of them exhibit a loud acoustic vibration
2. Fairly expensive vs cheap SMPS.
3. Quite difficult to obtain precise +/- rail on split supplies.
4. can be quite expensive designing highly filtered & almost noise free units!
5. Only around 50% efficient & exponentially falling with increased heat with power demands!
SMPS too have their quirks like,
1. High swithing noise!(specially in cheap units)
2. An unfiltered output may cause glitches in digital circuits or noise in audio circuits
3. EMI/RFI produced due to the current being switched on and off sharply.
4. Requires additional filtering.
Q:What effect will this "switching cycle" in SMPS have on analog circuits?
Q;How safe is SMPS in small anlog signal equipment?
Last edited:
SMPSs are used to power analogue circuitry all the time...and in very sensitive analogue circuitry. All power supplies create EMC.
They are as safe as linear supplies.
The switching cycle (like a mains based liner supply) will only have an affect if the PSU is badly designed.
Some more in depth reading regarding noise, EMC design is a critical part of any PSU design be it linear or switched, good design means good PSU from any topology; bad design means noisy power from any topology:
http://www.diyaudio.com/forums/power-supplies/193705-switcher-emc-design.html
http://www.analog.com/library/analogdialogue/archives/41-06/ground_bounce.pdf
There is a thread on eevblog, Laptop Power Supply adaptors - mains on the output! - Page 1
It concerns mains AC to DC supplies - isn't that what we are discussing here?
It concerns mains AC to DC supplies - isn't that what we are discussing here?
- Status
- This old topic is closed. If you want to reopen this topic, contact a moderator using the "Report Post" button.