Gentlemen,
What's the exact technical term for the switcher topology where you only have one large toroidal (typically) transformer right at the mains entry point which takes line AC and outputs it at a lower voltage level whereupon it's fed into the bridge rectifier and thence into a large storage cap whereupon it can be chopped up, regulated and finally filtered, there being no other transformers involved?
thanks.
Hmm, not sure there is an exact name for this. Sounds like an offline 60Hz transformer/rectifier/capacitor-input filter, and then a buck regulator. The regulator would have an inductor in it.
Jon
The new way to do this is with a small transformer running in the multi-hundreds of kHz and even into the Mhz region. The toroidal 60Hz transformers are Big, Heavy, and expensive.
True, but they're also the quietest option when you can't afford the least itty bit of noise riding on the output - and this is in a high-end signal generator where purity is a major criterion.
an antique switcher topology
In Europe they were often called "ironless transformer" because they used a small ferrite transformer. IIRC often used to drive LV halogen lamps or lamp chains but that is probably a market of the past now that everything goes LED.
I'm 'old school', most of my G-job SMPS's run at just over 20kHz ;-) ...Jim Thompson
That's nasty for animals. Mine mostly run north of 300kHz.
Once when our Shepherd mix was still with us I had a SMPS prototype go into choking. I couldn't hear it but the scope showed it. She gave me "the look" and left the office with a sigh that I am sure she did on purpose.
There are some VERY quiet power supplies in the switchmode camp. 'least itty bit' isn't how the manufacturer describes 'em, though.
Size and efficiency matter; a low-power supply can easily get LOTS of RF filtering from the big 60 Hz magnetics, and those are lossy at lower frequency, too. It's a 'free' broadband filter. A switchmode topology can be just as quiet, with similar bulk and expense, if filtering is designed into it. That filtering is the place to study, not the block-diagram.
In that case, if the power is low enough, skip all the switching crap, and use a linear regulator. If not, then serious shielding and filtering is required. This is some of the stuff that can keep you awake at night.
Jon
Seems to me you've got 2 separate bits there, a traditional iron psu and an sm voltage regulator.
NT
I really don't think so, Snowflake 1. If you know better, by all means post a link and disprove it.
Kind of looks a bit that way to some extent, I agree. The first bit is like you would expect to find in a linear supply; the rest is kind of buck converter style. But I'm looking for a vastly more pithy description I can google for the precise deets.
How about "Conventional transformer-rectifier capacitor-input unregulated supply followed by some switchers"?
That would probably be a lot more more expensive than buying a UL/CE MeanWell PFC offline switcher to make your highest regulated DC voltage, and then maybe some downstream switchers to make other stuff.
You kind of hit the nail on the head there description-wise, but I was hoping for something a bit more pithy to search under.
If the regulator is on the secondary side it's just a step down pr sthg... if the transformer is actively driven there's a few different terms. push-pull and half-bride being somewhat popular...
OK:
"hybrid"
High frequency actually makes a clean output easier. All of the components get small and cheap. I rarely do SMPS supplies under 2MHz anymore.
One of the few logical reasons I can think of for doing it the way you describe is if semiconductor switches with a high enough collector-emitter or drain-source breakdown voltage that could handle the peak rectified 220VRMS line voltage weren't available or too expensive, or the few devices that could withstand the voltage stresses and were affordable weren't fast enough to build an efficient flyback.
These were probably significant engineering challenges at one time but I ain't that old. I don't see why a PS topology using an iron step-down input transformer plus a chopper afterwards would be intrinsically lower noise than any other as far as the output signal is concerned, but it would be interesting to know why if it is the case.
That'd be a good vanity license plate though: SNWFLK1
------------------------
** There is no recognised name - it is not a SMPS at all.It's a conventional AC to DC supply with switching regulation of the DC.
.... Phil
Have something to add? Share your thoughts — no account required.
Ask the community — no account required