Diode and capacitor on mains input

What's the capacitance? At a SWAG, I'd say it was to protect a toroidal power transformer against saturation due to small amounts of DC on the mains.

Cheers

Phil Hobbs

It's 470uF. Yes tiny DC is the only thing I can think of.

Jon

Den onsdag den 27. maj 2015 kl. 12.40.34 UTC+2 skrev snipped-for-privacy@gmail.com:

dc blocker:

-Lasse

Mains toroids also have horrible inrush transients--sometimes many cycles long, I'm told--so the cap may be there to improve the settling behaviour. (I've never used one myself.)

Is the power transformer a toroid?

Cheers

Phil Hobbs

** Switching the AC supply on near zero volts produces a maximum magnetic i nrush during the first half cycle, peaking near the next zero crossing. Sub sequent current peaks are normally much smaller with toroidals. NTC thermis tors are often used in series with primaries to render such inrush surges h armless enough.

The only purpose for a diodes & electro cap network in series is to offset asymmetry in the supply voltage wave - which often causes moderate core sat uration and an annoying humming noise to be emitted.

FYI the electro does the job and the parallel diodes protect the cap from t oo many volts.

.... Phil

Those are not back to back but, I think I can safely say that is a voltage doubler line supply.

I am guessing it might be close to this ?

AC mains ((Ac Maints-1.4) * 1.414)*2 || -||+--------+------+>|+-----------+ || | | DC Volts | | | |+ - --- ^ --- + + | | | | | | === | GND == GND

Jamie

inrush during the first half cycle, peaking near the next zero crossing. S ubsequent current peaks are normally much smaller with toroidals. NTC therm istors are often used in series with primaries to render such inrush surges harmless enough.

t asymmetry in the supply voltage wave - which often causes moderate core s aturation and an annoying humming noise to be emitted.

too many volts.

Huh?

Related anecdote, FWIW:

Once designed a low frequency power supply (~10kW), in the mid audio range. H-bridge into stripwound toroid. During initial testing, we first had it DC-coupled, which predictably resulted in noisy operation and various sorts of drifty, saturation-ey operation.

After adding coupling cap (and damping network -- don't forget that), it self-biased just fine. Which is an interesting affair, because steel will self-bias, magnetically, all its own, no DC needed.

The startup transient went something like, Initial pulse -- waveform was technically flux-balanced, but in practice there's a few percent of residual.

First "swing" of saturation: initial imbalance walks over to one side, clips some flux. Mostly manifest as decreased risetime on the voltage waveform, and a "nipple" on the current waveform.

Swings back (partly due to damped recovery exhibited by the coupling capacitor), possibly passing through zero and saturating a bit on the other side. Returns to zero.

Over a longer period of time (seconds), it drifts off to one side. The side may be unpredictable, or determined by the initial magnetization history. Not sure. The implication however is that, materials with a "butterfly" B-H curve (which they'll never tell you about, you have to figure that out yourself), exhibit some sort of DC negative-resistance effect under AC bias, which causes them to walk towards saturation regardless.

Returning to the subject of saturation and noise: The original prototype unit we were working with, always saturated fairly quickly, and noisily. Even out of saturation (as such), it remained noisy.

The other units (which weren't subject to early, uh, saturation tests..) remained fairly quiet and didn't saturate to one side or the other so dramatically.

Moral of the story: when you see in datasheets (typically of NiZn ferrite cores) that you shouldn't subject it to saturation or mechanical shock, there might be some truth to that. On strip cores, too.

Tim

--
Seven Transistor Labs, LLC 
Electrical Engineering Consultation and Contract Design 
Website: http://seventransistorlabs.com

That's my vote too.

Interesting.

I had mystery problems with one rather large variac (7.5kVA as I recall) that would out of nowhere start to buzz like crazy that seemed independant of load. It wasn't defective, and had something to do with power factor correction caps, but didn't seem to be a resonance issue at 60Hz. Might this have been some core saturation issue?

Other appliances on the same line upstream can cause that.

One good example is a microwave oven on a low amp line an you are connected at the end of it. I've seen distorted AC wave forms on the line due to high amp devices currently operating.

Jamie