Good afternoon everyone

Take a look at this:

This is supposed to be an AC to DC adapter. The AC first enters the full bridge rectifier and the resistor and then it reaches the transformer. Since the output is a positive sinusoidal wave...it's just not DC. The insulation from the mains is also bad. How did it get CE approved? How can this stuff be marketed??

self-approved, what insulation test did you do? Cheap product needs cap.

@DanielTork The resistor is for the LED. The current coming out goes in one single direction, so it is DC. However did you come to the conclusion the transformer isn't isolating properly? And CE is a self assigned label that only ever gets taken away

@dirac16 A fraction of a fraction of a micro second before the left transistor switches on, what's the voltage across the capacitor, approximately/likely? And a fraction of a fraction of a microsecond after?

@Asmyldof That's more of a DC with a AC component. I don't think it's actually practical for devices. At least others, as the one it was design for can accept this. I thought that standard rules ask for a smoothing cap.

I was also expecting an optcoupler. It was made for charging the batteries of a toy car

@TonyStewart I can't say I've tested it, but the simplicity made me question it.

For charging batteries no cap is needed , the battery is the cap.

With sinusoidal waves? Never heard of this before.

@DanielTork -- using the battery as a cap is normal for cheap lead-acid chargers for car batteries and such

however, (!), how the heck are they not sending that transformer into saturation like crazy?

Why would that be?

Don't see why

the DC bias in the full-wave-rectified waveform

but for LiPo 7.2 @ 0.25A must have OVP peak clamp to be safe for battery

@Shalvenay The XFMR does not see DC but rather alternating narrow current pulses before peak voltage.

@TonyStewart betcha it's NiCd/NiMH not Li-anything

@TonyStewart if there was a cap there, I'd agree with you...

battery is bigger in Farads than any cap

It charges and discharges for itself...

let me know if you understood what I said

ah, I see

So tell me more about this

so the current would only flow when Vsecondary is greater than Vbattery

which is just before each peak voltage then cuts off

@Shalvenay If they were LiPo it woul be quite dangerous

the ripple voltage is inverse to the ripple current crest factor

*would

@DanielTork yeah

so an 8.4V clamp is necessary and a timer to remove it after a period of time

@TonyStewart How can the battery be used to replace the smoothing capacitor?

I don't really picture it

first understand that a battery is a capacitor with a chemical cell voltage in series. A 3Ah LiPo cell has more kFarads than big super cap

Okay

That makes me think i can use an empty battery as a supercapacitor

yes but limited by life charge cycle

the chemistry and electrodes share similarities

@DanielTork DC means the current goes in one direction. Which it does. it is DC. Period. If there's a capacitor it's smoothed DC. If you look with a scope it may say "it's 8V RMS + 6Vpp AC" , but that just means it's counting the ripple as AC. AC means alternating current. If the current goes in the same direction always there is no AC component to the current.

O...k different definition

So if it's alternating it's changing direction

And that's it

If I say the ripple is AC I'm wrong?

@Asmyldof almost , you might say the polarity does not change, but as we know from Fourier spectrum of polar pulses , it has a lot of AC

Actually AC component

@TonyStewart Agree with you

not wrong it is both AC and DC but net result is unipolar

If ripple voltage is 5% then peak current is 20x average current and duty cycle is 5% as a 1st order estimate

-Asmyldof loses it and prepares a lethal and crisp clear explanation to prove his point- :)

Ok, about the battery capacitor

I understood what you meant @TonyStewart

good

Please go on

@TonyStewart In frequency analysis there's components of (possibly mild 50(60)Hz), 100(120)Hz, etc, as with any ripple. The ripple is not an alternating current at the output, hence the adaptor has no AC component and is DC period.

Well Wikipedia says the same about direction

It has a pulsating component

That's the ripple

So alternating from alternating direction, right?

Brb

AC means alternating current polarity not alternating voltage so pulsed DC has AC although alternating voltage may or may not have DC

How do you define alternating current?

@TonyStewart In this case, the pulsed DC does not shift polarity, just instantaneous value

the voltage is not alternating but the current is, hence "AC"

usually specified as "ripple current"

AC RMS

This is true

Because the current waveform doesn't get clipped by the bridge rectifier

No, I'm wrong there

Current doesn't change polarity either

yes it alternates between charge and discharge

Indeed, but it doesn't go below 0

sorry but everything I said is true

tinyurl.com/lane6hk

Do you see alternating cap current polarity yet?

tinyurl.com/ko7q4hd

YEo

*Yep

ok

Thanks

That's below 0

That justifies everything

But

In the case of this circuit

There is no AC component @TonyStewart

then you must not be reading scope correctly

Well there is no capacitor here

Cap current is +8.47Apk and - 318 mA + and -

tinyurl.com/ko7q4hd

Here? you mean with battery?

Yes

The last lin is for battery?

*link

remember equivalent circuit for a battery?

Yep, what you said

Thanks

it is a cap with ESR and leakage R and ideal cell voltage in series determined by chemistry

I remember it, I can remember well the stuff I like :)

But just to clarify it, if the waveform of the capacitor wouldn't go below 0, then it wouldn't be AC component

Right @TonyStewart?

and it's V would keep climbing

it must alternate current to maintain a steady voltage

Ok thanks a lot for everything

so duty cycle and peak currents * time must balance for each polarity

Yet shouldn't the last simulation have a much bigger capacitor?

You said the battery has a bigger capacitance value compared to any other cap.

Oh sorry I didn't see those units well

It's ok

But the output waveform changes too much

Too large C

Anyone here knows a good ipv6 tunnel broker?

depends how you define features

Well, with a pop in germany and high reliability, thats probably more than you can ask for these days ^^

and free of course ^^^

beta.electronicdesign.com/analog/…

"That the company accepted the deal was surprising to an industry familiar with its engineering culture, disdain for corporate matchmaking, and fanaticism for analog design."