Power supply tank cap

I did plonk you now, you bloody utterly stupid Idiot

--
ciao Ban
Apricale, Italy

IEEEXplore only returns 12 papers with a Gaspari as an author, and only one of the was by an M. Gaspari; that paper was "An open framework for cooperative problem solving". I don't think that's the one you mean. Can you give a more detailed reference? I looked through the 2004 Industry Applications and didn't see a paper about rectifier-capacitor analysis; can you give a more detailed reference for that one as well?

This is an urban myth, the peak current is pretty much *independent* of the load capacitor size (exept its EMR). It is almost entirely only dependent of the (ohmic) interiour resistance of the transformer, diodes and cap. The tranny has a neglegible impedance influence at 60Hz, as you already found out. Only the *inrush current* when the cap is not yet charged is higher. The conduction angle stays basically the same and is only dependent on the Ri/Rload ratio. You cannot make rectangles out of sinusses or pulses. Make a spice simulation if you do not believe it. You can really model the tranny as an AC-source with the open circuit voltage plus an interiour resistance in series with the output.

Your basic mistake is to assume as Phil did, that you can charge the cap all the way to the peak sin voltage. This is not the case. The resistive losses prevent this. Out of this error your output voltage is a function of the cap, it is not. Only the ripple voltage is higher, the Vrms stays the same.

Which indicates that current flows also after the peak of the sin.

The proof of my thesis here is difficult to conduct, it involves piecewise integrations over hyperbolic curves and I will need some time to do this, if somebody is interested.

--
ciao Ban
Apricale, Italy

"Ban"

** So you completely missed this remark to the sheep fucker from me ?

" Funny how the rms value of the charging current pulse train stays nearly constant no matter how big the cap. "

Shame about your crap wog comprehension of the nuances of the English language.

** Been bashing that simple bloody fact into the Kiwi sheep f****rs thick as two planks head for years.

Something I know and you do not - wog boy.

** Again - that is your CRAP mis-comprehension of the nuances of English.

.......... Phil

Your question is ridiculous and the capacitor doesn't "raise" anything. There is so much slop in the transformer and filter components that any claim to precision is a pretense, delusion, or wish, and you have failed to account for a mains variation of typically 15%. A rough estimate of component values is to compute the equivalent load as RDC=VDC/IDC, select the filter capacitor to be > 100/(2pi*F*RDC), select transformer RMS current at 2.2*IDC, and select VA rating as 3xVDCxIDC. Then use big

10A diodes and ripple-rating on the capacitor(s) and observe maximum voltage ratings.

And stand WAY back when you plug that thing in.

"Fred Bloggs"

** It certainly raises the average value of the rectified AC voltage. ** For any given combination of physical components, the laws of physics apply just as precisely as ever. ** Good.

** Huh ??? Something very wrong here.

For 1 volt at 1 amp this gives a cap value of > 300,000 uF !

Maybe the constant should be 10 and not 100 ?

** Fine.

** Way too conservative plus *contradicts* the previous " 2.2*IDC " rule.

Since VDC = approximately 1.41 times the secondary volts AND the rated AC amps is 2.2 times the DC amps then the needed secondary VA is 2.2 / 1.41 = 1.56 times the DC watts.

This gives the *oft quoted and long proven ratio* of transformer VA to filtered DC watts of 1.6 times.

** Of course....

........ Phil

We can see your current state of knowledge, you don't need to elaborate.

16/4 =4R

C>66000uF wow!

I= 8.8Arms *12V= 106W

and

3*16*4= 192VA wow again!

So the tranny is supposed to have 106W or 192W, or is it VA??

better a ready made 10A bridge, or do you know any special discretes?

Well those 66mF caps will do this 10A ripple easily I think.

Hey Mr. Bloggs again an utterly helpful commentary.

--
ciao Ban
Apricale, Italy

Those specs ( not bothered to check your arithmetic) are the price you pay for brownout proofing, temperature extremes, and ultra long operating life. That little wisp of a thing you proposed is wholly unsatisfactory junk, lacks suitability for any serious application.

Right then allow for scale down to 0.8 for brownout and an additional

0.8 for transformer regulation and/or added secondary limiting/swamping resistance and you get something like 3.

At least you are familiar with O.A. Shade- too bad there wasn't a Shadioni or something for this Ban character to familiarize himself with.

The detail too much for you? RDC=12/4=3 ohms.

C in vicinity of 100/(377*3)=88mF, go with 2 x 47,000uF.

... makes no sense.

How hard is it to understand VA= 3*12*4= 150VA? So that makes it approximately 15Vrms @ 10Arms.

Vopenckt,max= 1.2(trans reg)*1.42(rms-pk)*1.15(mains)*15= 35WVDC capacitors and 100 PIV on rectifiers.

Yes they will, and output voltage ripple

It was supremely helpful compared to your unorthodox, unusual, unused and silly methodology. Stick with your small signal stuff.

I wonder what happens when the OP connects his appliance to this beast Well, now I finally understand why you were writing: "And stand WAY back when you plug that thing in."

--
ciao Ban
Apricale, Italy

Right- no telling what kind of mess he is creating for himself. Many people speak so highly of the ARRL literature, but from what I can see, it doesn't teach anything- the most clueless people in the world use it.

"Fred Bloggs"

** That is absolute bullshit.

What a lying rodent you are when caught out.

........ Phil

** That is absolute bullshit.

What a lying rodent you are when caught out.

........ Phil

** That is absolute bullshit.

What a lying rodent you are when caught out.

........ Phil

The handbook itself is of little use in anything having to do with actual circuit design (it barely even explores bipolar biasing, for example), but for a cookbook-type approach it's handy.

Strange you should write about this in the context of power supplies because IMHO it's strongest area is basic DC power supplies! You'd think even less of its approach to real RF design, I suspect :-).

Many of us grew up as pre-teens with access to zero technical literature outside the ARRL handbook, and I suspect this is why we still quote it 40 years later.

Tim.

I measured a small 50VA, 220V/2x18V toroid tranny that is working in many of my appliances, since 1984 I had no failure. Each side has a regulator(7815/7915) and 4700uF, load is 0.8A plus output current of 8 audio channels: primary 36.4R secondary in series 1.6R Total combined 1.6R + 36.4R/(220/36)^2 =2.6R with 1*12Veff that would be

0.29R. open circuit voltage with 227V input = 43.6V

Lets now make the calculation for the 4A 12V min.: assuming Diode Vforward=0.75V Vmin=(12V+1.5V*2/3 ripple) +15%= 14.3Vinf

Well I get the following values for +/-15% mains

Vrms= 12.7V; Irms=7.1A; P=90VA; Ri=0.225R C=25mF; Vmax= 21V; 10.8A peak ripple current with 10A bridge rectifier ripple is 1.5Vpp compared to 210mVpp of your solution

60% the space, heating and cost but not readily available. :-(

--
ciao Ban
Apricale, Italy

An interesting article that describes some pertinent considerations is available at: