Flyback transformer design confusion!!

What is it about stupid people that makes them feel that they should express an incorrect argument in incorrect terms and then use an incorrect concept to support their ideas.....?

Why not just state things correctly from the outset and explain it?

Of course that means you don't get the opportunity to TROLL someone up the garden path and then vacate the premises having lost the argument or..... god forbid, pounce on them with your final AHAAAAAA!!!!!

You also miss out on the opportunity to further confuse the issue by snipping, re-arranging or adding quotes/content as desired and hinting that you 'know' something or hinting that the other person does not.

So Charles, let's play the statistical game....

The thing you are missing in your statistical argument (ooooops sorry, obviously you know but you're not telling so you can look really clever) is.......?

Or are you tooo sure of yourself?

Perhaps it's time to go find yourself a nice political thread to wallow in.

Cheers

DNA

Unfortunately

Imax = Imin + dI

dI = Vin.Ton/L

Imax = (Imin + Vin.Ton/L)

Using the (incorrect) equation

E = L(Imax - Imin)^2/2Ton E = L(Imin + Vin.Ton/L - Imin)^2/2Ton E = L(Vin.Ton/L)^2/2Ton E = Vin^2Ton/2L

Which is...... wrong.

For discontinuous operation

E = (Vin.Ton)^2/2L

Which you would have got from the (incorrect) equation by setting Imin, which is what it is for discontinuous operation, to zero in the first place.

The above is, sort of, useful since you can rearrange it to get

Lpri = (Vin.Ton)^2/2E

Then, ignoring efficiency and energy transfer concerns, E = P/Fs

Lpri = Fs(Vin.Ton)^2/2P

'IF' you design for a maximum of 50% duty cycle on the primary side then

Ton = 1/2Fs

So

Lpri = Fs(Vin/2Fs)^2/2P Lpri = Vin^2/8FsP

However that would just be a first cut at the primary inductance.

You would use that along with a guesstimate of efficiency and leakage inductance effects to find a new power level so you can calculate a closer value for the required primary inductance.

I suppose in the general case.......

Ton = D/Fs

so

Lpri = Fs(VinD/Fs)^2/2P Lpri = (VinD)^2/2FsP

The boundary condition is the duty cycle for continuous operation which is given by.....

D = N.Vout/(Vin+N.Vout)

so

Lpri = (N.Vin.Vout/(Vin+N.Vout))^2/2FsP

And it all begins to get horribly complicated so I'm not going to bother.

There is the opportunity to vary turns ratio and duty cycle to trade off voltage stress on the primary switch versus that on the diode and vice versa and a must not forget is the energy/power lost leakage inductance since it resets to the primary side. There are two parts to it, E=LIleak^2/2, and the average current through Lleak during reset sitting on the reflected secondary voltage multiplied by the reset time. And thats all tied up with the reflected secondary and the effective primary clamp voltages.....

It's a whole new bucket of simple algebra but bollocks to it.

DNA

Hey DNA,

no no no no no. you and I are both clearly wrong, and the reason all our flybacks work (and the measured waveforms exactly match the calculated and simulated ones (I dont even bother to simulate them anymore)) is no doubt due to pure chance. that our maths is consistent and self-supporting is likewise an optical delusion.

Cheers Terry

Terry, Don't you think you are now starting to sound like a imposture.

Initially you wrote

ROFLMAO for a 50Hz ac power system driving a resistive load:

Vaverage = 0

Iaverage = 0

==> Paverage = 0

by your reasoning. which (given that my electric oven heats things up) is patently untrue, and is of course (in the case of AC) why we use RMS.

Then you wrote : Wiki defines Average as a central tendency of a set (list) of data here are ---many different descriptive statistics--- that can be chosen as a measurement of the central tendency.

And you claimed that only the arithmetic mean is an average despite what quoted that Wiki wrote.

Then in your answer to me you admitted that the rms is one of those descriptive statistics . If rms = average then by your own personal statement the stuff you wrote after ROFLMAO is wrong.

Personally I feel your sleazy attempts to cover up your own faults is now becoming humiliating.

I admit I am unable to judge on the rest of your debate, though initially I was backing you I am now starting to have doubts on your credibility.

The question we ask ourselves is for what reason should you use second rate methods to show that the guy is wrong even on a point where he is clearly right and you yourself have admitted indirectly ? People with such valueless attitudes like yours do not benefit neither the scientific community nor the human social environment.

But would someone like you care? Only you can answer that.

Cheers

Charles

snipped-for-privacy@gmail.com wrote: [snip]

absolutely correct.

you just left off the last sentence of the wiki definition:

"The most common method, and the one generally referred to simply as the average, is the arithmetic mean"

no, I never claimed it was the *only* one. as the wiki entry clearly states (in the line you snipped), the general meaning of the word "average" is arithmetic mean.

which I included in my penultimate statement:

"but LJ's statement "Power = average voltage * average current" is just plain wrong, given that an unqualified "average" refers to the arithmetic mean"

note the weasel word "given" and the associated qualifying statement.

I could however also write:

"LJ's statement "Power = average voltage * average current" can be viewed as correct, given that an unqualified "average" may refer to the Quadratic mean, or Root Mean Square"

but were I to do so, I would add:

"however given that "average" is generally taken to be the Arithmetic Mean, LJ's statement "Power = average voltage * average current" is somewhat misleading, and will give erroneous results unless the Arithmetic Mean is used, eg analysing sinusoidal waveforms"

because it is. but how is one supposed to tell which particular central tendency measure one is using?

this is the sticking point.

I can make analagous arguments to yours and say "median = average" and "mode = average". but thats really rather disingenuous.

when engineers use RMS quantities, they tend to mention it. OK, its often omitted when dealing only with sinusoidal waveforms (IOW its assumed), but when dealing with non-sinusoidal waveforms it is invariably mentioned. otherwise, its just begging for mistakes to be made.

in this particular instance, we have spent some time discussing arithmetic means and, like almost everyone, calling them averages (all of the charge balance calculations are explicitly arithmetic means), mostly because thats the common meaning of average, and also because arithmetic mean is a PITA to type repeatedly; only mathematicians are that pedantic.

I dont think its sleazy to assume, like almost everyone else, that an unqualified "average" refers to an arithmetic mean, especially when the formulae we have been bandying about have implemented said arithmetic means.

just follow the maths; its all there. If you like, cross-check it against Genomes post. He took a different approach, but we are describing the same thing, so you can "morph" his equations into mine, and vice-versa. If you cant, either or both of us has erred.

well he is not "clearly" right. I doubt you'll get many engineers to assert that RMS = average. mathematicians and lawyers though....

wow, you can divine my values from a wiki quote? cool trick!

If I was really pedantic, I'd point out there are enough nots in the above statement to negate your intended meaning, but I am not ;)

tell you what, instead of guessing my "values", how about you teach me something about electronics. I'm always keen to learn, and I've coughed up the bulk of the maths necessary to design a flyback converter (OK, I havent given the small-signal analysis here, but almost everything else)

even better, tell me about some mistake(s) you found in a circuit (your own or others, it matters not), and how you fixed them.

Cheers Terry

Terry, Your more compromising position is appreciated.

The point is your 3 mathematical expressions were extraneous avoidables because it was comprehensible that he was talking about rms voltages and currents.

And to be carefully dead on I would say when applied to sinusoidal voltages the expression Average Power = Average ( Voltage*Current ) is more accurate then

your expression

Average Power = rms Voltage * rms Current

My point is that emphasizing minor points of a distinct topic is shabby and deviating.

And your quotation of the Wiki article confirmed that rms values may be viewed as averages but you continually said that he was wrong which in itself was misleading or more "erroneous" as you admit now. The layman isn't aware of what an rms value is but LJ probably assumed readers here know its relation to the word average. I seem to recall he mentioned that.

I checked the first equation of Genomes post and it was a wrong citation of LJ's so I supposed the rest to be only folly.

I didn't consider seriously LJ's work but I may look into that if I can understand what he says and compare it to your work.

I was not attacking your values as a person but rather similizing your attitude in equating a legitimate statement as a presumable error as rather lowly.

If in any way my post is unmannerly towards you I apologize but I consider it the most chin up way of facing subnormal ethics for the interest of many.

Cheers

Charles

Wrong number sir! I said it was power not energy

Untrue man.

I just don't like typing too much. So if anyone is going through hell just come back and ask me.

lemonjuice

ALL INCOMING MESSAGES CONTAINING BARTOLI ARE DELETED BY SYSTEM

I live to serve :)

perhaps. but plenty of people read usenet who arent practising EEs, and LJ's statement is pretty much guaranteed to confuse them.

Bwahahahahaha!

they are in fact identical:

consider:

V = Vpk*sin(wt)

I = Ipk*sin(wt)

Vrms = sqrt(integral(Vpk^2*sin^2(wt))/T)

Irms = sqrt(integral(Ipk^2*sin^2(wt))/T)

where integral is from 0...T, and w = 2*pi*f where f = 1/T

we can rearrange these to give:

Vrms = (Vpk/sqrt(T))*sqrt(integral(sin^2(wt)))

Irms = (Ipk/sqrt(T))*sqrt(integral(sin^2(wt)))

you say toh-may-toh:

Pavg = (1/T)*integral(V*I)

Pavg = (1/T)*integral(Vpk*Ipk*sin^2(wt))

Pavg = Vpk*Ipk*(1/T)*integral(sin^2(wt))

I say toh-mah-toh:

Pavg = Vrms*Irms

Pavg = (Vpk/sqrt(T))*sqrt(integral(sin^2(wt)))*(Ipk/sqrt(T))*sqrt(integral(sin^2(wt)))

Pavg = Vpk*Ipk*(1/T)*integral(sin^2(wt))

Ego, neither is "more accurate", they are *both* *EXACT*

interestingly enough, this is basic high-school maths (read as: trivial)

and what do you call it when you say the exact same thing as me, then claim its not only different, but more accurate?

and laypersiblings (note non-gender-discriminatory term) wont read this?

that assumes LJ's post is right. Genome didnt "cite" LJ, he just wrote the relevant equations down. do the math (assuming you can), you will see that Genome is in fact correct, and so am I.

just dont make his mistake wrt delta_E.

bang in some numerical examples, its clearly wrong.

and really, flyback analysis is baby maths; any decent engineer can easily derive all the relevant equations from scratch, its trivial.

gosh, thats big of you.

wow, now my ethics are subnormal too.

another ad hominem attack (although, like the first one, couched in seemingly polite verbiage).

at least my maths (and hence engineering) is correct.

dont you have anything to say about electronics? I didnt think so.

but thanks for the laughable assertion that:

Average Power = Average ( Voltage*Current ) is more accurate then Average Power = rms Voltage * rms Current

Cheers Terry

LMAO of course @ Terry Man don't waste your time on that twit ... Remember one thing you can't recycle is time. After seeing his faulty spectale on ac voltages I understood if a man doesn't understand ac how can you even dream about talking to him about flyback transformers!

Besides only plants thrive under bullshit.

lemonjuice

ALL INCOMING MESSAGES CONTAINING BARTOLI ARE DELETED BY SYSTEM

A possibility. In that case you should criticize every engineering comment not understandable to the layman. I do not see you engaging frequently in this endeavor.

A very laughable statement to any amateur.

deluded.

Surely this is gut busting to say the least from someone who is howling "Ba ha ha ha".

Did you never learn how to write the correct expression for a

general sinusoidale voltage.

Did you criticize my High

school education.

Another mishap. My deepest commiseration.

Even with the wrong inputs the result is wrong . Either the calculus or the trigonometry or both went rotten.

I am sorry but the mistakes are so unbearable I reserve all my comments until you rectify them.

The starting distress above is a consequence of a misrepresentation of the key components of a sinusoidale waveform.

Again my deepest sympathy.

Obviously no. That is why I corrected your inaccurate formula for, um physics or chemistry? I don't seem to remember.

I loved it .

You split my sides from laughter with that flapdoodle.

You certainly cannot be American where commedy is more open.

My pity and I agonize with you if I have caused any pain but I wanted to remain in line with your comical interpretation.

Be well.

Cheers

Charles

Enough , enough, enough...Please.

Lemonjuice, Terry Given, snipped-for-privacy@gmail.com, genome, bartoli could you all please shut the hell up.

This thread has gone beyond the limits of mental sanity so we don't care anymore who is right because at the moment you're all behaving like a bunch of monkeys at the zoo.

snipped-for-privacy@MAILFORCE.NET a écrit :

Yeah. Welcome into the cage.

nobody forced you to read it....

Cheers Terry

snipped-for-privacy@gmail.com wrote: [snip]

it is indeed clear that you are deluded.

yet you offer no mathematical refutation, merely verbiage. fraud.

how can I, it appears to be missing.

what trig?

your comments appear to be worthless. the maths is correct.

if it were wrong, you ought to be able to prove it mathematically, which you conspicuously neglect to do.

I'd love to see you expand on this. just what key components of a sinusoid are missing?

I doubt you can tell the difference.

you certainly cannot be an engineer, where mathematical skills are extant.

note careful editing to try and make it look like I said that crap.

just sore sides from laughing so hard at your complete lack of understanding of high school mathematics.

I filtered lemonbreath and elephantiasis a long time ago.

Cheers! Rich

Hi Rich,

I think LJ just made a mistake, but elishaphat is either a troll or an idiot. fancy asserting RMS = average. I especially liked his complaints about my representing a sinusoidal voltage as Vpk*sin(wt) :)

Cheers Terry

Terry Given writes

How dignified I feel hearing Terry Given exalting me with such monumentale titles like Troll and Idiot. Continue with this high minded behaviour. But our ole boy goes contradicting himself again.

Certainly an inexact representation that will give you the an faulty expression for average power.

The formula you scribbled down

Average Power = V rms * I rms is a sign of carelessness. It is precise only for SOME cases!

Cheers

Charles

The expression below of Terry Givens is totally wrong and is an insult to engineering

Your Math is hoplessly wrong

Your expression for Power is wrong Mr. Terry Givens. I had used before the kinder expression inaccurate. Assuming you consider a case where your formula for ac voltages or current is correct.

I = Ipk*Sin(wt) would be the current according to the formula you wrote above.

the ac voltage across an inductor is

V = L di/dt = Ipk*L*Cos(wt)

Instantaneous Power across the inductor is

Ipk^2*L*Sin(wt)*Cos(wt)

Integrating the above to get the average power in 1 cycle gives you a value of zero. I hope you can see that as you say your understanding of high school math is very good.

Yet According to your formula

Pavg = Vrms*Irms

Average Power is NOT zero ... Wrong result for you Mr. Terry Givens.

Now who I'm really starting to wonder whether you are an engineer ... or a kid just playing around here on this newsgroup?

Not hard to guess what is really laughable or is it?

Cheers

Charles

Ooops. I noticed I had made a typo in my original message. There should be a "multiplied by w" in the Voltage expression.

This though doesn't change the whole substance of my original message.

Cheers

Charles

ALL INCOMING MESSAGES CONTAINING BARTOLI ARE DELETED BY SYSTEM