Analysis and Design of a Flyback; Part 1, How to Analyze and Model a Flyback Converter
HTML-код
- Опубликовано: 29 дек 2012
- At 24:21 The derived transfer function is for a continuous conduction mode flyback. The correct derivations is shown below;
/ @robertbolanos
Video 2. • Analysis and Design of...
Tutorial on how to analyze, design and simulate a flyback converter. Voltages and currents are calculated and then compared with a SPICE simulation. For comments, questions can be posted here or can be emailed to rbola35618@aol.com.
At 24:21 The derived transfer function is for a continuous conduction mode flyback. The correct derivations is;
Energy in the inductor define by W
W=(Lp*Ipri_peak^2)/2, (EQ1)
You then divide by time (Tperiod) to get power (P)
P=(Lp*Ipri_peak^2)/(2*Tperiod), (EQ2)
solve EQ2 for Ipri_preak yields,
Ipri_peak=SQRT(((P*2*Tperiod)/Lp)), (EQ3)
Power is related to P=Vout^2/Rload, substitute into EQ3
Ipre_peak=SQRT((((Vout^2/Rload)*2*Tperiod)/Lp)) (EQ4)
The equation for the inductor is,
Ipri_peak=(Vin*Ton)/Lpri (EQ5)
Equation 4 and 5 are egual to each other so now we solve for Vout/Vin
Vout=Vin*(Ton/Tperiod)*SQRT((RLoad*Tperoid)/(2*Lpri)), (EQ6)
For TopSPICE files request, email me at rbola35618@aol.com. These files can run using the TopSPICE demo available at www.penzar.com
May your simulations always converge ;-)
Robert
Thanks for this great series on Flyback converter design. I have all my interns review and study your material. Your instruction and techniques reflect your years of practical experience and excellence. Thanks for passing this knowledge down to future generations.
Thank you much for the kind comments. It make my day hearing this a very pleasant day.
Just letting your know that your 9 years old video is helping me now. Absolutely interesting and lucid. I am watching the entire playlist.
I just got my first power electronics design project and it's a flyback converter. This 10yr old video is amazing, and clears so many things. Thanks profe, you're an inspiration.
I am glad you found it useful
This is the way to do things. From someone who knows what they are talking about. No need for any fancy graphics and slick (Annoying) presentations etc. Brilliant.
Hey, very useful video! Find this very easy to understand compared to reading most books and notes out there. Great job, thanks!
Thank you so much Robert,I learnt a lot from this and the simulation files you have sent me.
You are very welcome
Thank you Lim! I appreciate your kind comments.
Robert
very nicely explained,thank u very much!!
Nice video. Good job on your Flyback explanation.
Hi Jadhav, thank you very much for the feedback! best regards robert
Thank you EmSysDev for your question. To set the turn ratio of the transformer, you use the "OFF" model shown in 18.20. If you will notice that when the primary switch is off (Mosfet), the secondary diode is turn on and therefore Vsecondary is equal to Vout. What you are really doing is setting the resetting voltage needed for the transformer to reset and avoid saturation. Most people use the equation that I used to set the turns ratio. That means that the reflected voltage (from primary) that the Mosfet will see is Vprimary (voltage across the primary winding) will be Vout/Nr. That reflected voltage will be aproximatly Vin. This means that the drain will see 2Vin; Vin in from the input supply added in series with the reflected Vprimary voltage which should be approximately also be Vout(refected to primary from secondary).
I hope this answers you question. If not, let me know and I will post a more detailed video. But I won't be able to do that until next Friday.
Hello Mr. Robert nice video explaining the flyback converter as it is also my final year thesis but if i want to calculate the current in the secondary capacitor ic.... will it be Vo/Rload or -Vo/Rload
subbed, you are amazing
What does Subbed mean?
Very great video.
Do you have a video that shows why as increasing frequency would give more power for a switching mode power supply ?
I would like to know the formula that shows the relationship between flux density and the frequency.
And the formula that shows the relationship between flux density and power.
This was a great video. But I'm not sure how you ensure that you have an output current of 0.5 amps?
Excuse me, may i ask what the dot stands phisically for a coupled inductors ? I mean, given a real "transformer", how can i identify the right connections ?
Furthermore, i think there's something not clear about deriving [EQ6] from [EQ4] = [EQ5] in your main comment, are they correct ?
Hi Robert !
Thank you for this interesting and beautiful work !
I have question, normaly the duty cycle (D) or frequency (f) must be change for regulate the voltage output , but in this video, when you want to calculate the primary inductance thou hast seen, D and F are constant, I don't understand this point . Can you explain me this ponit in details ! please !
thank you in advance
This is a very helpful video. I wish it contained a little more in terms of voltage and current ripples, but excellent nonetheless.
Hi Ali Etezadkha,
Thank you for the feedback. It is hard for me to judge how much detail to included in these videos. I am always afraid that if I make the videos too detailed and long that people might watch the entire video. I will try to include details on voltage and current ripples. I might do a follow up video. Question and comments can be sent to rbola35618@aol.com
Best regards, Robert
things start to get interesting around 10:51
Thanks
Thank you very much! I did not see this until today.
@24:00 the formula for turns is Vo=Vin*N*D/(1-D),but at 36:00 the formula is Vo = Vin * N.Why are there 2 formulas for the same transformer ?
Informative video. Thanks !
D=0.5 that gives Vo=Vin *N
Robert Bolanos,
I'm curious about something. If you've got a single stage power factor correction unit (without a capacitor between the full bridge diode rectifier & the flyback converter) how does the system implement power factor correction, even though a flyback converter demands discontinous input current for the primary side of the transformer?
I appreciate any response you give me.
Robert, you have done so much work on me with this video than you can imagine. Thanks a lot. I have a few difficulties though and I thank you in advance for the solutions I hope to get from you.
1) I tried the modelings too using Topspice and I am grateful that I could get it right to the point that I did before I began getting it not right with error messages saying, "missing parameter(s) on dc analysis specification".
2) I understand the way you obtained the ESR value. I looked up a few datasheets of capacitors with the hope to find ESR values but they contained no such information, and it makes me feel that they may be widely different from capacitor to capacitor, even if they are of the same type and value. Now, considering bulk manufacturing of a flyback, is an ESR value obtained for one capacitor using the method you showed on the video okay for the bulk?
3) To obtain an output of 1.2V from a 120V to 240V AC input using a flyback built with off-the-shelf transformers, I searched for transformers to use and I found one with a moderate primary inductance. However, the turns ratio would produce 5V at the secondary which, to me, seems to be far away from the 1.2V. Considering efficiency, having to obtain 1.2V with this turns ratio bothers me a lot and I sincerely need your advice on it. I will take whatever advice you give me.
Thanks. Akanimo.
1. Please send me your TopSPICE file to rbola35618@aol.com and I can take a look at it.2. Yes, I use a rule of thumb equation of esr=65u/cap (for electrolytic) to get an estimate of the esr. If you use tantalum caps, then use esr=6.5u/cap.3. You can still use and will get 1.2V. You will have a higher reset voltage.
hi, very good and informative video thank you!
I have a question: since the slope of I at the primary is V/L, does that mean that when the MOSFET goes off, and the current drops to zero instantly, we get a peak of very high negative voltage? (-infinity for ideal switch)
chopcooey Hi Chopcooey, When the Mosfet is turned off, the current thru the Mosfet goes to zero. Keep in mind that when the Mosfet was on, the energy is stored in the gap in the transformer. So when the Mosfet is turned off, the energy that was stored in the gap thru the primary winding will then be released back into the secondary thru the secondary winding. So if the Iprimary_peak was let say 1A, and lets say that the turn ratio is Ns/Np=12V/24V=0.5. Then when the Mosfet is turn off, the secondary will have a Isecondary_peak= Iprimary_peak*(24V/12V)=2A. The ramp will be reversed from the Iprimary and will start at 2A and then ramp linearly to 0A. Hope this makes sense.
Hi Rakesh,
Thank you for your comments. If you email at rbola35618@aol.com, I can send you the Topspice files that I used in the videos.
Robert
"May your simulations always converge" ;-)
Hi Robert,
Can I use inverse coupled transformer ? How this will effect the circuit?
Robert,
Thank you for all the useful information. I have designed one flyback converter. When i checked the hardware the current increases even at fixed pulse width and load. What may be the reason for that?
***** You may be saturating the transformer. Check the primary current. The current should look like a linear ramp starting from 0V to a Vpeak. If it has an exponential curve, the transformer is saturating. Rewind to more turns. Robert
Nice job, I think background noise is not much and it's bearable because voice is clear.
One question: If my Vin is a 12Vdc. How would I set my Vmin and Vmax as? (This is according to your design example). Thanks
Hi Robert ,the energy stored in the fly back transformer ,stored in the core or in the gap
Usually in the gap unless you are using one that has distributed gap. Remember that the gap is in the core.
What if i want the system to be in BCM mode then what should i do ?
at the very last of the video you mentioned that vo= vin*Nr, which should be vo= (D/(1-D))*Nr*Vin
Hi, I am building a Vin=12V to Vout=325V flyback converter and say my load is a laptop at about 60W and using a duty cycle of 0.5. In that case, the Iin(ave) for my situation will be 60W/12V=5A. So my Ipeak=5A/(0.5*0.5)=20A? Am I calculating correctly? How does the transformer primary induce such a large current of 20A? I don't think any typical lab power source can supply that amount of current right? Besides that, assuming that the above is correct, my mosfet switch should be rated for above 20A right?
That is correct. That is a disadvangtage of using discontinuous mode flyback. Typically when that current get that high, people will then use continuous mode flyback which will lower the peak current. The other option is to raise the input voltage so that the peaks current will reduce. Hope this make sense. Robert
Hi, thanks for the input. I have another question based on your video on transformer characterization. Don't mind if I ask this here too.
Let's say i calculated my Lmin to be 2.1uH to maintain in CCM. Npri is calculated to be 2 using the formula shown in video 6. As I am looking at ferrite core (ETD34), the AL value is given to be 2.7uH/turn. Since my Npri is 2, my primary inductance will be 5.4uH? So in this case, the actual primary inductance is 5.4uH and not the calculated 2.1uH right? But in all cases, it shouldn't matter as long as the inductance is greater than 2.1uH to maintain CCM right?
Lastly, for the actual winding of the transformer, the secondary is 180 degrees out of phase with the primary so the secondary winding should be winded in the opposite direction from the primary right?
Thanks a lot!
I'm very inspired by your Flyback series. Is there a way to gain access to the files that accompany the flyback series? I would really like to do your entire series of lessons for the flyback. Would I just get the simulation software and look somewhere online on how to add the simulation parameters like how you have it in your videos but looking at your video and writing the same thing for myself or? I see at one part of the video it instantly jumps to the simulation and I have no idea how you setup your simulation and what I need to do to set it up like how you did. I really appreciate your content. Its the best content I have found that demonstrates theory and proves it via simulation and later in the lab. This I have no seen anywhere else. Everywhere else its a large theoretical book and elsewhere more practical with non trusty theory without any simulation.
Yes, email me at rbola35618@aol.com and I will send you my files and notes.
@@RobertBolanos Thanks a lot for your response. I sent you an email. Have a good weekend.
Branko
Hello Sir, how the polarity of the primary winding can be reversed when the switch is OFF? Thank you
Hi Gilang, When the Mosfet is on, the inductor or the primary is storing energy in the core's gap by building up a magnetic field. When the Mosfet is turned off, the magnetic field start to callapes and induces a back EMF. The phenomenal is knows a Lenz' law. I hope this make sense. Cheers, Robert
@@RobertBolanos thank you so much Mr. Robert
In case we don't have input voltage and need to calculate the transfer ratio what is the way to calculate input voltage ?
Hi Mahmoud, please see the derivation on shown below the title. This is the corrent transfer function for a discontinous mode flyback,
Around 17.50 my understanding of inductors was that they will try to
keep same state when voltage is removed. Lenz's law etc. So
before the mosfet switches off the secondary circuit's current is in
clockwise direction. with primary also in clockwise direction (as per
dots). So after switch of the collapsing magnetic field will try to keep the
currents flowing in the same direction. But you are saying the current directions are swapped. What am I missing here?
Hi Kevin, This is a good question. When the Mosfet is on, there is current in the primary only; no current is flowing in the secondary because the polarity of the secondary reverse biases the diode. When the Mosfet is turned off. the primary current does not flow and the polarity voltage in the primary switches because of Lenz law and therefore the polarity also switches and this time forward biases the diode. Hope this makes sense. Best regards, Robert
For calculating power in the design example, shouldn't one be using Rms current instead of Average current?
Avg current equation = 0.5*D*Ipk
Rms current equation = Ipk*sqrt(D/3) ~ 0.577*Ipk*sqrt(D)
You usually use the rms current when calculating the power in a resistor such as Presistor=Irms^2*Resistor. You can also use Irms and multiply by Vrms. But to use this equation, you need to use the phase angle between the Vrms and the Irma. That being said, I use a simple approximation based on the fact that most non rms digital current meters display the average current and we simply multifly be the average current to get an approximate power. Roberts
Nice sir
Could you tell me how to calculate output capacitor of flyback
Hi Nitish, look for Analysis and Design of a Flyback video 21, output catacitors
@@RobertBolanos thank you sir
Yup, my audio is noisy. Got to fix that.
RB
How can we calculate primary secondary resistance inductance of isolated transformer
This is a little to be asking this question. However, you first need to calculate the number of turns of your winding and also select a core that will satisfy your specifications. The core datasheet will have the MLT (mean length per turn). You multiply the turns*MLT to get the extimated length of the winding. You then use Resistance=(1.72*10^-8 ohms*Meter*(length_in_meter/Area_of _wire)). Please watch the transformer videos. Go to my channel and you should be able to see the videos that I have produce to help you design a flyback. Hope this helps. Robert
@@RobertBolanos yes I saw and downloaded
Please I have not seen the link to all your fly back converter design tutorial
Tite Man please click on my name and it should take you to my RUclips channel and you should be Abe to see my playlist. You have to look for it at this tie.
hey can any one tell me that how can we choose a switching frequency ?
Hi Usman, Typical switching frequency is from 100KHz to 200KHz. As you go up in frequency, the physical dimensions of you inductor and capacitors get smaller while the switching losses increase. For example, the switching losses for capacitor are calculated by Ploss=Cap*Voltage^2*Freguency. So as the frequency is double, then the switching losses also double. So operating from 100KHz to 200KHz is a good compromise. I hope this helps.
Please send me the link for all your videos tutorial i can not find them online
Tite Man, you need to click on my name next to my picture and it should take you to my youtube channel. You then click on Videos and you should be able to see all my videos.
@@RobertBolanos thank you I have seen it . Did you do video on all the SMPS topologies
Can this be used as a step down buck converter, about 125 volts AC/DC input to 12 volts DC output?
Yes you can use it as a step down converter
@@RobertBolanos Thank You for replying back. I will watch all the video carefully.
It would be tremendously helpful if you can make a full lecture series of designing and building PCB planar transformers/magnetics. Something new and cool. I’m planning to design a cost effective multilayer PCB maker system, so I would be able to make planar transformers if only I knew the math and theory of planar transformers. Thanks.
@@ShopperPlug Hi Max, that would be a great subject to cover. Let me see what can do.
@@RobertBolanos Thank You 😊
Why does the average input current equals to power divided by the min input voltage? Shouldn't it be called maximum average input current? You r trying to do a worst case calculation for the inductor value right?
Hi le,
All power supplies are what are called constant power units. In other words, the input and output power relative remain the same. As the input voltage decreases, the input current will increase in order to keep the same power. as the voltage is decreased, the On my way! Will increase the Dutycycle to keep the same voltage . Assuming that the maximum Dutycycle is close to 50percent, the current will be at its maximum. We calculate the input average current because we know their is a relationship between the average current and the peak current. Once we know the peak current, we can then calculate the inductor go the flyback.
Robert
Robert Bolanos Robert, what I'm trying to say is why not use the max input voltage to calculate the input current. You used the min input voltage in this tutorial. Also, why not use the average input voltage to calculate the average input current? Thanks!
Kelvin Le Hi Le, If you use the max input voltage, and calculate the average current, then the value of the Lprimary that you calculate using the procedure will lead to a supply and will not regulate at Vinput minimum because the PWM will have reached it maximine rating
Sir can you please help me finding design for CO2 gas laser flyback transforme, I shall be very thankful to you.
You need to know the required voltage, current and power. Once you know the power, you can use the equations that I presented to calculate the primary inductance.
Please do you have video on buck converter design kindly send me the link
Tite Man , a forward converter is an isolated bulk. ruclips.net/video/GDE8j7g6s0U/видео.html
Where did the T_on term come from in i_p?
+Terry Price Can you be more specific please. Not sure I understand your question. RB
+Robert Bolanos Sorry for the late response: at ~21:26, you given an equation ofr i_p. where does this come from?
+Terry Price Hi Terry, I am expressing the current that goes thru the primary of the transformer and I labeld the current as ILp. When the mosfets is on, this is refered at Ton. So Ton means that the mosfets is turned on and it is appling Vin across the primary inductance of the flyback transformer. So the current during Ton is expressed as iLp =(Vin/Linductnc_primary)*Ton. Now keep in mind that Ton can also be expressed at Ton = Duty/Frequency = Duty*Tperiod. Both equations are equivalent. If you sub into the first equation then you get, iLP=(Vin/Linducntant_primary)*Duty*Tperiod.Hope this makes sense. Comments and question can be sent to rbola35618@aol.com Robert
Hello, thanks for the great video, can you please send me the text file you are using to explain
Need you email address
amineadse2019@gmail.com
Thank you so much
@@RobertBolanos thanks in advance
rbola35618@aol.com
@@RobertBolanos i have sent you an email sir.
where does 4 x Iinave = Ipeak? why 4? please gimme some explanation
Hi WBPM3Tesla, in video 2 I show that Iinave = 0.5*Duty*Ipeak. The 0.5 comes from the intergration of the current ramp that I show in the video. The 0.5 also represents the half the area of a square is the area of the ramp. Assuming a dutycycle of 0.5, then we get 0.5*0.5=0.25. We now have Inave=0.25*Ipeak. Solving for Ipeak, we get Ipeak=4*Inave. I hope this make sense
thank you very much for your response dat very helpfull for me. my paln to implementating this flyback converter for battery ballancing on lipo battery. 1 more question sir, 1/Tp. where do you get that? please the explanation. your video is very helpfull for my final project. thank you mr robert for your kind.
The 1/Tp comes from the switching frequency
"My your simulations always converge" .. My = May ?
High Hopes Thanks High Hopes!!!
at ~6: 00 -7:00 you state that Area x Ton = Ton x Vpeak . if (Ton x Vpeak =Area) then this state is mistake!!!
Hi haimt1464, yes it is a mistake. I guess I had a senior moment. The area during Ton is Area_left=(Ton/Tp)*Vpeak, while the Area during Toff is Area_right=Tp(1-D)*0V, so the right side egual goes to zero. The rest of the derivation should be correct.
Thank you very much for the quick response ,I am appreciate you knowledge and the willingness to share.
I am learning a lot from you ,keep the good work, and again thank you very much.
Hi haimt1464, Thank you spotting the error. I made an annotation on the video. If you into the 6:00 time frame, I made the annotation. Again, thank you for your feedback and spotting the error. Best regards, Robert
One question: If my Vin is a 24Vdc. How would I set my Vmin and Vmax as? (This is according to your design example). Thanks
Anh,
This is a good question. You want to calculate the voltage drop that you would have in the primary loop winding which could consist of the transformer resistance, the MOSFET Rdson, the input filter inductor. A rule of thumb that I use is to assume a 1 Volt voltage drop. So if you input voltage in 24 volts, then use 23 volts as you Vmin. Vmax is important to know because the input filter has filtering capacitors and you want to make sure you do not exceed their voltage rating. You can send questions to rbola35618@aol.com. Hope this helps
Yup, my audio is noisy. Got to fix that.
RB