0
\$\begingroup\$

I m studying a flyback converter. The flyback converter is already made. I m just trying to understand how it works. I m actually focusing on the flyback transformer. And in order to better understand how it is designed I need to know if it is working in Continuous Current Mode or in Discontinuous Current Mode. It is a flyback transformer with fractionnal turns. Here is the formula that I used for knowing if it working in CCM or DCM, the formula is not the best ... It is obtained by doing the approximation that the the transformer is ideal... If you have a better formula it would be really nice to hear it !

enter image description here

In the above formula I know each of the parameters, except Vs which is the output voltage of a transformer with one input and one output. Ve is the rectified input from a RMS 230V.

My transformer has several outputs. How can I know Vs ? What I have done : I estimated the power and the current of each output. Then I sum all of this for having the total output power and the total output current for having the "average" output voltage. What do you think about this method ? Do you have any advice ?

The result is a current which is very low (about 30 mA) where as the ouput power is about 40 W and the total current drawn is around 2A5. The primary inductance is about 1.5 mH. The frequency is around 75 KHz. I am relatively surprised that the limit between CCM and DCM is so low ...

Here is what I called a fractionnal transformer (the windings are fractionned in several outputs) (This is not the transformer that I m studying) :

enter image description here

Thank you very much :)

\$\endgroup\$
8
  • \$\begingroup\$ If you want to know if it's in CCM or DCM, measure the current through the transformer with a current probe! Of course that's not the best (or cheapest, considering the price of those probes) method, but it's certainly the most straightforward. \$\endgroup\$ – Hearth Mar 29 '20 at 15:01
  • \$\begingroup\$ You can't have fractional turns. They are whole numbers even if they don't look like whole turns physically. You haven't explained all the terms in the formula nor where you got it from. \$\endgroup\$ – Andy aka Mar 29 '20 at 15:02
  • 1
    \$\begingroup\$ That's called a multi-tap transformer. \$\endgroup\$ – Andy aka Mar 29 '20 at 15:12
  • \$\begingroup\$ I cannot do any measure... Alpha is the duty cycle and T is the period time. It is coming from here : fr.wikipedia.org/wiki/Convertisseur_Flyback it is in french ! good luck ;) \$\endgroup\$ – Jess Mar 29 '20 at 15:14
  • 1
    \$\begingroup\$ Does this help you find a better formula for DCM and CCM? Formulas lower down the page. \$\endgroup\$ – Andy aka Mar 29 '20 at 15:18
2
\$\begingroup\$

The easiest way is to look at the MOSFET drain waveform, \$v_{DS}(t)\$. If this is a high-voltage design, you can even avoid clipping the probe ground and simply carefully approach the probe tip close enough to the transformer and "see" the waveform by radiation. If you clip the oscilloscope ground and connect the probe tip to the MOSFET drain, make sure you have inserted an isolation transformer or supply the converter with an isolated dc supply. Observing the waveforms, reveals several cases:

  1. in discontinuous conduction mode or DCM, the transformer demagnetizes cycle by cycle. It means that at some point, the secondary-side diode stops conducting and the reflected voltage - the flyback voltage - disappears in the primary side. The capacitance lumped at the drain is thus charged at the reflected output voltage scaled by the transformer turns ratio plus the input voltage and needs to discharge to return to the input voltage. The energy stored in this capacitor is now oscillating back and forth between the capacitor and the primary inductance. This oscillation is typical of a DCM-operated flyback converter regardless of its control mode. The below simulation shows a typical shot:

enter image description here

The sense voltage (or the primary switch current) starts from zero and ramps up to the peak value, \$I_p\$.

  1. in continuous conduction mode (CCM), the transformer does not demagnetize cycle by cycle and some energy remains stored in the core. The secondary-side diode conducts during the whole off-time period and the reflected voltage does not disappear. The primary current no longer starts from zero but from some pedestal called the valley current \$I_v\$. The below shot shows the typical waveforms observed in CCM. The square drain-source waveform is typical of the CCM operation for the flyback but also for many other converters (as the DCM ringing):

enter image description here

  1. and finally, there is a third possible case which is also DCM but optimized in such way that the switch always turns on in the minimum of the drain-source voltage. It minimizes turn-on switching losses and can even bring them down to zero with true zero-voltage switching (ZVS). It also limits EMI radiations by reducing or even cancelling the brutal discharge of the lumped capacitance. This mode can either be a specific operating point at a certain load or input voltage or it can be the operating strategy: if you see the switching frequency changing with the input voltage or load but still observe the MOSFET switching right in valley switching, then this is a quasi-square-wave resonant converter, often named quasi-resonant converter or QR. Other names include borderline or boundary conduction mode (BCM). The below shot is the typical signature of this converter:

enter image description here

There are several reasons why operating a flyback in DCM or CCM but usually, if I take the example of a notebook adapter, the converter is designed to operated in CCM at low line and nominal power where conduction losses are smaller and the ac ripple in the output capacitor is minimized while DCM is naturally entered in high input voltage conditions. For a high-voltage power supply, DCM or BCM is usually adopted because the secondary-diode spontaneously turns off, without losses on the primary-side switch and you can use a lazy high-voltage diode for the rectifying job.

\$\endgroup\$
3
  • \$\begingroup\$ Thank you for your explanation ! It was really interesting ;) What I conclude is it is difficult to estimate theoritically if the transformer is working in CCM or DCM. Nevertheless, when designing a transformer it is important to know if this one will work essentially in CCM or only in DCM. The design won't be the same .... \$\endgroup\$ – Jess Mar 30 '20 at 9:20
  • 1
    \$\begingroup\$ With pleasure. You can easily determine if the transformer operates in CCM or DCM: you determine what is called the critical load or input voltage for which the converter operates in boundary condition. Below this value the converter operates in DCM, above it, in CCM. This is all detailed in the green book I wrote amazon.com/Switch-Mode-Power-Supplies-Second-Edition/dp/… \$\endgroup\$ – Verbal Kint Mar 30 '20 at 9:33
  • \$\begingroup\$ Thank you ! I asked the book to my manager. \$\endgroup\$ – Jess Mar 30 '20 at 10:02
1
\$\begingroup\$

How to know if a flyback converter is working in CCM or DCM

Comparing the two transfer functions: -

enter image description here

So, if you can get your hands on an oscilloscope and look at the voltages to determine the duty cycle (D) and the operating frequency (\$F_{SW}\$) then, knowing what the turns ratio (N) is you can see which TF fits the bill more.

Alternatively, you might be able to tell if there is a flat line section in the secondary voltage. If so, it tells you that the device is operating in DCM due to the "hold" period present in the DCM switching cycle. The picture below is for a 1:1 flyback converter hence, primary and secondary currents have the same scale: -

enter image description here

enter image description here

\$\endgroup\$
5
  • \$\begingroup\$ Thank you for your explanation. What I conclude is it is difficult to estimate theoritically if the transformer is working in CCM or DCM. Nevertheless, when designing a transformer it is important to know if this one will work essentially in CCM or only in DCM. The design won't be the same .... \$\endgroup\$ – Jess Mar 30 '20 at 9:21
  • 1
    \$\begingroup\$ @Jess A lot of on-line articles can imply that a converter operates in either CCM or DCM but it's quite often that they operate in both and smoothly transition. In CCM, if the load becomes very light (or Vin becomes high) it will inevitably have to drop into DCM operation (unless it uses a synchronous rectifier on the secondary). And, virtually any DCM will go through a stage of CCM as it powers through dozens of cycles immediately following a power up event. The design has to be the same!! \$\endgroup\$ – Andy aka Mar 30 '20 at 9:29
  • \$\begingroup\$ Thank you for this comment ! I have no experiencies ... I read from this document (ti.com/lit/ml/slup127/slup127.pdf) that the design won't be the same according to the operating mode of the flyback. Indeed according to the operating mode, the losses implied into the design are not exactly the same. The repartition of the losses is different. In CCM the flux swing is lower than in DCM so the core losses are lower, etc ... In any case you re right ! \$\endgroup\$ – Jess Mar 30 '20 at 10:08
  • 1
    \$\begingroup\$ @Jess it's not the flux swing that determines core losses, it's the absolute peak value so, although in CCM the change in flux will be smaller, it might still have a high peak value and that will cause core losses just the same as in DCM. \$\endgroup\$ – Andy aka Mar 30 '20 at 11:03
  • \$\begingroup\$ I agree with you :) Thank you for the remind :) \$\endgroup\$ – Jess Mar 30 '20 at 12:21

Your Answer

By clicking “Post Your Answer”, you agree to our terms of service, privacy policy and cookie policy

Not the answer you're looking for? Browse other questions tagged or ask your own question.