Continue to Site

Welcome to EDAboard.com

Welcome to our site! EDAboard.com is an international Electronics Discussion Forum focused on EDA software, circuits, schematics, books, theory, papers, asic, pld, 8051, DSP, Network, RF, Analog Design, PCB, Service Manuals... and a whole lot more! To participate you need to register. Registration is free. Click here to register now.

Off-line Power factor correction

Status
Not open for further replies.

Winsu

Full Member level 3
Joined
Oct 25, 2017
Messages
174
Helped
2
Reputation
4
Reaction score
4
Trophy points
18
Activity points
1,691
I know that to achieve an ideal power factor correction power converter, the converter needs to operate in DCM during the whole line half period. I don't know why it is like that. Is it something to do with the cancelation of magnetic fields in the inductor?. If it was like that, the average current in the inductor in CCM is also cero. Anyone could give me a hint please?.

Thanks,
Winsu
 

it is because of the gain difference as Vin goes up between CCM and DCM ...
 

For example for a buck boost the duty cycle needs to meet this criteria to operate in DCM Vout > (D/(1-D)) * Vin. If D meets this criteria the converter is in DCM and therefore it is an ideal power factor corrector converter. This is the theory but I don't understand the logic behind that, I can't see the reason. Thanks.
 

I agree with Easy peasy.

The transfer function changes between DCM and CCM, so it is much more difficult to make the regulator loop if both modes are allowed.
I think it could be possible with a fully digital regulator that can detect the change between DCM and CCM, and quickly change the regulator equations.
Remember that we want minimal distortion in the current, so a "normal" regulator is probably not good enough.

To be in DCM, the Vs (Volt*second) when taking energy from the inductor must be higher than the Vs when putting energy into it.
D = switch on, so Vin is applied to the inductor
The rest of the time is (1-D), where Vout is applied to the inductor until the current reaches zero.
This means that Vout * (1-D) must be greater than Vin * D if the mode must be DCM.
Vout * (1-D) > Vin * D
Divide by (1-D) on both sides:
Vout > Vin * D / (1-D)
 
Last edited:

you can have a fixed PWM in DCM on a converter and get somewhat good power factor - but if you try to do this in CCM ( cont cond mode ) the gain rises more than proportionally with the rising Vin and you get very peaky and undesirable input waveform...

- - - Updated - - -

The L6562 chip has BCM and other improvements for good power factor ... study the app notes etc ...
 

Thanks for your answers. I am getting a better insight into this, but I haven't got the answer that I was looking for yet. A perfect power factor is when the real power equals the apparent power or in other words the reactive power equals cero. Why it is better achieved when the converter is working in DCM rather than in CCM?.
 

I believe the question has been answered by several contributors, but you probably didn't understand the answer.

It seems necessary that you read more about PFC operation to understand how it basically works.

Specifically, a PFC designed to operate in CCM will necessarily transit to DCM at low instantaneous input voltage and also at light load. This brings up the problem of jumping modulator gain and difficulties to maintain a sinusoidal input current. In contrast, if it's designed to work in DCM at rated load and peak input voltage, it will always operate in DCM.
 

Hi All,

Thanks for your answers. I agree that I definitely need to learn more about PFC. I think the main target of a PFC is make the converter function as a pure resistive load, then the power factor is optimal . In order to do that the current average at the input and the voltage at the input stage must be proportional. By operating in DCM the inductor current is zero before the switching converter period is finished therefore we ensure that the average current is proportional to the voltage input. If inductor current never is zero ( CCM) the gain rises further proportionality and the power factor becomes poor. In other words, if the inductor never demagnetizes completely the average current won't be in phase with the input voltage input. I guess this is the reason...
 

I'd just understand it as a general problem where the system has a step change in its performance on the boundary between CCM and DCM.

Although feedback loops theoretically solve this in real life any step change in the system will 'get through' the feedback loop to some extent. If you avoid the step change staying in DCM all the time you eliminate the underlying problem.

The fact is PFC's don't switch very fast and/or their low cost controllers don't have terribly high performance feedback loops.

As mentioned a digital controller could be designed to solve this problem pretty easily.
 

Thanks asdf44. I am not facing any practical problem that I need to resolve, it is purely educative . I came across with a book that explains off line PCF LED drivers. It specifies that to achieve an ideal off line PFC LED driver the converter needs to be working in DCM mode. I understand now ( or at least that is what I think ) that a converter that transits from CCM to DCM won't have a good power factor but I am not 100% sure about the reasons.... I guess the reason is because as voltage and current are not proportional ( and not in phase) when CCM it causes that the power factor gets poor.
 

Sorry but you are fixated on “DCM is needed for PFC”…in real terms you can get adequate pfc IN bcm, or ccm, or a combo of dcm and ccm…..take an average mode pfc like using LT1248….that can transition from dcm to ccm and its still perfectly ok power factor.
Here is my SMPS course…….(you can find it in the following link)….there is a section on PFC…there is an average mode pfc there that goes from dcm to ccm and the power factor is fine.

If you use a “BCM PFC controller”, then that is always right on the verge of ccm/dcm, and if you average the inductor current, then it’s a sine in phase with voltage…so yes , nice pfc but not the only way to do it.
Ill try and dig out the average mode pfc sim for you….but maybe you will find it first…its in the free LTspice.

the link is somewhere in this...

https://massey276.wixsite.com/electronicsdegree
 

If you want to control a converter in CCM to draw a sine wave of current from the mains - then the control is more complex and you need to sense the current - DCM avoids all this as the gain with Vin is constant ...
 
  • Like
Reactions: treez

    T

    Points: 2
    Helpful Answer Positive Rating
sorry here is the link to the smps course with the section on pfc...

https://drive.google.com/open?id=0B7aRNbu3Fes4TU92Mkw3YlA3ams

...it shows annotated ltspice sims of many types of pfc's...which you can run and evaluate.

ltspice is free down load from analog.com.

.....the non smoothed , post rect dc bus gives you a voltage half sine train, which you can divide down and use as the reference in your current error amplifier, so that your current is sinusoidal and IPW the voltage. There is a sim like this in there.
At low powers, pfc is often done in dcm as it avoids the dreaded reverse recovery problem, but thats not to do with the power factor

- - - Updated - - -

please do tell me if any of the sims dont work, and ill tweak them and send them back to you.
 

Hi Treez,

Thank you very much for the course it looks very complete. I will try to get back to you when I have had the opportunity to work a bit on that, thanks again.
 

Status
Not open for further replies.

Similar threads

Part and Inventory Search

Welcome to EDABoard.com

Sponsor

Back
Top