http://www.onsemi.com/pub/Collateral/MC34262-D.PDF
http://www.youtube.com/watch?v=bX8IIjf15qY
This was supposed to be a small side piece in the Manson PSU Teardown, but it was almost 10 minutes, so I decided to separate it out to keep the teardown as short as possible. You should watch the PSU teardown to get the context of this where the chip is used.

And let's take a look at the Motor Roller slash ON Semiconductor sorry still keeps calling the Motorola every time I see the old Motor Roller prefix of MC but the MC 34264 uh 262 sorry power factor uh controller it's just an 8 Pin dip package comes Ino as well. It's got a Uh 2% internal band Gap reference, zeroc current detectors, uh, quadrant multipliers, and uh, various other stuff. So let's take a look at the Uh internal simplified block diagram here. as you can see not a huge amount of Uh circuitry in there.

There's an internal Uh band Gap voltage reference that's an over voltage comparison comparator which comes from the Uh voltage feedback which goes into an error amplifier. There's a compensation pin pin uh, the input from the multiplier is here. there's the output uh driver and there's a current sense uh input as well with the zero current detect input because it has to uh s it has to uh know when there's zero current and we'll take a look at that and there's a Power Pin and that's pretty much it. There's not much else, but we'll see if we scroll down here and we go to typical configurations we can actually have a look at.

um, it's well worth uh reading this. it's got introdu and how a power factor correction system actually works and we'll have a look at what happens if you don't have a power factor correction circuit. uh, you've got the basic Uh AC uh Mains coming in here full wave Bridge rectifier as we saw in the schematic. and if the power factor correction Uh, circuitry wasn't there, you got the bulk storage capacitor.

So Fullwave Bridge rectifier straight into the Um storage capacitor generating a high voltage DC which then goes into your Uh DC Todc converter which then Powers your load. Now the problem with that is that looking at the waveforms here, you can see the AC Uh line waveform here, which is the dashed, the outer dashed one there. that's the ideal, the 50 HZ or 60 HZ Mains input. and because uh, you, the energy is stored in the bulk capacitors here.

the voltage Peak you're only going to draw um, a current Spike So this waveform here this spiky one here is the current uh drawn during the peak period of the waveform and that's quite. It's going to be a large current draw, so you're actually going to drag down the line voltage the Main's voltage just a little bit and it's going to Sag as they say as they call there they call that line sag and so you're going to get these current pulses at the positive and negative Peaks at the main input and that's not very good at all for Uh for getting a power factor correction of one. And in fact, they tell you here that this configuration can give you uh, commonly a power factor ratio of 0.5 to7 and that's not great at all now. I Don't think this is the place to actually get into a detailed discussion on uh, how a Power Factor Correction Preon converter Works a current mode one like we've got here.

If you want to, uh, get the details of this, by all means, download the data sheet. it'll be linked in um in the Uh notes there. So, but this is basically um, what we've got. We've added a PFC Power Factor Correction Preon converter between the bridge rectifier and the bulk storage capacitor we had before, and the rest of you can DC to DC converters all the same.
And of course, you got the high frequency bypass capacitor in here. and you would have noted that on the Uh on the schematic for this um, Uh model as well. But we're looking at the MC 34362 Power Factor Correction Preon converter and there's basically series inductor here with a mosfet uh, pulling that down to ground and an output diode and you'll note that that is, basically um, a boost converter. That's basically that's exactly the same configuration um, as you get for a boost DC Todc converter.

but the whole basic concept of it is it's going to uh, switch the mosfet down here so that you get a half sinus soidal, um, average, uh current. basically. uh, drawn from your inductor current. So that's a series induct series current through your inductor.

You're going to have these Peak values, but they're going to average out and smooth out, um, over your AC Mains or half of your AC Mains waveform like that. So instead of having just one big spike here like you would get without uh, this power factor correction circuit in here, you get multiple spikes spread out over half of the sinusoidal waveform and that's and that averages out and then include and that improves your power factor. Uh, correction. And if you take a look at the rest of it here, it tells you all about the different functional block diagrams inside the device.

and I Highly recommend. Uh you. It makes for some great bedtime reading. Got the error amplifier, the multiplier which is pretty much the key to how it all works over voltage, comparative, zero current detectors, current sets, compara latch how it all works, and uh, it's great stuff.

I Highly recommend it And bingo down here. We've got our design equations and uh, there's a whole bunch of them down here and well, you might have, um, seen, uh, me, go through similar equations like this before for right? DC to DC converter uh devices And the problem. One of the problems with uh, this is that, uh, let's say you're trying to calculate the required uh, well, you know you want it. Wants to know the required converter output power? Well, what is that? Because this is a generic bench power supply, the the user could be taken anywhere from zero output power right up to the maximum 600 watts or so.

So um, really, uh, these values are going to be all over the shop and you have to um design and do various tradeoffs for various Um output, uh, powers and things like that. So um, maybe you would take it, you know, at the maximum output power, but you'd have to go through the formulas to see how much uh, your power factor and other things traded off at lower Uh output. Powers So um, you would have to go through the formulas. Be my guest.
Great stuff to do at bed time. I Love it! And uh, how? the internal block diagram and the basic Um application circuit is down here. there's the mosfet. there's the Uh Transformer exactly how we see it there.

and uh, here's a typical table of some power factor controller test data and we've got power factors of 0.999 down to 0.996 so it's going to be pretty close to one. So these things work pretty well. and if you take a brief look at the application circuit here, you can see the main inductor T here. but it's actually a Transformer because it's got a secondary winding here that goes down to ground.

and that uh Taps effectively allows you to tap off the Uh output, read the output current, and that's exactly what happens on Pin five here, which detects uh, zero output current. Pin Eight is the VCC Uh pin, so it just filters that there and Powers the actual device and Pin three down here is our multiplier. uh input and you'll notice the uh, it's just a voltage divider here from the directly from the uh, uh, I ified Main's input. So that's reading.

That's going to be reading the the harmonics off there that come from the full wave Bridge rectifier and that goes in to the multiplier and the rest of the device. And there's our output driver which drives the mosfet which pulls that down to ground once again through a current sense resistor there. So um, that's basically all there is to it and we see that uh got some typical waveforms here, which is quite nice and uh, if you want to, uh, take a look. Typical test things: error amplifiers, error output comp Ah, current waveform Spike suppression all sorts of things.

Highly recommend you. Uh, download this. Oh, look at the old school layout. look at the old school taped layout there I Love that brilliant gez that' be home anywhere in the 1970s, but it can't really be that old.

Nobody care cared about power factor correction back in the 70s did they? except for huge, uh, maybe big, uh, industrial stuff or things like that? Certainly not at the Uh bench. uh, Power Supply Lab power Supply level I Don't think anyway? Um I Highly recommend you. Uh, download this data sheet and uh, uh, you know, print it out and uh, have some good bedtime reading. It's most interesting.


Avatar photo

By YTB

19 thoughts on “Eevblog #273 – power factor correction with the mc34262”
  1. Avataaar/Circle Created with python_avatars Thomas Maughan says:

    Yet another video that fails to reveal the magic smoke.
    Problem statement: A simple rectifier into capacitor power supply draws current only when the incoming sine wave happens to have a higher voltage (plus 2 diode drops) than the capacitor. At that moment, the capacitor starts charging and its effective resistance nearly zero thus current demand is very high, but only for a few milliseconds.

    Solution statement: Find a way to spread out that spike that happens 120 times per second; find some way to extract energy from the entire sine wave not just the peaks.

    Complicating factor: The filter (bulk) capacitor already has 300 volts on it. At the moment the incoming sine wave is 20 volts. How are you going to step up that 20 volts to 300 so it will impart some energy to the capacitor? You cannot just use a 15 to 1 step up transformer; if you do that, watch the smoke happen at the peak of incoming voltage!

    The magic is a FLYBACK regulator; known here as a boost regulator but it depends on the flyback effect of the inductor. This is the same phenomenon that generates the spark in your gasoline powered internal combustion engine.

    Inductor charging half-cycle:

    For a few microseconds, the inductor is shorted to ground which seems like a bad thing but while shorted to ground it is building a magnetic field. Then you suddenly open that path to ground.

    Capacitor charging half cycle:

    The magnetic field starts to collapse. It wants to push electric current in the same direction that it has been flowing. What is special is that it does not care about voltage; it will push X number of electrons up a very steep hill if necessary. So, even though the capacitor already has 300 volts on it, and you charged the inductor at a moment where the incoming sine wave had only 20 volts, when you interrupt the ground path, the inductor will push 20 volts no matter what; making a spark if it needs to. It pushes that 20 volts right onto the 300 volts already on the capacitor and THAT is how you glean energy from the 20 volt (or any volt) portion of the incoming sine wave and scoop it up on top of your storage capacitor.

    Detailed nuance: We don't actually fully charge the inductor, that would be bad, and we don't completely dump the charge either. So we "top off" the capacitor with these little impulses and in that manner spread out the consumption of CURRENT from the source so that the current approximately follows in shape and phase with the source voltage.

    It is by its very nature current limited since the thing that charges the main storage capacitor is the FLYBACK impulses from the inductor.

    You doubtless notice an interesting phenomenon; at the source peak voltage, the duty cycle on the inductor pulse width modulation is LEAST. That's because it only takes a few microseconds of high voltage to "charge" the inductor; but as the voltage goes down following the sine wave, the pulse width will expand in order to maintain the magnetic field charge on the inductor. It will expand across all cycles as the load increases.

  2. Avataaar/Circle Created with python_avatars James Hancock says:

    Just discovered this. I’ve also read some abstracts on using synchronous rectification and using that to control pf too in a more simplified design. Have you done a video on that? Couldn’t find anything.

  3. Avataaar/Circle Created with python_avatars Datsit says:

    Hi Dave
    I sometimes notice power factor is low at low power consumption, why this happens?

  4. Avataaar/Circle Created with python_avatars Avin Raikwar says:

    I was wondering what will happen if we replace the AC input with DC one , for e.g instead of 240v AC we supply 345v DC , how would this circuit will response ?

  5. Avataaar/Circle Created with python_avatars wiedapp says:

    I think you did right to separate this part from the other video.
    Whoever searches for PFC will find it (like me) and get a huge chunk of pointers on how to design their own DC PSU circuits.
    As an electrician in an industrial company in Germany I know how dirty a power grid can get if machine suppliers are not careful, waveform wise that is.
    The choice of the type of VFD can make a huge difference in that regard.

  6. Avataaar/Circle Created with python_avatars Samuel Aditya says:

    What happen to spinning power meter if the load are just half wave rectifier with capacitor power supply. Will the disc still spinning since the current form is just on one side or DC.

  7. Avataaar/Circle Created with python_avatars Robert Lake says:

    Verify your electrical circuits on the go! Track down: androidcircuitsolver/app.html

  8. Avataaar/Circle Created with python_avatars Lokesh Naik Lokesh says:

    Sir can I get the MATLAB simulation of this pfc…

  9. Avataaar/Circle Created with python_avatars Colton Baldridge says:

    Just came here from the LED flickering video, and considering it's 1AM in the states, this is gonna be my bedtime reading for tonight 🙂

  10. Avataaar/Circle Created with python_avatars Matthias H. says:

    Thank you so much for this interesting video

  11. Avataaar/Circle Created with python_avatars Destructo says:

    i would love if you could somehow plot the input volt and amp curves on a device with a poor power factor. (seeing that sag) and perhaps multi channel to see the difference the chip actually made, side by side.. or do a video on different methods of active PFC, how they work, etc.

  12. Avataaar/Circle Created with python_avatars sys admin says:

    I thought I might be a little bit nerdy but this night time topic puts you way above me 😀

  13. Avataaar/Circle Created with python_avatars ncrdisabled Submarine vet says:

    I worked for KW Controls which had UPS system from .5 kva to 1000kva They had TUPPS system transistor uninarutable power system . I was trained to work on all the systm including QA testing which includes power factors on all the systems . This was the late 80s I worked there for 6 years . They also had Piller systems

  14. Avataaar/Circle Created with python_avatars lez briddon says:

    bridge rectifier into caps is nice and efficient, but not for the power company, and maybe also for their consumption metering / billing systems…

    I'll stick with the old ways unless theres a real need for anything else that improves the equipment in use.

  15. Avataaar/Circle Created with python_avatars Fir3Chi3f says:

    I love your idea of bedtime reading Dave!

  16. Avataaar/Circle Created with python_avatars deathventure says:

    Motorola is a multi-department corporation. Each department takes a different sector of manufacturing. Motorola spun off two companies from their semi-conductor branch, one being ON Semiconductor, and the other being Freescale. The reason for spinning off these other companies can vary, but it usually ends up being cheaper for them.

  17. Avataaar/Circle Created with python_avatars legogunguy001 says:

    OK thanks.

  18. Avataaar/Circle Created with python_avatars CampKohler says:

    If you listen carefully, most professional announcers have a big bass component and a matching treble component. The two sound great together, but if you put it through a lo-fi path (POTS line), these roll off at both ends and you can barely recognize the speaker. Dave's probably has a big mid-to-high-freq peak, so he's tailored for those 99-cent earbuds. Just the thing for iPod listeners and utilitarian comm channels. 🙂

  19. Avataaar/Circle Created with python_avatars CampKohler says:

    Money is the reason. With a big transformer, you need uh, a big transformer. That's lots of iron and copper. The DC-DC starts by generating AC at 100KHz or whatever, which allows little transformers to be used. It's also smaller and lighter, allowing more product to be jammed into a container, which is important because it all comes from China. If you are going to have regulation ckts anyway, might as well have them doing the DC-DC thing at the same time.

Leave a Reply

Your email address will not be published. Required fields are marked *