Great video Dr. Ben-Yaakov! Thank you for putting this together. I was looking for a simple, straightforward discussion of the pros and cons related to split resonant capacitor and your video did an excellent job of providing a straightforward description. Thanks!
Thank you for the great video. The only thing I have to mention is that actually, you do not need any additional DC link bypass capacitance when you use split capacitors C1 and C2. You can put even big inductance between C1/C2 and Vin and there will be no problem here, the charge balance will be satisfied for C1/C2 (average current is zero). Might be that RMS is a little bit higher for C1/C2 but still no problem here. You can easily simulate this but also practically ALL HB converters I faced did not have additional big DC link capacitors with split C1/C2 configurations, only some ceramic for EMC. Thank you for ideas.
@@sambenyaakov I also see only two bulk capacitors in a typical ATX power supply. Perhaps with more stringent power factor controls in place, we will see more cases where a capacitor straight across the bus is found.
I recently opened up an EVGA 650W PSU that had a half bridge and two DC blocking capacitors but strangely enough they were not in the split configuration I was expecting. One capacitor went directly from the transformer to the negative side and the other also went from the transformer to negative but with a current transformer in between - I assume so they could use a much smaller current transformer. So far, it's the largest PSU I have seen use a half bridge - any bigger and they go with a full bridge design.
DC by capacitor rescues the DC source from draining continuously. During each charge cycle, polarisation current of capacitor cT only flows.Therefore CT capacitor improves inverter efficiency.
Thank you for the great video @sambenyaakov . where can i find the next riddle mentioned in the vedio difference between Series and paralle Resonant converter riddle? Plz share the link below.
Is the method of unifying the alternating current that we get from the secondary coil always the same in every method that you use to activate the primary coil, or can you change it as you wish?
In my opinion in the second circuit: when Q2 is on, there is no reverse current in the primary of the transformer. You have to add a capacitor. Thank you professor for this excellent analysis.
Hi first of all want to thank you for your amazing explanation. My question is: you are saying that in the case of LLC, we don't need a bulk capacitor since the Cr doing the Job. when I'm looking at the TI application of UCC25800-Q1 they use Bulck and Cr but they put TWO 1/2Cr on the secondary side. would like to hear your opinion about it. Thanks you anyway 🙏
Thanks for note. Where (minute in video or slide number) do I say that there is no need for a bulk cap? As for C/2 see th-cam.com/video/T8qWWwsMOK8/w-d-xo.html
Sir, i am very greatfull to listen to your videos. Diodes in bottom circuit will see serious voltage spike at turn off as they are chopping inductor current. So snubber on diodes is mandatory.
@@sambenyaakov Sir, at 4:33, a current flowing into Q1 splits into C1 and C2. It is obvious that the portion of current in C2 will flow back to the negative terminal of the source. But What happens to a portion of current entering into C1? How it will reach a negative terminal of source? It is very uncomfortable to digest sir.
I saw recently this design, with what you sad in Mind, would you say the connection of the middle of the Buss(in red) is not optimal, they have (in Blue) Capacitors across to manage those currents but i would not have connected the middle of the 0.1 Capacitors to the Main Bulk Capacitors. Or do i see this wrong?
This looks OK. The blue caps to suppress spikes across bus and the red for steering the ac current. It is desirable to put ceramic caps in parallel to electrolytic for the high frequency components. O.1u seems to me too low though.
would this not be a problem as you mentioned around 10:54, without a Cb does the ripple current not become a problem. And yes 0.1µF is not much the electrolytic will probably see quite the current.
hello. I have 2half bridge on IR2153 in my DC/DC power. And I dont understand that these Voltage is not dividade by 2. When IR2153 start working one capacitor has 0.9 Uin second 0.1Uin
Good point. But...Lf is a current source. When reflected to transformer primary it will (by superposition ) cause the currents indicated by the arrows.
I think you're missing resistors that are typically in parallel with each of the "voltage divider" capacitors. Typically there is a resistor (~100k Ohms) in parallel with each of the caps. This creates a voltage divider with Vin/2 in the middle and Vin/2 across each cap. When each of the FET's turn on, the transformer is hit with a constant voltage from the caps which creates the current ramp waveform through the transformer. Using the inductor equation V = L di/dt and solving for the constant inductor voltage you get an integral of a constant over time which is a ramp for the inductor current. So typically there are resistors in parallel with the capacitors to set the voltage.
Well, you are mistaken. In fact,as pointed out in video, there is no need for two capacitor. One is OK (so where would you put the resistor "voltage divider"). Try to watch the video end to end with open mind. If you then have a question, I will be happy to discuss.
@@sambenyaakov I'm referring to R1 and R2 in this image: www.seekic.com/uploadfile/ic-circuit/201162412958301.jpg. The resistor divider is also mentioned in the answer to this stackexchange post: electronics.stackexchange.com/questions/259406/dc-dc-half-bridge-converter-capacitor-selection. With the two capacitors the duty cycles need to be almost exactly the same for the two transistors. Reference: www.runonielsen.dk/Half_bridge_control.pdf.
@@sambenyaakov Here is the quote from Runo's Power Design: "In the half bridge the two switches must run with the same pulse width, so that the midpoint between capacitors C1 and C2 will be at half of the input DC voltage."
@@sambenyaakov You also might want to explain, in the resonant LLC, that the Lr Cr are important because they create a filter that filters out the higher order harmonics, leaving behind the fundamental harmonic from the drive pulses. The resulting (filtered) waveform is a sine wave.
![[TH] 2024 PMWC x EWC Group Stage Day 2 | PUBG MOBILE WORLD CUP x ESPORTS WORLD CUP](http://i.ytimg.com/vi/vwIQyCmNIBU/mqdefault.jpg)
Great video Dr. Ben-Yaakov! Thank you for putting this together. I was looking for a simple, straightforward discussion of the pros and cons related to split resonant capacitor and your video did an excellent job of providing a straightforward description. Thanks!
Thanks Cody .
and again your video is the best explanation of how half bridge converters work!
🙏🙂
Great tutorial and very helpful!
Thank you professor Sam!
😊
Incredible videos, very interesting. Greetings from Saint-Petersburg, RU. Keep up the cool channel.
😊
Thank you for the great video. The only thing I have to mention is that actually, you do not need any additional DC link bypass capacitance when you use split capacitors C1 and C2. You can put even big inductance between C1/C2 and Vin and there will be no problem here, the charge balance will be satisfied for C1/C2 (average current is zero). Might be that RMS is a little bit higher for C1/C2 but still no problem here. You can easily simulate this but also practically ALL HB converters I faced did not have additional big DC link capacitors with split C1/C2 configurations, only some ceramic for EMC. Thank you for ideas.
Thanks for participating. You do need a capcitor since the two caps will never be exactly equall.
@@sambenyaakov I also see only two bulk capacitors in a typical ATX power supply. Perhaps with more stringent power factor controls in place, we will see more cases where a capacitor straight across the bus is found.
@@peterdouglass6058 👍
Your videos are awesome. I hope learn as much as I can.
😊
Thanks.
Very helpful professor. thank you so much. We are all your students from all over the world.
😊🙏
I recently opened up an EVGA 650W PSU that had a half bridge and two DC blocking capacitors but strangely enough they were not in the split configuration I was expecting. One capacitor went directly from the transformer to the negative side and the other also went from the transformer to negative but with a current transformer in between - I assume so they could use a much smaller current transformer. So far, it's the largest PSU I have seen use a half bridge - any bigger and they go with a full bridge design.
Interesting . Thanks for sharing.
Bottom dc cap missing
Thanks again for each of your video, they are really useful
👍 Thanks.
Thanks
Great video Dr. Sam.
Thanks Swaraj
DC by capacitor rescues the DC source from draining continuously. During each charge cycle, polarisation current of capacitor cT only flows.Therefore CT capacitor improves inverter efficiency.
I am sorry but I do not understand what is this has to do with this video.
Thank you for the great video @sambenyaakov . where can i find the next riddle mentioned in the vedio difference between Series and paralle Resonant converter riddle? Plz share the link below.
Will tr to locate😊, no promise.
Is the method of unifying the alternating current that we get from the secondary coil always the same in every method that you use to activate the primary coil, or can you change it as you wish?
In my opinion in the second circuit: when Q2 is on, there is no reverse current in the primary of the transformer. You have to add a capacitor. Thank you professor for this excellent analysis.
To which minute are you referring to?
Hello Professor,
I answered your question: "What's wrong with one of these circuits", maybe I'm wrong! It was the 18th minute of your video.
👍
thanks for the video Dr
thanks
Nicely Explaination
😊
Hi first of all want to thank you for your amazing explanation.
My question is: you are saying that in the case of LLC, we don't need a bulk capacitor since the Cr doing the Job. when I'm looking at the TI application of UCC25800-Q1 they use Bulck and Cr but they put TWO 1/2Cr on the secondary side.
would like to hear your opinion about it.
Thanks you anyway 🙏
Thanks for note. Where (minute in video or slide number) do I say that there is no need for a bulk cap?
As for C/2 see th-cam.com/video/T8qWWwsMOK8/w-d-xo.html
The bottom circuit is missing a DC blocking capacitor...
👍
Thanks for a Great Video. Curious to see answer for Riddle - What is wrong with one of these circuits?
Isn't this an answer?
Sir, i am very greatfull to listen to your videos.
Diodes in bottom circuit will see serious voltage spike at turn off as they are chopping inductor current. So snubber on diodes is mandatory.
Hi Venkata, Please indicative which slide number or or minute you are referring to.
@@sambenyaakov Sir, at 4:33, a current flowing into Q1 splits into C1 and C2. It is obvious that the portion of current in C2 will flow back to the negative terminal of the source. But What happens to a portion of current entering into C1? How it will reach a negative terminal of source? It is very uncomfortable to digest sir.
I saw recently this design, with what you sad in Mind, would you say the connection of the middle of the Buss(in red) is not optimal, they have (in Blue) Capacitors across to manage those currents but i would not have connected the middle of the 0.1 Capacitors to the Main Bulk Capacitors. Or do i see this wrong?
Please clear to which slide #, minute, you are referring to.
O i forgot to attach the images, sorry,
imgur.com/a/RQTXw0k
This looks OK. The blue caps to suppress spikes across bus and the red for steering the ac current. It is desirable to put ceramic caps in parallel to electrolytic for the high frequency components. O.1u seems to me too low though.
would this not be a problem as you mentioned around 10:54, without a Cb does the ripple current not become a problem. And yes 0.1µF is not much the electrolytic will probably see quite the current.
hello. I have 2half bridge on IR2153 in my DC/DC power. And I dont understand that these Voltage is not dividade by 2. When IR2153 start working one capacitor has 0.9 Uin second 0.1Uin
It is probably a start up problem you need soft start/
@@sambenyaakov I am uisng to do it ir2153 driver...
I do not understand, on slide 7 current is flowing from +Vin to again +Vin why? The current is always flowing from high potential to low...
Good point. But...Lf is a current source. When reflected to transformer primary it will (by superposition ) cause the currents indicated by the arrows.
In half bridge inverter, changing capacitors, does it change natural resonance
Which capacitors?
Series inductor-transformer primary provide DC path to ground. Cores will saturate. Unlimited DC current will flow. Part(s) will fail.
🤔
I think what is wrong is that in the PR converter there is no DC blocking.
Thanks. Please see posted answer.
Explain full bridge pwm controller
I think you're missing resistors that are typically in parallel with each of the "voltage divider" capacitors. Typically there is a resistor (~100k Ohms) in parallel with each of the caps. This creates a voltage divider with Vin/2 in the middle and Vin/2 across each cap. When each of the FET's turn on, the transformer is hit with a constant voltage from the caps which creates the current ramp waveform through the transformer. Using the inductor equation V = L di/dt and solving for the constant inductor voltage you get an integral of a constant over time which is a ramp for the inductor current. So typically there are resistors in parallel with the capacitors to set the voltage.
Well, you are mistaken. In fact,as pointed out in video, there is no need for two capacitor. One is OK (so where would you put the resistor "voltage divider"). Try to watch the video end to end with open mind. If you then have a question, I will be happy to discuss.
@@sambenyaakov I'm referring to R1 and R2 in this image: www.seekic.com/uploadfile/ic-circuit/201162412958301.jpg. The resistor divider is also mentioned in the answer to this stackexchange post: electronics.stackexchange.com/questions/259406/dc-dc-half-bridge-converter-capacitor-selection. With the two capacitors the duty cycles need to be almost exactly the same for the two transistors. Reference: www.runonielsen.dk/Half_bridge_control.pdf.
@@sambenyaakov Here's an example of a real life power supply design that uses the resistors: th-cam.com/video/pffOJdCQ7kw/w-d-xo.html
@@sambenyaakov Here is the quote from Runo's Power Design: "In the half bridge the two switches must run with the same pulse width, so that the midpoint between capacitors C1 and C2 will be at half of the input DC voltage."
@@sambenyaakov You also might want to explain, in the resonant LLC, that the Lr Cr are important because they create a filter that filters out the higher order harmonics, leaving behind the fundamental harmonic from the drive pulses. The resulting (filtered) waveform is a sine wave.