I really enjoyed the video, Steve. looks as if the design will be just what you're needing. I'm looking forward to seeing if it works out for you. Thanks for sharing.
When you are doing layout of switcher it is important to make parasitic circuit elements work for you rather than against you. In the circuit in the video, the output ov the bridge rectifier should go straight to the positive terminal of C1. The connection to the input side of L1 should go straight from the positive terminal of C1. This makes the connection a sort of "U" path. It is very much like what you are trying to do when using "Kelvin connections" when measuring low value resistors. Similarly the output side of L1 should go straight to the positive terminal of C2 and the connection to the high side of the transformer should go straight from C2 to the transformer pin. Doing this makes the parasitic inductances of the tracks work for you with regard to high frequency currents rather than against you. In reality in this particular design it probably makes little difference. The circuit is reasonably compact to begin with and aluminum electrolytic capacitors are pretty dismal as capacitors at high frequency. If there were noticeable difference it would likely be as a small improvement at the fundamental switching frequency. In lots of designs of switchers caps with good high frequency characteristics are used in parallel with electrolytic caps. There layout is more critical to get full benefit of the (typically rather large and expensive) additional capacitors. Spool core inductors tend to be relatively poor at high frequency, too, though it varies a lot with the winding. With a multi-layer winding there can be significant capacitance between the input and output, and often in the worst possible place. Toroids can be better if the winding is "progressive" or a single layer, but they are more expensive. The circuit also has narrow tracks in the output and wide ones in the input. That is backwards. Peak currents in the input circuit are quite small but can be very large in the output. Remember that when the switch turns off the current in the output winding instantly rises from zero to the (current that was flowing in the primary) times (the primary to secondary turns ratio). Again the output of the PI filter is from the wrong place. It should come directly from the positive terminal of the output cap (C8 ??)
I don't know if it's possible, ie, it might be too different, but maybe put all the features on one PCB, ie, opto feedback, and populate them one at a time so as to directly compare if there is an advantage. Prob overkill - lol -.
Great content. Pity you haven't got a means of zooming it when you are showing text on your screen as my laptop screen isn't that high a resolution and its quite small too. I despair when a creator tries to show program listings. lol.
Is there any reason you opted to use rounded corners instead of 45deg for the traces? Also there are some sharp angles. This is not meant as criticism, I have been wondering for some time now if rounded corners are better. My bad if I missed it in the video.
The only place where curved corners are really advantageous is in high frequency design where you want to keep the characteristic impedance of a track constant - you are trying to make a "transmission line." Curved corners make this easy because the width of the track is constant around the curve. With an "L" shaped corner the outside peak would typically be trimmed off to minimize the impedance change, but that is a compromise. If you can find old Motorola RF applications notes you'll see trimmed corners on tracks. You'll also find PCB tracks designed for specific inductance as part of impedance matching networks. There are certainly high frequency content in the waveforms found in switchers, but keeping track impedance constant isn't normally a concern. Usually you are trying to keep connection impedances as low as you possibly can.
Thanks for sharing. I have a question you might know, are these design ideas on power integration web site free for commercial use? I want to implement one for my circuit but couldnt be sure its legality
An interesting overview of the tools and processes used in this PSU design. Does this take the fun or the drudgery out of the equation? Delete where applicable. Thanks for the video!
Almost an infomercial with commercials. Unsubscribed due to lack of convergence: happy you be so smart, but the is little connection to karaoke videos.
![ผัวเถื่อน - ยูริ โตเกียวมิวสิค [ OFFICIAL MUSIC VIDEO ]](http://i.ytimg.com/vi/P08ORRVDGl8/mqdefault.jpg)
Can't wait to see how you wind the transformer.
It's nearly impossible for anyone not using a 4K screen to see what you were doing with the design tool.
I really enjoyed the video, Steve. looks as if the design will be just what you're needing. I'm looking forward to seeing if it works out for you. Thanks for sharing.
Very interesting video. Thanks !
I love learning new circuit designs and techniques! I wasn't aware that the home hobbyist could design circuits like this. Thank you, Steve.
I can see the text on an iPhone at 720 resolution.
Fascinating stuff and I'm looking forward to making the transformer....cheers.
I struggled watching this on a laptop screen, soo small texts. Cant imagine watching something so small on phone screen. Nice info though
Will make sure everything is much larger next time!
Thanks for this, been wondering about flyback design for a while.
When you are doing layout of switcher it is important to make parasitic circuit elements work for you rather than against you.
In the circuit in the video, the output ov the bridge rectifier should go straight to the positive terminal of C1. The connection to the input side of L1 should go straight from the positive terminal of C1. This makes the connection a sort of "U" path. It is very much like what you are trying to do when using "Kelvin connections" when measuring low value resistors. Similarly the output side of L1 should go straight to the positive terminal of C2 and the connection to the high side of the transformer should go straight from C2 to the transformer pin. Doing this makes the parasitic inductances of the tracks work for you with regard to high frequency currents rather than against you.
In reality in this particular design it probably makes little difference. The circuit is reasonably compact to begin with and aluminum electrolytic capacitors are pretty dismal as capacitors at high frequency. If there were noticeable difference it would likely be as a small improvement at the fundamental switching frequency. In lots of designs of switchers caps with good high frequency characteristics are used in parallel with electrolytic caps. There layout is more critical to get full benefit of the (typically rather large and expensive) additional capacitors.
Spool core inductors tend to be relatively poor at high frequency, too, though it varies a lot with the winding. With a multi-layer winding there can be significant capacitance between the input and output, and often in the worst possible place. Toroids can be better if the winding is "progressive" or a single layer, but they are more expensive.
The circuit also has narrow tracks in the output and wide ones in the input. That is backwards. Peak currents in the input circuit are quite small but can be very large in the output. Remember that when the switch turns off the current in the output winding instantly rises from zero to the (current that was flowing in the primary) times (the primary to secondary turns ratio).
Again the output of the PI filter is from the wrong place. It should come directly from the positive terminal of the output cap (C8 ??)
Well done!
When you created your pcb where did you get your schematic and footprint for the transformer? Can you give me a part number?
I don't know if it's possible, ie, it might be too different, but maybe put all the features on one PCB, ie, opto feedback, and populate them one at a time so as to directly compare if there is an advantage. Prob overkill - lol -.
Great content. Pity you haven't got a means of zooming it when you are showing text on your screen as my laptop screen isn't that high a resolution and its quite small too.
I despair when a creator tries to show program listings. lol.
I'll bear this in mind for next time!
Are you missing the filter cap which is supposed to be between primary and secondary? I mean the class Y one.
Love the curvy traces.
Is there any reason you opted to use rounded corners instead of 45deg for the traces? Also there are some sharp angles. This is not meant as criticism, I have been wondering for some time now if rounded corners are better. My bad if I missed it in the video.
No reason, it's just what the default was when I started layout in KiCAD!
@@sdgelectronicsThank you!
The only place where curved corners are really advantageous is in high frequency design where you want to keep the characteristic impedance of a track constant - you are trying to make a "transmission line." Curved corners make this easy because the width of the track is constant around the curve. With an "L" shaped corner the outside peak would typically be trimmed off to minimize the impedance change, but that is a compromise. If you can find old Motorola RF applications notes you'll see trimmed corners on tracks. You'll also find PCB tracks designed for specific inductance as part of impedance matching networks.
There are certainly high frequency content in the waveforms found in switchers, but keeping track impedance constant isn't normally a concern. Usually you are trying to keep connection impedances as low as you possibly can.
Thanks for sharing. I have a question you might know, are these design ideas on power integration web site free for commercial use?
I want to implement one for my circuit but couldnt be sure its legality
❤❤❤
An interesting overview of the tools and processes used in this PSU design. Does this take the fun or the drudgery out of the equation? Delete where applicable. Thanks for the video!
Nice
P R O M O S M 😈
long time, no content 👋
Almost an infomercial with commercials. Unsubscribed due to lack of convergence: happy you be so smart, but the is little connection to karaoke videos.