Great video. A few notes: - The HV blanket should cover all parts up to and including the left side of the transformer and optoisolator. Neutral is a high voltage net if the outlet or power cord is wired backwards or the neutral opens up upstream of the outlet - I would add the voltage ratings for C16 and C23 on the schematic as they are both high voltage caps - I would also add a bleeder resistor across C23 - The clearance in the GND-P pour for C23 positive pin looks too close for 388v (275V * 1.414)
That HV blanket is used for a clearance rule, the rules is set between all nets in the HV net class and all other traces, so it covers the case you are referring to since that rule takes priority. But anyways, the build of this is in progress and I'll do another video on it once it's finished.
Can you provide some additional insight into how the feedback path works? I'm trying to understand the selection of bias resistor values on both sides of the optocoupler. Thanks!
Sure, this kind of feedback method uses a shunt regulator to bias the LED side of the optocoupler at a specific value. The shunt regulator has a threshold at which it starts admitting power delivery to the optocoupler, which is why you have that resistor divider. On the primary side, the resistors are chosen such that he FB pin is biased at a target value. So if you are below the shunt reg. threshold, then the current in the optocoupler drops, and then the voltage reaching the FB pin also drops, this then drive adjustment of the PWM signal internal to the switcher.
Hi Zach, Altium Rule quickly question, Can it set a rule to check a spectial VIA(not thru pad hole) with only one line in all layers? The correct case is that there must be copper or wire segments connected in the two layers, Thanks
Great video very informative for hardware design engineers I would like to know that how to isolate the input AC line from DC voltage and asking how much the AC trace thickness should be thank you
AC trace sizing (thickness and width) should be based on the amount of current it will need to carry. In this example the flyback converter is stepping down to a low voltage so it will have very low current on the primary side. Still, it is typical to make the traces on the input large enough to handle any surge current when the device first powers on or if there is a fault somewhere on the primary side.
Very nicely explained. The only thing I would add is there should be the typical isolation slot (under U9 and T1 in this case) to increase creepage distance.
thanks sir for this clear explanation , I have question about return current path and having a solid ground plane under traces . why we don't care in power electronics board about return current path ?
We do care about the return current having a well-defined location, including in power electronics. In this design, the primary ground is being placed under the switcher with a direct connection to GND and to the bridge rectifier to complete the circuit. The input line and neutral form the remainder of the return path on the primary side. On the secondary side, I also placed a GND pour region to form the return path for the 3V3 output. If this were a switching converter with an inductor, we would also want to place ground below the switch node to control where the return current exists and to control radiated emissions. You can learn more about this aspect in this video: ruclips.net/video/6ind9vopKZs/видео.html
I am confused. A flyback transformer needs no switches, so why are you designing this with noisy switches. Flybacks are used for audio because they have no noisy switching. Why are you using features of a Switch mode power supply in a flyback transformer? Can you explain please.
@@richardsardini5585 Okay, flyback transformers are used in Switching power supplies. In the past they were commonly used to generate the EHT for Cathode ray tubes. The same concept is applied here in designing flyback converters. The "fly-back" concept comes from switching DC voltages at a high frequency across a coupled inductor. Therefore, without the said switches, the flyback transformers cannot perform their intended task.
@@richardsardini5585 Flyback works only on DC voltages...In the offline flyback converters, the mains AC is first rectified to about 340V DC (i.e for the 240V RMS mains). This DC is then switched on and off at a very high frequency across a coupled inductor (which we now call the flyback transformer).
It is not a requirement for the design to function, but it does increase creepage distance if low creepage is a problem, and it makes the boundary between the isolated regions very easy to see for someone who looks at the board, so nothing wrong with adding it. I have seen boards that include the cutout and some that do not.
Great video. A few notes:
- The HV blanket should cover all parts up to and including the left side of the transformer and optoisolator. Neutral is a high voltage net if the outlet or power cord is wired backwards or the neutral opens up upstream of the outlet
- I would add the voltage ratings for C16 and C23 on the schematic as they are both high voltage caps
- I would also add a bleeder resistor across C23
- The clearance in the GND-P pour for C23 positive pin looks too close for 388v (275V * 1.414)
That HV blanket is used for a clearance rule, the rules is set between all nets in the HV net class and all other traces, so it covers the case you are referring to since that rule takes priority. But anyways, the build of this is in progress and I'll do another video on it once it's finished.
I really enjoyed this video, very informative. Thank you for sharing.
Can you provide some additional insight into how the feedback path works? I'm trying to understand the selection of bias resistor values on both sides of the optocoupler. Thanks!
Sure, this kind of feedback method uses a shunt regulator to bias the LED side of the optocoupler at a specific value. The shunt regulator has a threshold at which it starts admitting power delivery to the optocoupler, which is why you have that resistor divider. On the primary side, the resistors are chosen such that he FB pin is biased at a target value. So if you are below the shunt reg. threshold, then the current in the optocoupler drops, and then the voltage reaching the FB pin also drops, this then drive adjustment of the PWM signal internal to the switcher.
Hello. Where to find the 3D model and footprint of the transformer ?
The switch U8, I don’t prefer make the optical coupler output route under switcher U8.
But anyway it will work❤
Hi Zach, Altium Rule quickly question, Can it set a rule to check a spectial VIA(not thru pad hole) with only one line in all layers?
The correct case is that there must be copper or wire segments connected in the two layers, Thanks
Great video very informative for hardware design engineers I would like to know that how to isolate the input AC line from DC voltage and asking how much the AC trace thickness should be thank you
AC trace sizing (thickness and width) should be based on the amount of current it will need to carry. In this example the flyback converter is stepping down to a low voltage so it will have very low current on the primary side. Still, it is typical to make the traces on the input large enough to handle any surge current when the device first powers on or if there is a fault somewhere on the primary side.
It would be nice to see a simulation of that flyback in Altium
Very nicely explained. The only thing I would add is there should be the typical isolation slot (under U9 and T1 in this case) to increase creepage distance.
Lucky for you guys I am building some of these modules and the custom transformers for another video! stay tuned....
thanks sir for this clear explanation ,
I have question about return current path and having a solid ground plane under traces . why we don't care in power electronics board about return current path ?
We do care about the return current having a well-defined location, including in power electronics. In this design, the primary ground is being placed under the switcher with a direct connection to GND and to the bridge rectifier to complete the circuit. The input line and neutral form the remainder of the return path on the primary side. On the secondary side, I also placed a GND pour region to form the return path for the 3V3 output. If this were a switching converter with an inductor, we would also want to place ground below the switch node to control where the return current exists and to control radiated emissions. You can learn more about this aspect in this video: ruclips.net/video/6ind9vopKZs/видео.html
I really enjoyed the video. You really know you're way around hardware designing very well. It would also be cool to see it assambled and running
Build is in progress, I am waiting on the transformer assembly and bare boards now
Didn't find any blog link to design the transformer..
The blog link has been added into the description
I am confused. A flyback transformer needs no switches, so why are you designing this with noisy switches. Flybacks are used for audio because they have no noisy switching. Why are you using features of a Switch mode power supply in a flyback transformer? Can you explain please.
I can't imagine how a flyback transformer would work without a switch.
@@WaswaSayari I asked for an answer, not a stupid question.
@@richardsardini5585 Okay, flyback transformers are used in Switching power supplies. In the past they were commonly used to generate the EHT for Cathode ray tubes. The same concept is applied here in designing flyback converters. The "fly-back" concept comes from switching DC voltages at a high frequency across a coupled inductor. Therefore, without the said switches, the flyback transformers cannot perform their intended task.
@@WaswaSayari So, an AC to DC flyback would not need switching as the current does the switching for it. Is this a correct assumption?
@@richardsardini5585 Flyback works only on DC voltages...In the offline flyback converters, the mains AC is first rectified to about 340V DC (i.e for the 240V RMS mains). This DC is then switched on and off at a very high frequency across a coupled inductor (which we now call the flyback transformer).
Shall I add any board cutout under isolation barrier? if added what will be the pros and cons of that?
It is not a requirement for the design to function, but it does increase creepage distance if low creepage is a problem, and it makes the boundary between the isolated regions very easy to see for someone who looks at the board, so nothing wrong with adding it. I have seen boards that include the cutout and some that do not.
@@Zachariah-Peterson when shall is use pcb cutout , are there any use cases?
Thanks in advance.
Great explanation!
Where is the link how to design transformer?
A link to the blog has been added to the description
You can also watch this video: ruclips.net/video/-72g1V6v0GQ/видео.html
Someone asked and you showed
I have asked so many times to show arduino pcb creation on altium and you never show
😞😞
Phil's lab, another Altium contributor has a great video on designing a raspberry pi from scratch: ruclips.net/video/X00Cm5LMNQk/видео.html
Are you looking for a video on a custom Arduino PCB layout?