Thank you Sam ! I just now realize that the circuit at 6:39, (the circuit with the diode + inductor + capacitor) is exactly what the input to a boost converter looks line if you accidentally switch on the input voltage without a PTC inrush-current protector.... Incidentally, that's exactly how I blew up a PFC Totem board a few weeks ago 😞. I wish I knew then what I know now...
That's why, in my opinion, good designs include large 50Hz bypass diodes across the coils during energization. They will take the current surge AND avoid the resonance of the circuit and the associated overvoltage. If I remind correctly, on the totem pole circuit, you put a pair of these diodes between the coil-connected AC terminals and your DC bus terminals. These diodes will present less impedance to the surge than your fast diode + coils, but won't conduct any current in normal operation (so no recovery). They don't need to be rated for the full current, and even a dumb 1N5408 (1000V 3A) can handle more than 100A peak of inrush for a couple of like cycles. By the way, don't you mean NTC instead?
@@pbs1516 In my totem pole PFC, I use a PTC for inrush current protection. That would be Rpre in Sam's schematic at 2:07. That resistor nicely limits the inrush current, and serves as short-circuit protection as well. The body-diodes of the totem pole mosfets then serve as a rectifier that charges the bus capacitors gently. No further diodes are needed. The only thing I did wrong was accidentally switch S2 on while S1 was still open on my prototype board. That's not a good idea as I found out the hard way LOL.
Excellent video! Please sir could you make detailed video about self excited lr multivibrator fluxgate sensor using op amp. I couldn't find any source about the calculations involved.
@@sambenyaakov yes all of it about 3 times 😋 that's what got me interested and I searched the internet and only found a single research paper that u cited. However couldn't get the full pdf as it required login or something. Searched everywhere but to no avail about anything, I think my only last hope is u professor😃 I am a subscriber to ur channel and I found it the most detailed channel about ee on RUclips.
Maybe the wording in my comment was bad. I've worked with designs similar to your first diagram, but instead of a pre charge resistor, there was just a single resistor used to limit current into the capacitor. Additionally, instead of a battery it was rectified line voltage. Doing fancy precharge switching is too expensive/unreliable, so we just use a single power resistor in series to protect the capacitor from exceeding dv/dt ratings. Worst case is the condition where your power cuts out, and returns at peek of line when the capacitor is discharged. There are not many suppliers that offer power/surge resistors with peek pulse power ratings with ample margin, and some don't even include the pulse power/ time duration curve in their datasheet . From what I've seen KOA has the best resistors and exceeds their ratings when tested better than other mfgs. Obviously there are other ways around this by going to a larger package size, but everything has its tradeoffs. @@sambenyaakov
@@Cydget Using a series transistor in the common way is dangerous due to violation of the SOA. See new idea: ruclips.net/video/05QprDgrP6E/видео.html ruclips.net/video/-08rwDt5Wp8/видео.html
very nice insight and foundational one
Thanks
This is a very important technique (with Diode) for efficient energy harvesting. Thank you very much.
Yes, thanks
Thank you Sam ! I just now realize that the circuit at 6:39, (the circuit with the diode + inductor + capacitor) is exactly what the input to a boost converter looks line if you accidentally switch on the input voltage without a PTC inrush-current protector....
Incidentally, that's exactly how I blew up a PFC Totem board a few weeks ago 😞.
I wish I knew then what I know now...
That's why, in my opinion, good designs include large 50Hz bypass diodes across the coils during energization. They will take the current surge AND avoid the resonance of the circuit and the associated overvoltage.
If I remind correctly, on the totem pole circuit, you put a pair of these diodes between the coil-connected AC terminals and your DC bus terminals. These diodes will present less impedance to the surge than your fast diode + coils, but won't conduct any current in normal operation (so no recovery). They don't need to be rated for the full current, and even a dumb 1N5408 (1000V 3A) can handle more than 100A peak of inrush for a couple of like cycles.
By the way, don't you mean NTC instead?
@@pbs1516 In my totem pole PFC, I use a PTC for inrush current protection. That would be Rpre in Sam's schematic at 2:07. That resistor nicely limits the inrush current, and serves as short-circuit protection as well. The body-diodes of the totem pole mosfets then serve as a rectifier that charges the bus capacitors gently. No further diodes are needed.
The only thing I did wrong was accidentally switch S2 on while S1 was still open on my prototype board. That's not a good idea as I found out the hard way LOL.
A very nice twist on that classic capacitor charging paradox
👍🙏
Excellent video! Please sir could you make detailed video about self excited lr multivibrator fluxgate sensor using op amp.
I couldn't find any source about the calculations involved.
Thanks. Have you seen ruclips.net/video/ry6j32Q2VBw/видео.htmlthere are 3 parts.
@@sambenyaakov yes all of it about 3 times 😋 that's what got me interested and I searched the internet and only found a single research paper that u cited. However couldn't get the full pdf as it required login or something. Searched everywhere but to no avail about anything, I think my only last hope is u professor😃
I am a subscriber to ur channel and I found it the most detailed channel about ee on RUclips.
👍
Thanks
👍🙏❤
Cool, maybe this design could save money by allowing for cheaper series resistors that have a smaller peek pulse power rating.
The analysis discussed IS small signal. Unless I missed something. There are many videos in my RUclips channel on control.
Maybe the wording in my comment was bad. I've worked with designs similar to your first diagram, but instead of a pre charge resistor, there was just a single resistor used to limit current into the capacitor. Additionally, instead of a battery it was rectified line voltage. Doing fancy precharge switching is too expensive/unreliable, so we just use a single power resistor in series to protect the capacitor from exceeding dv/dt ratings. Worst case is the condition where your power cuts out, and returns at peek of line when the capacitor is discharged. There are not many suppliers that offer power/surge resistors with peek pulse power ratings with ample margin, and some don't even include the pulse power/ time duration curve in their datasheet . From what I've seen KOA has the best resistors and exceeds their ratings when tested better than other mfgs. Obviously there are other ways around this by going to a larger package size, but everything has its tradeoffs. @@sambenyaakov
I also havent seen your previous video yet, so I might still be off topic
@@Cydget Catch up 😊😊
@@Cydget Using a series transistor in the common way is dangerous due to violation of the SOA. See new idea: ruclips.net/video/05QprDgrP6E/видео.html ruclips.net/video/-08rwDt5Wp8/видео.html