I've been evaluating a number of brick solid state DC out relays and have noticed something most never think about. The fast turn off circuit doesn't work if the control voltage slowly drops. This problem only happens with turn off. The gate drive can easily drop and stay at low voltages which force the FET to operate in a linear region. I had assumed that the on/off delays would provide a hysteresis allowing use with analog circuits. That cause the FET to burn up with relatively low current. Battery and self powered PV circuits could be prone to failure without voltage protection circuits. Thought your viewers might find this helpful.
Hi my Professor, this is a very useful video. I been trying to do a model just like this. This will save me alot of time and effort. That you so much for the video and sharing your knowledge!!! With warm regards, Robert Bolanos
Curious, I measured the short circuit current of the classic 1N914A while being intensely illuminated by the LED flash of my phone in direct contact with it, and with fine adjustment of its position, the peak short circuit current was up to 1.3uA, though 1uA was more readily achievable.
Hi Sam! Do you have some materials over emi filters calculation? Quite balagan in this topic, many topoligies with no strict calculations. And also not much info how you deal with dm/cm noise reduction in actual design.
@@sambenyaakov I did a LTspice simulation using a SST201 (JFET) and a 100k resistor. Works flawless. Full solar current for charging, fast discharging (even faster than charging). A simple discharge resistor would work MUCH worse. What kind of issues are you thinking of?
This driver should be used with care, in my prototype it lead to thermal runaway as gate voltage drops until finally MOSFET failure. Also due to their price practical use may be questionable, tiny dc-dc can be cheaper
Wouldn't it be better to have that kind of circuit to provide power to charge a capacitor and then a faster system for control that uses that energy to toggle the gate? BTW: the energy efficiency is atrocious 0.2% !
@@sambenyaakov You could power a CMOS gate and decoupling capacitor with the solar current and set the switching threshold high enough. This way, the MOSFET would switch very fast ON and OFF. But yes, the average gate power is limited to the power provided by the solar cells.
Hello dear sam I have learned from your videos so much Have you ever worked on reaonant gate drivers and current source gate drivers ? Which model do you recommend ? Do you any article that did some analysis on them ?
Love these! There are so many weird use cases beyond just driving gates. Another good one: IL300 Linear analog optocoupler
Indeed ruclips.net/video/b23tQZ2KCp8/видео.html
omg hi marco
I've been evaluating a number of brick solid state DC out relays and have noticed something most never think about. The fast turn off circuit doesn't work if the control voltage slowly drops. This problem only happens with turn off. The gate drive can easily drop and stay at low voltages which force the FET to operate in a linear region. I had assumed that the on/off delays would provide a hysteresis allowing use with analog circuits. That cause the FET to burn up with relatively low current. Battery and self powered PV circuits could be prone to failure without voltage protection circuits. Thought your viewers might find this helpful.
Thanks for sharing. I was nit aware of that.
Hi my Professor, this is a very useful video. I been trying to do a model just like this. This will save me alot of time and effort. That you so much for the video and sharing your knowledge!!! With warm regards, Robert Bolanos
Thanks Robert
Curious, I measured the short circuit current of the classic 1N914A while being intensely illuminated by the LED flash of my phone in direct contact with it, and with fine adjustment of its position, the peak short circuit current was up to 1.3uA, though 1uA was more readily achievable.
Interesting
Informative, Thank you very much.
Thanks. I appreciate it.
Hi Sam! Do you have some materials over emi filters calculation? Quite balagan in this topic, many topoligies with no strict calculations. And also not much info how you deal with dm/cm noise reduction in actual design.
It is a mess. I have nothing to suggest.
@@sambenyaakov wow didn't expect that bad
Check the SOA of the MOSFET with these slow switching times.
Very good point. Thanks for pointing this out.
Application note 31 from Panasonic tells the "secret" of the fast turn off circuit. A depletion mode FET.
I don't think that this is a D mode MOSFET and there are some issues with the circuit as shown in the application note.
@@sambenyaakov I did a LTspice simulation using a SST201 (JFET) and a 100k resistor. Works flawless. Full solar current for charging, fast discharging (even faster than charging). A simple discharge resistor would work MUCH worse. What kind of issues are you thinking of?
@@JacquesMartini What happens if there is no load, no capacitance at the output? Still. Also, I wonder if a jfet can be mixed in the IC.
This driver should be used with care, in my prototype it lead to thermal runaway as gate voltage drops until finally MOSFET failure. Also due to their price practical use may be questionable, tiny dc-dc can be cheaper
Why did it runaway? Can you share the catalog number of inexpensive DC DC converters?
@@sambenyaakov because more temperature - less gate voltage - higher resistance - more temperature... Etc . Catalog LCSC
Wouldn't it be better to have that kind of circuit to provide power to charge a capacitor and then a faster system for control that uses that energy to toggle the gate?
BTW: the energy efficiency is atrocious 0.2% !
Yes, but then you need an isolated driver and you still are limited by the rate of switching on and off.
@@sambenyaakov You could power a CMOS gate and decoupling capacitor with the solar current and set the switching threshold high enough. This way, the MOSFET would switch very fast ON and OFF. But yes, the average gate power is limited to the power provided by the solar cells.
Hello dear sam
I have learned from your videos so much
Have you ever worked on reaonant gate drivers and current source gate drivers ?
Which model do you recommend ?
Do you any article that did some analysis on them ?
I did work on resonant gate drivers awhile ago. Look up papers by David J. Perreault of MIT
@@sambenyaakov thank you appreciate it
You are missing the most interesting part, the turn off ciruit. Simple but clever!
Why won't you enlighten us? I doubt if the manufacturers provide the full details of the turn-off circuit design.
@@sambenyaakov Where is my comment? I did post some links to a data sheet and a ltspice file. Search for Application Note 031 from panasonic.
Did you get a new mic? Sounds a ton better.
🙂it is the internal microphone array of the Surface 9 pro
@@sambenyaakov Nice. Sounds much clearer and has a much better noise floor than the previous one you were using.
@@gsuberland 👍👍
👍🙏❤️
🙏👍🙂