I've been looking into circuit protection for some time, and this is really cool. You could toss in a fuse or a breaker for overcurrent protection as well. Great video!
Ideal diodes are great, but usage depends a lot on the system. SLVAE57B is a good application note comparing different circuits. For anti-spark, and especially to prevent voltage spikes due to downstream capacitance and cable inductance, usually it's better to just use a precharge circuit
@@vincentnguyen8083 thank you for letting me know! I will have a look at that application note 😁 What do pre charge circuits involve? How do you get the delay? I’ve seen it done with relays before, but I don’t really like relays, and I know a similar thing is possible with MOSFETs, I just haven’t looked hard enough to see one yet. Why is it better to use such a pre charge circuit vs this one? Does it charge even slower allowing for even less current spike?
Wait, but the p channel circuit from the beginning can be made the same with an n channel mosfet, just have to put the mosfet in the negative rail. Like we put the drain to V-, the gate goes to +V via resistor+zener(if we expect more than ~15V), and then the source is the output ground that goes to our device. When i was making a dc/dc charger for lifepo4 batteries that circuit worked well. The gate capacitor trick is interesting, probably can adapt it to the simpler circuit too. But a chip like this one is probably the way to go if one wants low losses high side switch protection.
@@victorman2227 cool, thanks for letting me know! I think the chip has a charge pump built in though so it may be able to work at lower voltages and still have the MOSFET fully turned on… Thanks so much for this information though 😁
By the way, don’t use ideal diode mode for motor driving systems. Use RPP mode. Ideal diode mode prevents the motors from returning energy to the battery (blocks reverse current), causing the voltage to rise up to dangerous levels. I’m very sorry, I should’ve mentioned this.
So just clarifying, RPP mode would also protect against someone accidentally pushing the robot (causing flyback current to the motor driver)? Are you also thinking of adding overcurrent protection - that would be a great addition for the next video, no? ;) Looking forward to it either way (as I think your videos are well presented and very informative, so keep up the good work)!
@@darrenjcosborne RPP mode allows the main circuit to not be damaged during flyback current, correct :) Overcurrent protection would be good, I'll have to figure out how to do that though (sounds simple enough). Thank you for commenting :) Have a great day.
I've used similar ICs from Linear for high current application. For most products i work on a SOT-23 P channel MOSFET tend to be a better option. I usually work with low power embedded devices that draw less than 2A.
@@rnayabed KiCad for the circuit board design. I did not do the analysis myself, I was in contact with onsemi and they were kind enough to actually make the circuit and test it, for free.
Hey, I'm a big fan of yours and also have passion for mechatronics. Can you please tell from what sources did you learnt all this bc I wanna learn it too! Thanks
@@AXL_ARSH Online tutorials, data sheets, technical reference manuals, application notes, anything documenting devices which I would like to use / learn concepts from. I couldn’t name anything specific though for my whole knowledge, I must’ve read hundreds of documents by now and I would still consider myself a beginner lol. If you want to know how I learnt specific things, I might be able to remember which documents I read for that
just make sure you dont use FinFet for power aplications where linarr(slow turn on) mode is possible. FinFets have low junktion to case thermal conduction. Ak even thou it may say 50A its gona blow in instant once you try to pass 10A in linear mode. Junktion itself has usually verry low thermal mass, causing it to heat >150C within ms, effectivly burning it to dead short.
If you check out this datasheet: www.infineon.com/dgdl/Infineon-BSC022N04LS6-DataSheet-v02_01-EN.pdf?fileId=5546d462689a790c0168bd1094dc481b The drain current when the case temperature of the MOSFET is kept at a constant 25 degrees is 139 amps, but when the thermal resistance is specified (Rthja = 50C/W), as in, the thermal resistance between the junction (die) of the MOSFET and the surrounding air, this rating quickly decreases to 27A. It all comes down to cooling. However, if the MOSFET you find has a current rating of say, 1.5x your maximum current draw, when Rthja is being taken into account (not all MOSFETs have this), then that's fine. Because some MOSFETs don't have the Rthja Id rating, instead specifying the Tc Id rating, which is unrealistic, I tell you to times your system's drain current by 7 and look for the Tc Id rating from that, to make an estimation of which MOSFETs would have a Id Rthja right for your system, since that is normally what the manufacturers specify on the product page. Thanks, let me know if you have any more questions :)
@@gergelybencsik8626 Yep! You just need to find the Id rating with the right specified thermal characteristic: not Tc, but Rthja (or R theta ja is also written with the theta symbol). Hope this helped!
@@h1tec They are describing the crowbar circuit. The resistor can also be a fuse or polyfuse. Zener serves dual function overvoltage protection and reverse bias protection.
@@h1tec Not per se. A crowbar circuit like this is my favourite way to retrofit last-recourse protection into classic computers of the 80s where ICs are rated to 5.5V but are irreplaceable, while voltage regulators and old PSUs are known to sometimes fail short through a linear regulator to output. Commodore PSUs were often potted so you can't really repair or service them all too well, but people like keeping things original.
@@Hypoengg this is smart! For high powered circuits which draw lots of current, N channel MOSFETs are the way to go though 😁 I think there are ways to make 4 MOSFETs act as an inverter for this purpose with a specialised IC, GreatScott has covered it before I’m pretty sure. I’ve thought of doing this before but only after I made this circuit and had it manufactured 😂
@@SianaGearz correct. You can do the same thing with MOSFETs with a specific controller IC though, I think GreatScott has covered it. Using full-bridge rectifier for low voltage, low current circuits which are meant to be low-power and efficient shouldn't use this in my opinion. High voltage, low current, maybe it will be ok (like wall output)
@@h1tec Yeah i should do active rectification for bicycle hub dynamo. I want to build a dynamo charged powerbank into the steering bar tube. Also shunt regulation works well for this particular use, allowing the bicycle to freewheel more. It also ideally needs some sort of active power point control to improve harvesting.
3 problems: 1)bridges has usually 1.4V drop. on 5V its verry big drop. 2) Ground return path is impossible, most applications use negative as ground return for signals(USB) 3) some PSUs have no PE tied to negative. Instead its connectet do AC side with 2 capacitors leaving negative 1/2 of AC voltage(verry low current). If now you have full bridge and you connect negative after bridge to ground(ex scope goound wire), its gona boost your DC supplt to 1/2*1.4V AC voltage unless your circuit draws enough current to overcome parasitic current from AC side.. in 240V applications i daught any low voltage device can tolerate around 150VDC at input.
"I am glad your a Genius; know that your doing something extraordinary(Meaning? You understand and also build Circuits; You have my support. I love creative minds who share similar interest. Know that you have a friend in the sky, and? I encourage you to build the best Robot." 👍
I do wonder what the target audience was for this chip, I can see the appeal for a hobbyist, but for a product it seems either overkill or lazy. I.e. in low current applications it's entirely unnecessary, in high current applications your batteries will have keyed connections anyway preventing reverse polarity, and then using this chip just for the anti spark and enable features seems like a waste. I suppose it could appeal to lazy or time crunched engineers? Idk it's just odd to me, maybe there's some sketchy foreign market for these things? Or perhaps I'm underestimating the time crunched / lazy engineering market (likely true).
@@xxportalxx. yeah, it was meant for hobbyists. Antispark connectors like the XT90-S cannot be used in space-demanding applications. You could also set up a low voltage cutoff with this circuit, with a comparator, voltage divider and small MOSFET, so it is very versatile and useful for people wanting to design PCBs themselves
@hxtec32 yeah perhaps drones and power tools would use it? It would have to be a situation where the solution gives it a bit more premium feel, but engineering time is at a premium (I.e. a semi custom solution becomes preferable). Then again perhaps I'm underestimating the problem. Most of my engineering experience lately has been in industrial spaces where the engineers would just throw a third party, warranteed, module at the problem haha
Hmmm.. I might do something like that, consolidating all of the online calculators into one. But, I'll have to learn how the existing calculators work first though 😂 Thanks!
Honestly speaking I don't like these gate time tricks since you do create a heater. Why not to use an external soft-start circuitry? Why not to modulate these switches instead by compararator and current shunt or just use instead cheap mcu? You can basically use a damn NTC for this job or 2 of these things starting first via resistor and then normal case.
No, this is not "best". Best depends on your needs, price point, and component availability. If you can't get a component, that is the worst solution. If it is too expensive or overkill for your actual circuit's needs, it is just suboptimal.
I mean, fair. But, it is much much much more efficient past a certain current rating and applies to the same sort of circuits that many other UA-camrs claim is 'THE way' to protect against reverse polarity. Also, to be completely honest, I needed views on this video which I spent around 20 or so hours making. If I didn't name it 'BEST' or say 'Is this the BEST?' no one will click on it and watch it. What would you use instead for >5A circuits? Do other solutions you've seen on UA-cam have the same anti-spark power switch function? I was trying to fill a gap in the search term 'reverse polarity' because no one else (at least popular) mentioned N-channel FETs, which I thought was a no-brainer. Also, for low-power circuits, even though it is likely quite a small difference, the lower resistance of the N-channel FETs can help save more power. No hate or anything by the way. Have a good day 😁
Do you like the changed style of this video? Reply to this comment with your thoughts... Thanks 😀
I like it, thank you.
@@arcticpilotshow4440 thanks for letting me know 😁
bro I like it soo much
@@lukejijohn1476 thanks!
Stuff I shoulda learned 25 years ago. Good info for getting back to it.
@@SlinkyD thank you 😁 Hope it goes well!
I've been looking into circuit protection for some time, and this is really cool. You could toss in a fuse or a breaker for overcurrent protection as well. Great video!
@@MMuraseofSandvich true! Thank you! 😁
Your videos are very well organized and it's cool that I have now found a better solution for reverse polarity protection.;)
Thank you! 😁
This is an excellent video! Thanks for the exceptionally thought out and organized information.
Glad you enjoyed it! Thank you 😁
Ideal diodes are great, but usage depends a lot on the system. SLVAE57B is a good application note comparing different circuits. For anti-spark, and especially to prevent voltage spikes due to downstream capacitance and cable inductance, usually it's better to just use a precharge circuit
@@vincentnguyen8083 thank you for letting me know! I will have a look at that application note 😁 What do pre charge circuits involve? How do you get the delay? I’ve seen it done with relays before, but I don’t really like relays, and I know a similar thing is possible with MOSFETs, I just haven’t looked hard enough to see one yet. Why is it better to use such a pre charge circuit vs this one? Does it charge even slower allowing for even less current spike?
Wait, but the p channel circuit from the beginning can be made the same with an n channel mosfet, just have to put the mosfet in the negative rail. Like we put the drain to V-, the gate goes to +V via resistor+zener(if we expect more than ~15V), and then the source is the output ground that goes to our device. When i was making a dc/dc charger for lifepo4 batteries that circuit worked well. The gate capacitor trick is interesting, probably can adapt it to the simpler circuit too. But a chip like this one is probably the way to go if one wants low losses high side switch protection.
@@victorman2227 cool, thanks for letting me know! I think the chip has a charge pump built in though so it may be able to work at lower voltages and still have the MOSFET fully turned on… Thanks so much for this information though 😁
Beware of the external connections bypassing your ground. Say your device has USB in/out or an audio connection sleeve is grounded.
By the way, don’t use ideal diode mode for motor driving systems. Use RPP mode. Ideal diode mode prevents the motors from returning energy to the battery (blocks reverse current), causing the voltage to rise up to dangerous levels. I’m very sorry, I should’ve mentioned this.
So just clarifying, RPP mode would also protect against someone accidentally pushing the robot (causing flyback current to the motor driver)? Are you also thinking of adding overcurrent protection - that would be a great addition for the next video, no? ;) Looking forward to it either way (as I think your videos are well presented and very informative, so keep up the good work)!
@@darrenjcosborne RPP mode allows the main circuit to not be damaged during flyback current, correct :) Overcurrent protection would be good, I'll have to figure out how to do that though (sounds simple enough). Thank you for commenting :) Have a great day.
I've used similar ICs from Linear for high current application. For most products i work on a SOT-23 P channel MOSFET tend to be a better option. I usually work with low power embedded devices that draw less than 2A.
Fair enough, I would do the same 😁
Dude! Awesome video! Subscribed!
Thank you! 😁😁😁😁
Really good. I think the EN pin is active at 2.5 or so volts though, not 5v according to the data sheet. But nice job explaining!
Thanks! I think I read that it might’ve been better to enable at 5V though 😁
the style is funnier indeed :P
what is the name of the software you are using to design the circuits? can you also share software for analysis, etc?
@@rnayabed KiCad for the circuit board design. I did not do the analysis myself, I was in contact with onsemi and they were kind enough to actually make the circuit and test it, for free.
Hey, I'm a big fan of yours and also have passion for mechatronics. Can you please tell from what sources did you learnt all this bc I wanna learn it too! Thanks
@@AXL_ARSH Online tutorials, data sheets, technical reference manuals, application notes, anything documenting devices which I would like to use / learn concepts from. I couldn’t name anything specific though for my whole knowledge, I must’ve read hundreds of documents by now and I would still consider myself a beginner lol. If you want to know how I learnt specific things, I might be able to remember which documents I read for that
just make sure you dont use FinFet for power aplications where linarr(slow turn on) mode is possible. FinFets have low junktion to case thermal conduction. Ak even thou it may say 50A its gona blow in instant once you try to pass 10A in linear mode. Junktion itself has usually verry low thermal mass, causing it to heat >150C within ms, effectivly burning it to dead short.
@@Murphydeffa-oq8lm what do you mean by FinFet?
So... why 7x current rating? Is that for peak pulse current? Isn't that a bit much?
If you check out this datasheet: www.infineon.com/dgdl/Infineon-BSC022N04LS6-DataSheet-v02_01-EN.pdf?fileId=5546d462689a790c0168bd1094dc481b
The drain current when the case temperature of the MOSFET is kept at a constant 25 degrees is 139 amps, but when the thermal resistance is specified (Rthja = 50C/W), as in, the thermal resistance between the junction (die) of the MOSFET and the surrounding air, this rating quickly decreases to 27A. It all comes down to cooling. However, if the MOSFET you find has a current rating of say, 1.5x your maximum current draw, when Rthja is being taken into account (not all MOSFETs have this), then that's fine. Because some MOSFETs don't have the Rthja Id rating, instead specifying the Tc Id rating, which is unrealistic, I tell you to times your system's drain current by 7 and look for the Tc Id rating from that, to make an estimation of which MOSFETs would have a Id Rthja right for your system, since that is normally what the manufacturers specify on the product page. Thanks, let me know if you have any more questions :)
@@h1tec So in short, it all comes down to cooling and thermal design, got it. Very interesting.
@@gergelybencsik8626 Yep! You just need to find the Id rating with the right specified thermal characteristic: not Tc, but Rthja (or R theta ja is also written with the theta symbol). Hope this helped!
peace be upon you sir
@@DimasFajar-ns4vb thank you 😁
I would like an anti spark solution for 12s
I’m sure there are different ICs with higher voltage ratings, but how many amps are we talking?
@@h1tec amps is about 100, but this is dealt with an appropriate MOSFET with heatsink
Much easier way is resistor on input voltage and a zener clamp at 5.6V and after the switch and ur done
@@FriendlyIntentions what do you mean?
@@h1tec They are describing the crowbar circuit. The resistor can also be a fuse or polyfuse. Zener serves dual function overvoltage protection and reverse bias protection.
@@SianaGearz oh ok, thanks. would this still do the on/off switch part?
@@h1tec Not per se. A crowbar circuit like this is my favourite way to retrofit last-recourse protection into classic computers of the 80s where ICs are rated to 5.5V but are irreplaceable, while voltage regulators and old PSUs are known to sometimes fail short through a linear regulator to output. Commodore PSUs were often potted so you can't really repair or service them all too well, but people like keeping things original.
@@SianaGearz cool!
mmmm... Cool but i like to live life to the fullest😂 explosions are fun
anyways, yeah new style is very cool
@@santynolo 😂 thanks for letting me know!
Neat!
@@KofiAsare0 thank you!
You didn't justify the 7x current rating anywhere?
See the pinned comment and its replies. Sorry for not explaining that part clearly enough. Have a great day 😁
I just use a bridge rectifier, it not only protects the circuit but also makes sure that circuit works irrespective of the supply polarity.
@@Hypoengg this is smart! For high powered circuits which draw lots of current, N channel MOSFETs are the way to go though 😁 I think there are ways to make 4 MOSFETs act as an inverter for this purpose with a specialised IC, GreatScott has covered it before I’m pretty sure. I’ve thought of doing this before but only after I made this circuit and had it manufactured 😂
But it's very ineffiicient at low input voltages and has high dropout
@@SianaGearz correct. You can do the same thing with MOSFETs with a specific controller IC though, I think GreatScott has covered it. Using full-bridge rectifier for low voltage, low current circuits which are meant to be low-power and efficient shouldn't use this in my opinion. High voltage, low current, maybe it will be ok (like wall output)
@@h1tec Yeah i should do active rectification for bicycle hub dynamo. I want to build a dynamo charged powerbank into the steering bar tube. Also shunt regulation works well for this particular use, allowing the bicycle to freewheel more. It also ideally needs some sort of active power point control to improve harvesting.
3 problems:
1)bridges has usually 1.4V drop. on 5V its verry big drop.
2) Ground return path is impossible, most applications use negative as ground return for signals(USB)
3) some PSUs have no PE tied to negative. Instead its connectet do AC side with 2 capacitors leaving negative 1/2 of AC voltage(verry low current). If now you have full bridge and you connect negative after bridge to ground(ex scope goound wire), its gona boost your DC supplt to 1/2*1.4V AC voltage unless your circuit draws enough current to overcome parasitic current from AC side.. in 240V applications i daught any low voltage device can tolerate around 150VDC at input.
it would blow up like a fuse at....this is simple but...
i like your video, hey if you want some help in your projects i can help you out and together i think we can build cooler projects
"I am glad your a Genius; know that your doing something extraordinary(Meaning? You understand and also build Circuits; You have my support. I love creative minds who share similar interest. Know that you have a friend in the sky, and? I encourage you to build the best Robot." 👍
Thank you so much! I appreciate it 😁
I do wonder what the target audience was for this chip, I can see the appeal for a hobbyist, but for a product it seems either overkill or lazy. I.e. in low current applications it's entirely unnecessary, in high current applications your batteries will have keyed connections anyway preventing reverse polarity, and then using this chip just for the anti spark and enable features seems like a waste. I suppose it could appeal to lazy or time crunched engineers? Idk it's just odd to me, maybe there's some sketchy foreign market for these things? Or perhaps I'm underestimating the time crunched / lazy engineering market (likely true).
@@xxportalxx. yeah, it was meant for hobbyists. Antispark connectors like the XT90-S cannot be used in space-demanding applications. You could also set up a low voltage cutoff with this circuit, with a comparator, voltage divider and small MOSFET, so it is very versatile and useful for people wanting to design PCBs themselves
@hxtec32 yeah perhaps drones and power tools would use it? It would have to be a situation where the solution gives it a bit more premium feel, but engineering time is at a premium (I.e. a semi custom solution becomes preferable). Then again perhaps I'm underestimating the problem. Most of my engineering experience lately has been in industrial spaces where the engineers would just throw a third party, warranteed, module at the problem haha
Do you plan to program a web application, calculating everything for lazy people? Perhaps you could earn some money by ads.
Hmmm.. I might do something like that, consolidating all of the online calculators into one. But, I'll have to learn how the existing calculators work first though 😂 Thanks!
Honestly speaking I don't like these gate time tricks since you do create a heater.
Why not to use an external soft-start circuitry?
Why not to modulate these switches instead by compararator and current shunt or just use instead cheap mcu?
You can basically use a damn NTC for this job or 2 of these things starting first via resistor and then normal case.
No, this is not "best". Best depends on your needs, price point, and component availability. If you can't get a component, that is the worst solution. If it is too expensive or overkill for your actual circuit's needs, it is just suboptimal.
I mean, fair. But, it is much much much more efficient past a certain current rating and applies to the same sort of circuits that many other UA-camrs claim is 'THE way' to protect against reverse polarity. Also, to be completely honest, I needed views on this video which I spent around 20 or so hours making. If I didn't name it 'BEST' or say 'Is this the BEST?' no one will click on it and watch it. What would you use instead for >5A circuits? Do other solutions you've seen on UA-cam have the same anti-spark power switch function? I was trying to fill a gap in the search term 'reverse polarity' because no one else (at least popular) mentioned N-channel FETs, which I thought was a no-brainer. Also, for low-power circuits, even though it is likely quite a small difference, the lower resistance of the N-channel FETs can help save more power. No hate or anything by the way. Have a good day 😁