Spent so much time looking into these modules and chips, incredibly happy you made this video! The only thing I wish you did is compared reverse polarity protection vs reverse current protection (two important aspects of diodes, but I think all these modules perform both) and how certain "ideal diode" P-MOS circuits only perform reverse-polarity protection.
One thing I'm curious about is the frequency limits, or the amount of time it takes to switch the MOSFET on and off if the input voltage polarity reverses/switches off. You'd think for such large MOSFETs it would be pretty slow since it seems like the available gate drive is quite low, particularly the ideal diode type. Not necessarily an issue for 50/60 Hz power, but repetitive transients could be interesting.
The MOSFET can switch very fast but because the control circuit is design for low power consumption its ability to turn the MOSFET on and off rapidly will be the real speed limit. Versions with a 3 wire which has a enables them to operate continuously could be design to switch much more rapidly, but generally 10's of KHz would be realistic which is not a challenge at all for the MOSFET.
Hi. Great breakdown, I can now see I got a bit lucky using some of these and not burning things down, because I did not understand entirely how they work. Can you please explain a bit more about the maximum voltage difference between anode and cathode? If it is stated at 40V, it means i can not connect a 48V battery charger to a battery through it, because when the charger turns off, there will be 48v difference between anode and cathode? In fact my use case is to connect two boost converters in parallel to transfer power from a 12V battery with solar charger, to a 48V battery, and I intend to use ideal diodes to isolate boost converters' outputs from each other. Thank you
Is the 2% recharge time enough to cause issues with electronics downstream? Would putting a smoothing capacitor inline downstream alleviate that? Also, how would something like this behave if it were to fail? Would it fail open-circuit or closed? How would you tell if it failed full open? Would a TVS to ground ahead of it be helpful in surge events to not fry the ideal diode? So many questions!!! Very cool electronics!
Can you parallel them to gain higher amperages? More parts = more failure points? How do they perform at higher amperages and what happens if they go over amperage,
Any follow-up video would be interesting to compare the reverse leakage (if any) and the quiescent current to ground. Don't want to be leaking the battery into the "diode" just to hold it in an off state.
I'd like to make use of these modules, but there is limited data from the sellers, some are far from ideal. Looks like most can be used in a battery charging application, but for reversed supply polarity protection their operation is not clearly specified. It seems the ones with ground reference are the ones to choose for stability, but leakage and current consumption are also factors to consider.
Many of the "ideal diodes" (quite the misnomer) provide reverse polarity protection, but *not* reverse flow protection. The latter is often the much more important function, especially in PV/battery charging situations. You can do "reverse polarity protection" by just making sure you connect wires right (so you don't need a diode at all), but you can't prevent reverse flow by "just making sure". If in doubt, don't get them.
I think you're right. All the chip pins (that I can trace) line up with the LTC4412 datasheet. Interesting that it's not called an ideal diode. It's quite hard to find in a Google search.
Hi, great video as allways. Does this mean that if the diode feeds a capacitor which maintains the voltage on the output, the chip cannot recharge the capacitor and is this a problem?
You know those gray square rectifiers with 4 connections that turn ac to dc, could one of those be use with a bunch of solar panels wired in parallel going to the ac side of the rectifier, would that stop power going back to the panels at night
That's just a device with four regular diodes inside. You could use it, but you have all the disadvantages of a normal diode. Actually, I think you'd be going through two diodes each way, so you'd have double the disadvantage.
Yes, you'd be better off using just a single appropriately-rated diode. The only reason to use a bridge rectifier for DC would be if you wanted to allow connecting the input with either polarity.
Diodes pass Voltage not current, unless under load, there is no current. However, the diodes have a Current limit. Current does not flow, it is static in any circuit, Current creates a Magnetic field which can be measured by a Coulomb Meter or a Resistance shunt, where you read a Voltage to calculate the Current present under load. The subject is of course up for debate, but without Voltage and a load, expressed in Watts, there is no Current present, it does not suddenly start flowing, as there is no place for it to come from.
Not to be picky but current is the flow of electrons so current flows from the cathode to the anode, not the other way around. Yes, current flows counter to the arrows on schematics.
No, that is not how the term "current" is typically defined or used in modern electronics contexts. In circuits, electrical current is pretty much universally defined as flowing from positive to negative. The fact that it happens to be _caused_ by a physical phenomenon that involves some elementary particles moving the opposite direction is basically irrelevant, and does not change what the term means or how it should be used when working with electronic circuits. (And really, you could just as legitimately say that "current is the flow of valence shell vacancies" (aka "holes"), in which case it would physically be from positive to negative too. It's all just a question of perspective.)
@@foogod4237 So you are right except I am right about the physics. You are going to get into problems later on if you have the very basic physics of things wrong. So, in a rube circuit, the plate is the source of the electrons and the cathode is what catches them at the other end? And they heat the plate and not the cathode? Or do electrons go the other way in tube stuff? You are really not doing your further understanding of how devices work until you understand how current flows, and I am sorry to say, that it flows from the cathode to the anode. It is just that simple.
Very interesting comparisons and video, It looks like the P channel modules do not have the switching issue between the internal diode and mosfet of the N channel types, which must create a lot of supply line switching noise. Where as the internal resistance of the P channel mosfets is higher, so dissipate more power than the N type. It worth noting mosfets have a positive power coefficient so the hotter they get the less they conduct, so in theory they will self regulate.
It's only the 74610 controller that switches the MOSFETs off every 2.66 seconds. The LTC4359 controller drives N-channel MOSFETs and doesn't do the switching (but it does need a ground reference).
Odds I will ever use an ideal diode in a project I actually get too, 20:1 Odds I will watch some more videos on this channel instead of doing projects, 1:1
I hope we will see a final "versus table" and some tests to finally one/two/three models recommended because of the price. Thanks a lot for your work!
Spent so much time looking into these modules and chips, incredibly happy you made this video! The only thing I wish you did is compared reverse polarity protection vs reverse current protection (two important aspects of diodes, but I think all these modules perform both) and how certain "ideal diode" P-MOS circuits only perform reverse-polarity protection.
One thing I'm curious about is the frequency limits, or the amount of time it takes to switch the MOSFET on and off if the input voltage polarity reverses/switches off. You'd think for such large MOSFETs it would be pretty slow since it seems like the available gate drive is quite low, particularly the ideal diode type. Not necessarily an issue for 50/60 Hz power, but repetitive transients could be interesting.
I’d like to see this too…
The MOSFET can switch very fast but because the control circuit is design for low power consumption its ability to turn the MOSFET on and off rapidly will be the real speed limit. Versions with a 3 wire which has a enables them to operate continuously could be design to switch much more rapidly, but generally 10's of KHz would be realistic which is not a challenge at all for the MOSFET.
Hi. Great breakdown, I can now see I got a bit lucky using some of these and not burning things down, because I did not understand entirely how they work. Can you please explain a bit more about the maximum voltage difference between anode and cathode? If it is stated at 40V, it means i can not connect a 48V battery charger to a battery through it, because when the charger turns off, there will be 48v difference between anode and cathode? In fact my use case is to connect two boost converters in parallel to transfer power from a 12V battery with solar charger, to a 48V battery, and I intend to use ideal diodes to isolate boost converters' outputs from each other. Thank you
I'm definitely not going to get a load of these straight away. Probably have my tea first
Is the 2% recharge time enough to cause issues with electronics downstream? Would putting a smoothing capacitor inline downstream alleviate that? Also, how would something like this behave if it were to fail? Would it fail open-circuit or closed? How would you tell if it failed full open? Would a TVS to ground ahead of it be helpful in surge events to not fry the ideal diode? So many questions!!! Very cool electronics!
Can you parallel them to gain higher amperages? More parts = more failure points? How do they perform at higher amperages and what happens if they go over amperage,
Any follow-up video would be interesting to compare the reverse leakage (if any) and the quiescent current to ground. Don't want to be leaking the battery into the "diode" just to hold it in an off state.
I'd like to make use of these modules, but there is limited data from the sellers, some are far from ideal. Looks like most can be used in a battery charging application, but for reversed supply polarity protection their operation is not clearly specified. It seems the ones with ground reference are the ones to choose for stability, but leakage and current consumption are also factors to consider.
Many of the "ideal diodes" (quite the misnomer) provide reverse polarity protection, but *not* reverse flow protection. The latter is often the much more important function, especially in PV/battery charging situations. You can do "reverse polarity protection" by just making sure you connect wires right (so you don't need a diode at all), but you can't prevent reverse flow by "just making sure". If in doubt, don't get them.
Fascinating. Thanks.
Julian, the top-right Blue-colored PCB module may be using LTC4412 (marking LTA2).
I think you're right. All the chip pins (that I can trace) line up with the LTC4412 datasheet. Interesting that it's not called an ideal diode. It's quite hard to find in a Google search.
Thanks Julian
Hi, great video as allways. Does this mean that if the diode feeds a capacitor which maintains the voltage on the output, the chip cannot recharge the capacitor and is this a problem?
As long as a reasonable amount of current flows, the 74610 will work as intended.
2:20 - it is always a joy to snip of these thick legs after you soldered them😀
II think the best is to use a ideal diode with a GND connection.
I want to replace the 3 diodes behind the 60V PV-module with this type of diodes.
You know those gray square rectifiers with 4 connections that turn ac to dc, could one of those be use with a bunch of solar panels wired in parallel going to the ac side of the rectifier, would that stop power going back to the panels at night
A bridge rectifier you mean? You'll only be using two of the four diodes and they could get pretty hot.
That's just a device with four regular diodes inside. You could use it, but you have all the disadvantages of a normal diode. Actually, I think you'd be going through two diodes each way, so you'd have double the disadvantage.
Yes, you'd be better off using just a single appropriately-rated diode. The only reason to use a bridge rectifier for DC would be if you wanted to allow connecting the input with either polarity.
Perfect
I wonder where you store all the mailbox stuff (you got over the years) in your tiny British house?
In a large 3-dimentional matrix. Finding stuff is near impossible.
He sends it to Dave Jones in Australia of course! Julian's handwriting is so bad a lot of it ends up in Austria though. 😼
Diodes pass Voltage not current, unless under load, there is no current. However, the diodes have a Current limit.
Current does not flow, it is static in any circuit, Current creates a Magnetic field which can be measured by a Coulomb Meter or a Resistance shunt, where you read a Voltage to calculate the Current present under load.
The subject is of course up for debate, but without Voltage and a load, expressed in Watts, there is no Current present, it does not suddenly start flowing, as there is no place for it to come from.
Not to be picky but current is the flow of electrons so current flows from the cathode to the anode, not the other way around. Yes, current flows counter to the arrows on schematics.
I'm a conventional kinda chap ;)
No, that is not how the term "current" is typically defined or used in modern electronics contexts. In circuits, electrical current is pretty much universally defined as flowing from positive to negative. The fact that it happens to be _caused_ by a physical phenomenon that involves some elementary particles moving the opposite direction is basically irrelevant, and does not change what the term means or how it should be used when working with electronic circuits.
(And really, you could just as legitimately say that "current is the flow of valence shell vacancies" (aka "holes"), in which case it would physically be from positive to negative too. It's all just a question of perspective.)
@@foogod4237 So you are right except I am right about the physics. You are going to get into problems later on if you have the very basic physics of things wrong.
So, in a rube circuit, the plate is the source of the electrons and the cathode is what catches them at the other end? And they heat the plate and not the cathode? Or do electrons go the other way in tube stuff?
You are really not doing your further understanding of how devices work until you understand how current flows, and I am sorry to say, that it flows from the cathode to the anode. It is just that simple.
Very interesting comparisons and video, It looks like the P channel modules do not have the switching issue between the internal diode and mosfet of the N channel types, which must create a lot of supply line switching noise. Where as the internal resistance of the P channel mosfets is higher, so dissipate more power than the N type. It worth noting mosfets have a positive power coefficient so the hotter they get the less they conduct, so in theory they will self regulate.
It's only the 74610 controller that switches the MOSFETs off every 2.66 seconds. The LTC4359 controller drives N-channel MOSFETs and doesn't do the switching (but it does need a ground reference).
Need pwm 5 video
Teao2diode?,” Anyone?”Hello Julian
Odds I will ever use an ideal diode in a project I actually get too, 20:1
Odds I will watch some more videos on this channel instead of doing projects, 1:1
wtf,no links in description?
😪😪😪 Search.
Yeah, here's the link www.aliexpress.com/w/wholesale-ideal-diode.html?spm=a2g0o.home.search.0