Thanks! 😃

Tell me about it, we've all been there. That's why I always use my current limiting PSU now set to about 10mA initially!

Great point, Chris. Seems crazy that we can put the battery in backwards. Then again, why are audio power supplies (eg for guitar wah-wah pedals) with a negative centre whilst the rest of them are positive centre? Just to blow up our Arduinos, that's why.

Good call! Simple to do too.

Glad you like them! There are 199 videos for you to enjoy at the time of writing this! Plus the beginner Bacon Bytes videos too. Have fun watching them.

You and me both, Hans.

Glad it was helpful!

Glad it was helpful!

Glad you think so!

I appreciate that!

No doubt! Probably wrote it in bed instead of watching TV! Clever chap!

​@@RalphBacon in 1905, Philo Farnsworth was not yet born, so he could not have invented television.

And you've stated exactly the reason, Jeff, why I don't like LOW side MOSFET switches - the ground reference can keep changing and some devices I've had go bananas when that happens (technical term, obviously). No reason not to do this in the HIGH side and keep it all stable.

@@RalphBacon I agree! Although low side switching should be treated as low side switching and nothing else. That means, ground is switched and should never be referenced to any other ground. In a car that would imply that ground is not the vehicle and that means you'll have to isolate devices that are electrical (galvanically) connected. An NPN power transistor in a TO-3 case where the case is the collector is another example. We would like to mount it on a heat sink, and the heat sink to the chassis which is ground. Now, NPN-transistors with the collector to ground is not the best design, to say the least. :-) So we need to isolate it from the heat sink. A pain in the... ;-)

If there is a capacitor on the gate of the protection MOSFET it might hold its charge (or still be charging) when you plug the lights in again. Without seeing your circuit I'm guessing a bit!

@@RalphBacon Thanks for your reply, I think it boiled down to a bad batch of ESP32s

The protection diodes (not on all pins, BTW) clamp the voltage and could allow current to flow to VCC, true. But the body diode is also going to conduct too, surely? After all the Drain/Source of the output MOSFET is directly connected to the pin. Perhaps two ways for the Arduino to be parasitically powered?

Yes, found this out the hard way.

I'm glad you liked this video, Gene, but let's consider your MOSFET. If you were to draw the maximum current of 30A through this device, using a 5v Gate voltage it would have an RDS(on) of 0.0118Ω (max). So the power you are dissipating is I2R = 10.62W. The maximum RthJA in free air is 62°C per Watt so it would reach 620°C - a heatsink is required! The most you can dissipate _without_ a heatsink is max junction temp (175) - ambient temp (25) divided by RthJA so that is 150/62 = 2.4W. Or about 15A. A relatively modest heatsink for a TO-220 would suffice. I'll let you calculate the exact °C/W you need (and be conservative). You could also use a small fan to cool the heatsink, PWM controlled from a Tiny85, now that would be cool (literally).

@@RalphBacon Wow this helps so much.. I think I'm starting to get this. I dont mind a heat sink or even a small fan. I rarely go that high in amperage but if and when I do It is nice to know I would be covered. That PWM code you helped me with would be perfect for that...

In this circuit there is no need for a pullup/down for the Gate as it always connected to GND, Peter, no switching required (or intended). Is that what you meant?

@@RalphBacon duh, of course it's not needed... In this configuration the gate would never be floating, so no need for esd protection. Brain fart moment resolved! Thanks! 😂

Glad it was helpful!

Glad you liked it!

As you say, the body diode is usually ignored, Michael, to a large extent, as it is a by product of the MOSFET design. But it really does need to be considered, not just in designs like this that actually leverage it to provide a useful function!

The image is already burned into my brain from the previous video. I need to lie down...

@Mai Mariarti A valuable mnemonic. Thank you. I just hope my prostrate doesn't start acting up again.

I'm glad you liked this video, it deserved more than what I gave it in the BB series. Glad you got the PCB board too (I replied to your other comment).

Indeed, could save your bacon. Or Bacon. 😉

@@RalphBacon In my case it wood save my Gordons 😂🤣👍🏻🖖🏻

Yes, you are right. We can easily calculate the power as P= I * I * R so if the RDS(on) is 0.06 and we are passing 100A that's 600W! Even 10A is 10 * 10 * 0.06 = 6W which (probably) requires a heatsink. Theory vs Real World application! PS Posted your PCB today.

@@RalphBacon Great thanks. now i'll have to buy some 555, in all my years in electronics design I've never used a 555 timer!!!!! I think they never made a mill spec version ;)

Don't forget it is a *surface mount* 555. Hang on, never used a 555? I thought _everyone_ started off with a 555 and that's what got them hooked? Oh, mil spec? You've let the cat out of the bag now!

@@RalphBacon Cat left the bag years ago, Don't know why I never used one before i started work, but all electronics I've gone since is Avionics or Medical based. so tight on the component specs. will be fun to play with one. since pic's became cheap and easy to get, If you wanted a delay an 8pin pic was the easy option.

Noted, but I have got to say putting anything in the low side is often fraught with unexpected problems because it causes (or _can_ cause) the GND reference to change unexpectedly which can upset a lot of electronics. I've tried it and have come to the conclusion that high side works 100%.

@@RalphBacon Thank you Ralph! I don't disagree however, I am experimenting with a low loss (or zero loss) generator cutout relay for my 1925 Studebaker house-car . The electrical system is 6 volts positive ground and the generator puts out a maximum of 15 amps - going downhill with a tailwind. This looks like an OK application for an N-MOSFET ergo the reason for my original question :-)

Oh well, if you have a positive GND setup then it is totally different. I have never tried this but this circuit shows a possibility (scroll down to figure 3) www.electronicdesign.com/power-management/article/21801509/reversepolarity-protection-in-automotive-design

My pleasure!

Speak to you soon!

My apologies about the timing of this video, Bora! Better late than never! Yes, put a 12V zener and a 100K resistor in the gate to protect it. Running it at 19v is getting too close for comfort! I will check out the LM3999AH too.

Some MOSFETs are designed to work with low voltages, Mark, and have a Gate threshold of between 1V and 2V. But a MOSFET driver can be a simple BJT and then you have all the switching power of a MOSFET at your disposal (assuming the BJT by itself is not an option - it could be just to switch a relay on, for example).

I used to be indecisive but now I can't make up my mind if that's true.

The 100K Gate resistor simply ensures the Zener can clamp the Gate voltage correctly without passing excessive current. I would not skip the zener if there is any chance of the input voltage exceeding the Gate to Source voltage maximum (and as I said in the video, 16v is not uncommon, so you would clamp it to 12v). For the sake of a few pennies, keep it simple!

Glad you enjoyed it!

It should, as long as the MOSFET you choose conducts with a Gate voltage of 3v. Datasheet will tell you this.

@@RalphBacon Thanks! After a lot of research, it sounds like a AO3401 would fit the bill perfectly for an ESP-32 and a NDP6020P can be used for a something drawing a lot of current at 3.3v. Both have a gate voltage maximum of only about 1V.

Yes, they do say the folks in OZ are a bit backward... Oh. Just a joke, Mike, keep sending the files!!!

It all depends on the current flowing but P-Channel MOSFETS can handle huge amounts of current (even if a heatsink might be wise at that point) so I would think that rather than wasting your solar energy heating up a diode it would be much better using a MOSFET and using that energy to heat up the house (or the hot water). Glad you like my videos, Kurt, great to hear from you.

Just saw this! Only 3 months late 🤷‍♂️ The Zener diode protects the MOSFET when it conducting by clamping the gate voltage to the Zener voltage (eg 12v). When the voltage is reversed, the MOSFET is switched off because the Gate Source voltage for this case will be positive (voltage drop over the Zener diode). See section 3.3 for a more elaborate explanation here: www.infineon.com/dgdl/Reverse-Batery-Protection-Rev2.pdf?fileId=db3a304412b407950112b41887722615 It's been a while since I did this video; I hope this clarifies things?

Thanks for that, Ted, I'll scour eBay to see whether there is anything a bit cheaper out there but which still does the job.

@@RalphBacon I'm sure it's out there. That's the beauty of eBay!

When you "dual power", I'm guessing you perhaps mean two 9v batteries in series? With the centre connection deemed 0v and the other two terminals +9v and -9v? I'm thinking that an N-channel connected "backwards" (just as the P-channel was connected "backwards" here) with the Gate at "zero" (actually +9v relative to Drain) with the Source connected to the load. Someone must have done this. Let me see if I can find a circuit.

OK, if you scroll down to figure 3 that is what I had in mind: www.electronicdesign.com/power-management/article/21801509/reversepolarity-protection-in-automotive-design

But you would need _both_ parts to protect the circuit from reverse battery connection. Theory only, never been tried by me!

Nice of you to say, Jeff. Great to hear from you, as always.

Cheap solution: You could add in a single IN4001 diode into the incoming VCC line. However, that would reduce any intended input DC voltage by about 0.7v which you might have to take into account. Better solution: Put in a proper bridge rectifier, it's a single component with 4 leads, so it's easy enough to install. Plus an electrolytic or two! Same caveat applies. Rectified voltage might be well over 12v, so you could also consider a voltage regulator. Yes, it gets more complicated! Or design a detection circuit that blocks AC. It will go through a capacitor so you can detect it that way (after rectification) and do not switch on the main circuit. Eg using my auto on/off circuit with a simple Si4599 dual MOSFET. If you get stuck let me know, I might add this to my video questions!

@@RalphBacon I looked up bridge rectifier and build one with IN4007's that I had on hand. Works great! The 12VoltAC that I have in my shop are as high as 15voltAC, so I am ordering some 7812 Voltage Regulators to help smooth out the voltage. I already use 5V and 3.3V regulators to directly supply MCUs or radios as needed. Thanks for the advice.

It depends on just how much power you have available, Hans. If you have 1.8 volts then an ATtiny13/85 could boot up (no crystal, run from the internal oscillator at 1MHz) and keep monitoring the external voltage until it reaches the magic value (3v3 or 5v, for example). Check out video #171 ukposts.info/have/v-deo/eXGopGeGa6llzZc.html which might have some ideas for you. Let me know if this helps. If not then we might have to go the zener diode /MOSFET switch on route.

So, just to summarise and clarify, Jason, the *VGS* is the *maximum voltage* that the MOSFET can tolerate between the Gate and the Source or it will meet its maker. The *Vth* (sometimes, and confusingly, written *VGS[th]* ) is the *minimum voltage* the Gate needs to START the MOSFET conducting. To ensure it is fully switched on, we usually aim for 5v for TTL (Logic Level) level MOSFETS as otherwise it's 10V which is too much for us Arduino users. The P-Channel MOSFET IRLML9303TR you have picked has a typically higher RDS(on) of 270mΩ (just over 1/4 of an Ω) at -4.5V Gate voltage but that's not too bad for a P-Channel FET. The VGS is 20V and the VGSth is -1.3 to -2.4V so good for Arduino level voltages. Good luck with your project.

@@RalphBacon Thanks!

Great to hear! Personally, I wouldn't put the protection in the negative side as the changing ground potential can upset some electronics but I guess yours is working fine!

I see what you mean. Maybe some sort of bridge control (the type that controls dc motors) would work.

Glad to hear that! 👍🏻

@@RalphBacon i used to put 40 volt schitky diode in series toprotect from reverse voltage..........but as current rises in my Buck converter , diode gets too much heat........this Mosfet solution is very efficient and has no heating............

Only if you know what the input voltage is, and then it becomes voltage dependent. For the price of a zener (pennies) this is input voltage independent and thus a more flexible design.

Well, there is a pre-mounted version with through holes (on a PCB) that I use on my breadboard. AliExpress do them (search for them, as costs - and postage - vary considerably) but I've also found them on Amazon now and again: www.aliexpress.com/item/33010369472.html Are you sure you could not solder these? Use flux first, then solder just a corner of the chip to its pad to get it positioned square, then do the other pins. I use 2x - 4x reading glasses as magnification glasses when doing this too!

Lol, I don't think my hands would be steady enough, I have a hard enough time soldering through hole! Wow the pricing on Amazon compared to AliExpress is crazy! many thanks for the link, getting some on order now. Cheers!@@RalphBacon

As long as the MOSFET can take the voltage (at 4.2volts max I think you're safe with anyone) and the current (at about 100mA for an Arduino ditto). Personally I like the Si4599 SMD MOSFET pair. Use the P-Channel for this and the N-channel elsewhere in your circuit. Just make very sure your MOSFET is TTL or better so that the 3.4volts that the 18650 can get down to before cutting off will still be enough to switch the power on.

It depends on what sort of battery you are charging. It looks like you might be using a Li-Ion (eg 18650 or LiPo battery) in which case you MUST use a proper charging circuit to stop overcharging which would irreparably damage the battery. If you use a protected battery that helps but there are many charging circuit boards from Banggood and AliExpress that will help you. You can certainly use a PSU of 9v as the charging circuit will reduce the voltage via a buck circuit to what is required. It may protect you from reverse polarity protection too. You need to research this in more detail.

I always got told off by my Uncle when I tried to cut my potatoes with a knife. Shock! Horror! Not done in Germany, always use the fork in the right hand. I think that has gone by the board now, though, thank goodness.

@@RalphBacon How do you insult a French cook? By cutting an omelette with a knife.

Putting N-Channel MOSFETs into the low side is a bit tricky, Rene, because it means the GND reference can change. This, in turn, can upset the thing you are running. You won't be able to control the MOSFET from the device you are powering if you do it that way but maybe a simple polarity protection circuit like this _might_ work. Only one way to find out. But I'd recommend getting a few (cheapish) P-Channel MOSFETs!

That MOSFET would work but it is quite chunky (technical term, obvs). You could go to the RS Components site and set some filters on the MOSFETs page that allows the voltage (5v?) and current (5A) and see which ones might suit. But size-wise you can get some very small 8-pin chips that would handle this with ease (eg half an Si4599).

Thanks for responding Ralph, to be honest the 3 pin MOSFET looks much less complicated to wire into my circuit. I would have no clue where to start with the Si4599 how is it that the chips handle it better the the 3 prong version?@@RalphBacon

Well I think i figured it out after looking at the data sheet. Many thanks for your help, that was a really useful video. You gained another sub and a like! Cheers!!@@RalphBacon

Glad you figured it out; that Si4599 (or equivalent AO4606) is just two MOSFETs in one package (as you now know, of course), one P-channel and one N-channel.

I am curious to know though.. does the Si4599 work better than it's larger counterpart that you use in your video? Is it a case of bigger is not always better?@@RalphBacon

Not a single component that does it all, but there are SMD components for all parts mentioned here (MOSFET, Transistors, LEDs etc).

Pretty much, Leroy. But it never hurts to read the datasheet _from the manufacturer who made it_ just to be sure that critical specs (for your project) are all as you would expect.

Let's say it this way: most parts have a "primary spec", their main intended use, that is often the same from different manufactures (say BC546 -> 80V 100mA small signal NPN transistor). It gets complicated with secondary and tertiary specs. And that can be really iffy if you do not use a part for its main/in a simple use case or at the edge of such a secondary spec. Take for example some MOSFET, but your application does not depend on the simple V/A/RDSon spec, but on avalanche energy it can take.

@@janseiffert7799 Thanks, Jan. That helps a lot.

Many good examples of using the more commonly available and actually better for the job N-channel mosfets can be found by merely using Google or similar. Most of them are mirror images of the one that Ralph showcases and explains so well.

Because it is usually much harder to break all ground connections to a circuit. You often have other ground connections to the circuit through a case for example or the circuit is connected to another one and needs a common ground.

Putting an N-channel MOSFET into the LOW side _can_ work too, but I've found that some components/modules are sensitive to the GND reference changing as a result (you're putting in resistance into the ground path, after all). Perhaps modern day components do this better than when I last tried this!

Some devices uses negative inputvoltage for their PSU. like -48V. This is because you have to also require 10-30A.

@@RalphBacon Thanks for yours and other commenter's clarifications regarding low side switching, it has worked very well the few times I've done it, but it certainly wouldn't be suitable for every application, as has been well explained by others on here. I didn't at first believe that misusing a mosfet like this could possibly work, without liberating some quite expensive magic smoke, so I was quite nervous the first few times I 'trusted' the method, and deliberately reversed the polarity on my projects, only for absolutely nothing at all to happen, but even now it still seems like witchcraft compared to the old way of using diodes in series.

Is that 6mA DC or AC? You can put that through an optical isolator LED which will have an opto-transistor as part of the package. This will then go HIGH when the LED is lit. For an idea how this works look at my video #18 (ukposts.info/have/v-deo/nGmWqohkgmx3oo0.html ) on relays. You don't have to use a relay, of course, but you will get to understand how the optoisolator works. This makes it safe to use too. Instead of closing a relay you could turn on a MOSFET that is connected to a digital GPIO pin on your relay that goes HIGH (or LOW, doesn't matter which) when you heater is on. Or just use that relay module for total simplicity! Does this make sense? Watch the video before replying!

@@RalphBacon its an industrial current transformer with 4-20ma dc output, 0-100 amps and the current being measured is about 20 amps.I think I could just put a resistor in series with the loop current and connect that to a digital input. My first plan was to measure with A0. but then I'd have to turn on a digital output at some threshold in order to trigger an interrupt.

In what way do you mean this, Syed? Do you mean directly manipulating the port pins (or the whole port) rather than addressing the pins via pinMode and digitalWrite?

@@RalphBacon if you can use port register rather than pin mode or digital read or write as pinmode consume 250 or 255 bytes

Well, yes, it saves a few bytes here and there. Look at my video #175 ukposts.info/have/v-deo/eaehprChjmeikZ8.html to see whether that helps you.

It is best to switch high side with microcontrollers (most electronics) because otherwise the ground potential keeps changing (relatively speaking). But some P-channel MOSFETs are good value, just required a really good search!

There are modules out there that do this, or you can design your own with a MOSFET. Basically, if the USB voltage is present it will turn off the P-channel MOSFET that allows the battery to connect to the circuit. When the USB voltage is absent, the gate to the MOSFET will be LOW and thus allow the battery current to flow. Because the (5v TTL Gate Voltage) MOSFET is such a low resistance when turned on you lose nothing from the battery power pack (unlike a diode that drops up to 1v).

@@RalphBacon Superb answer, thank you. We are making a cost sensitive product so probably going with your individual mosfet suggestion rather than a module or an ideal diode IC.

A second LED connected the other way around would be simpler.

As I mentioned elsewhere I didn't want the LED diode's forward voltage drop to interfere with anything. This kept it simple (like me). And what Mr. Einstein said.

@@Roy_Tellason yes but the light comes from somewhere else

@@cmuller1441 But if you do that dual led setup before reverse voltage protection and use green + red leds for power indicator. fast way to notice it's connected wrong.

@@LimbaZero The led was supposed to be the load after the reverse protection... Making it work both way illustrates that the reverse protection works... Also LEDs have a limited reverse voltage breakdown. If you apply that to a large voltage you may kill them. You'll have to add a diode in series to protect them... You can find a full bridge in a single chip so my solution is simple. Here you have to add a second led and 2 diodes. This is a mess.

I've seen similar circuits using the negative rail (the TP4056 battery charger effectively does this, as it switches the battery negative in and out of circuit) but you must be careful with the ground potential changing; microcontrollers don't like that one bit.

Is your stepper motor a bipolar or unipolar type? That is, is it being controlled by an SN754410ne H-Bridge controller, or a 2004 Darlington Array (transistor)?

@@RalphBacon It's bipolar, and controlled by an DH860S Driver. Thanks

The controller you are using is quite sophisticated and would not allow you to connect the +ve to the -ve line, although you could do so with a "standard, homebrew" H-Bridge controller if your code is a bit rough (please don't ask me how I know this). For a good (and simple) explanation of how H-Bridges work see here: bit.ly/HBridgeSimple

@@RalphBacon Thanks Ralph, but I was looking for a more general explanation regarding switching a digital pin on the Arduino, that is connected to a negative pin on "what ever" to HIGH. The example I gave was just that, and it works. I was hoping that you could shed some light on "why" this works, and in what sort of situations sending HIGH to a digital pin connected to a negative terminal would be the correct connection.

@@RalphBacon The wiring diagram I have followed is in this UKposts video: ukposts.info/have/v-deo/oYmQaI-Go6ai2n0.html

Would a relay be fast enough to protect the rest of the circuit? In theory, your idea works but I would want to detect (eg on a 'scope) for how long the incorrect polarity was present before the relay kicked in. There we are, your homework for this week!

There is still the voltage drop across the diode to confuse matters - I've learned my lesson, keep it simple!

And many cheap circuits do "just use a diode" but suffer the voltage drop across the diode as a result. So if you're hoping for 5v you have to supply 6v - not the standard phone charger wall wart voltage. Using a MOSFET ensures no voltage drop and no heat generated, all due to the incredibly low "on" resistance of the MOSFET. I don't agree with you, Jer, that the MOSFET complicates everything at all! It's just another transistor at the end of the day. Simples!

He just did. Go back and rewatch.

@imho2278 I asked this question 2 years ago. And someone answered it 2 years ago. I was asking why use a mosfet at all, and instead remove the fet and use a diode to prevent reverse polarity. But I know now that this video is not just about stopping reverse polarity. It is more about how a mosfet works, and voltage control. Im still learning. 😀

@@jerm5772 Also diode allows current only one way, while mosfet is like a switch so current can flow both ways (while providing voltage polarity protection). So for example if there was a motor and it would be hand-driven, then with diode blocking the current the BEMF voltage can be high and would need to be clamped on the board (e.g. using beefy TVS diode), however if you use mosfet then the BEMF voltage will be clamped at battery level and you get some charging. Additionally, in that example, going diode route could require actually quite large diode due to motor current x diode voltage drop.

They obviously watched my video and took it from there, Joey!

@@RalphBacon proper thing haha.

Well, well. With a name like "diode" you would not expect a P-Channel MOSFET to be lurking in there. For others, here's a quick link to one that supports 7A: www.electronicdesign.com/markets/automotive/article/21807223/ideal-diode-supports-battery-protection-to-7-a Pricing for 1000 starts at $2.55 so I suppose individually they are twice that price.

You can do that too, yes, should work.

For one, they are hard to find anymore and; two, they can't handle much current.

David beat me to the two huge reasons! Find me a germanium diode, Clifford, that can take at least 1A and is available at somewhere like DigiKey, Mouser, RS Components! You can only find "new old stock" diodes like the OA81 (ah, memories) from eBay! And that allows about 5mA before it melts. Anyway, even if germanium diodes were available the MOSFET design is still more efficient with next to zero voltage drop across it (as you saw).

That's called the Alternative Diode Use for Fire Making. Well known. 🤷

What's the forward voltage drop of a Bypass Diode though?

So there we are, a Real Life example of how things can go wrong. Sorry for your 3D printer, Norm, but thanks for sharing that rather unfortunate turn of events. This circuit could protect things in future.

Yes, about that. The in-built diode in the MOSFET drops closer to 1V though.

Don't worry about it, I wish you well and hope you get the PTSD treated or medicated so that you feel better.

@RalphBacon I'm doing better the last year believe it or not. I'm finished with trying new meds. I've been on every ssri and antipsychotic med you can imagine over the past decade. I've went with an out of pocket solution which costs me 200 a week but I'm not poor so it's worth it. Workers comp would pay half but they want free access to my personal files and there's alot there of no concern to them and when people have that access across all facets of life they weaponize it. The best man at my wedding and I at his held me under his thumb and also bullied me with it for years until I stood up to him. Something I could have sweet, sweet revenge if I wanted because I do not break confidence and he would at best be without a pension or job. He's such a coward he looks the other way if I see him and must feel guilt as I was told he had me removed from wedding photos. Hahaha how fucked is that. Oops sorry. I am not embarrassed of what I am as I do what I can to help people and I'm fiercely honest and always make amends, even if I've done nothing wrong. Being forging and kind is rare and from personal experience, also not something people look for in a person anymore which I find so counterintuitive to how I was taught to behave. I was recently accused for trying to take advantage of someone when trying to help and as a result gave up alot of money at once which I had obtained by selling an item dear to me as a musician. I'm many things but dishonest as you've seen is not one. On an unrelated note you should develop a pure programming related principles course that you put behind a paywall. Most makers are kings of nested if statements. It reminds me of a time in university when a friend was taking a visual basic course and wrote a card shuffling block of code that was over 300 lines long. When he left the room to shower I rewrote the shuffle code in about 30 lines without any predictable behavior because of how horrible random numbers were at the time back in 01 and he snapped when he came back in because I told him I deleted his code but I had just unchecked the box to make it visible. He ended up getting his first A and became a better programmer instead of being a know it all like his whole family were haha. He thanked me later as I took the time to explain it to him. Teach a man to fish right haha. Pcs were so slow back then and the exe brought his amd 300 to its knees when he hit the deal button. What a horrible language that was. I always wrote everything in borland c beforehand when it came to learning then I'd have the logic nailed down. We hadn't adopted c++ 98 stl at that time in university. All I did at home was code then stopped for almost 20 years. Sorry. My point is most makers suck at coding because it's so easy to stitch together blocks from other's work. I don't do that and you don't either unless it's pointless not to. I know you're busy but people are taught that pointers and such are no longer needed because they write using interpreted languages which were written in c haha. I'd pay for it too. I'm not perfect. You have a lifetime of experience.

In the old days, yes. I've replaced many diodes in friend's equipment when they let the magic smoke out of the diode. The idiot diode relies on the diode junction failing short circuit, which they do 99% of the time, and taking the fuse out before any serious harm is done. It's that 1% that makes this a really poor choice, plus the fact that it requires a part replacement to get the equipment operational again (not necessarily a bad thing, consider some of my friends).

A diode wired in reverse across the supply is a cheap (but bad) design IMHO. It depends on a fuse blowing somewhere (do you put fuses into your Arduino project? I thought not). If I didn't know better I'd say this would be prevalent in Asian designs. And as others have mentioned it then requires a fix! Hmm, return to base, and pay $20 to have a 10c diode replaced. Ironically, in a car you have sufficient volts to waste one across a (high power) diode but even so a MOSFET is a better design!

@@RalphBacon actually its a very American thing. Unless it's just that a lot of American car radios and CBs were Asian imported, that I don't know, but all my CB gear had the diode fitted, and I have repaired many of other people's who were obviously colour blind.

@@lezbriddon Yes, I think it was used in every CB radio made in the 1970's (some great radios came from that era). It wasn't just an American thing though, I've done this fix on more Unidens than I care to think about. I think the reason it was so popular is that it only cost the manufacturers a few cents to implement, whereas a more sophisticated approach that didn't limit voltage or current was very expensive in 1970. @Ralph S Bacon Actually the fuse usually blew before the diode did, so quite often the user could figure it out and get it working themselves. For those that hooked it up in reverse with no fuse, at least the equipment was repairable and not a total write-off. Selection of the fuse was critical - pick a value that was too high or a 'slow' version by mistake, or simply have some really bad luck and you were in a world of hurt. Very often when I did this repair I would change the blown 10 amp fuse for a 2 amp fast one, which is all a 15 watt radio really needed.

Thanks for posting, Yogesh.

Hang on, Flemming, I'm not a god? And not a wizard either? Gee, that sucks. 😔 I guess I'm just a lowly UKposts maker. Oh. 😀 That's alright then! Phew! It did teach me a lesson though, to watch my own video through! And not to rush things.

@@RalphBacon Sorry to ruin your "self image", or selfie as they might call it today 😂