Hi. Just a quick aside on the power supply: a few people mentioned what happens if you say charging a battery with this power supply and you disconnect the input voltage while you still got that battery on the output? I.E You're feeding a voltage back into the power supply and you switch off the input. What happens? will the regulator blow? and uh, this is, uh, quite a common scenario for uh, traditional linear power supplies. And let's take a look at it.

If you got your voltage Reg Regulator Like this, you got your input voltage. Doesn't matter where it comes from, you got the output voltage of your power supply. If you're feeding back in a supply from a battery, say you're charging it or something like that and it's going back in. What happens if you've still got the output battery connected, but you disconnect this input? Supply And well, you can blow up your regulator and a lot of uh, uh, Power suppliers will typically or traditionally have a reverse protection diode on there so you don't uh, blow the pass transistor in the voltage regulator and my design doesn't have that.

and uh, I didn't include that because the none of the app notes for the Lt380 actually show um, that including their own lab power supply circuit. So I figure it's a new design, should be robust enough that it doesn't need that. But H Let's test it all right now. What I've got here is I've got my power supply circuit built up.

not that we actually need the circuit itself because we're just testing the regulator effectively. Um, but I've left the circuitry there as is what I've got is I've got it hooked up to two external uh Power Supplies up here. Uh, this one here is the Uh input uh Supply and this one is uh, feeding a voltage back out into the output of the voltage regulator and what these Uh meters are measuring. I've set the output voltage to 5 Vols here.

So that's the output voltage of our regulator. This is our input voltage of our regulator and this is the Uh current being fed back into the input from the external Uh power supply just so we can monitor how much current's actually going back in when we switch this thing. um, on or off and play with it now. of course, if I switch off my input voltage, I've got no input voltage here and my output voltage is nothing.

Okay I Not so I'm not um, actually feeding in the output voltage. All right. So what I'm going to do now is I'm going to switch on this output voltage. So we're feeding a voltage back into the regulator and this will monitor the current.

Remember, this is monitoring the output voltage of the regulator, so we should be able to force that output voltage higher than the 5 volts that it's trying to actually regulate. So I'll switch that on now I've got it set to 5 Vols so it doesn't really change at all because they're the same voltage. But let me wind that up a bit and uh, let's see if we can. There we go.

Bingo I can actually I'm driving that output voltage from my external Supply My external Supply is six. Well, it's saying 6 point. It's measuring 6.25 and sure enough, it is there. There It is 6.26 So I'm feeding in that external Supply Now what happens? Let's actually, well, let let's just leave it there.

Let's say it's 6.2 Vols doesn't really matter. and let's switch off our input voltage and see what happens here. We go and it's feeding 8 milliamps back in. Uh, but it's doing.

It's doing nothing. There you go, we're still getting an input voltage being fed back through the regulator. So even though our circuit is actually disconnected, if we have a look at our circuit here, this switch is actually disconnected. And we're and we're measuring our input voltage.

which is 5.57 because it's being fed back through the regulator. Remember there, there is no diode there at the moment. Okay, so we're feeding the voltage back in and well, let's see if that has killed it. So I'll switch my input voltage back on I'll switch my load voltage off.

No, it survived that quite well. Not a problem whatsoever. Let's wind the output voltage all the way up to okay. I've got a 12vt input voltage.

Let's wind the output voltage up to 12 as well. So there you go. I've got 12 volts output voltage and let's switch off. Let's switch that on and force the output voltage up to 12 volts.

Bang! We've done that. It's only feeding, you know, 10 milliamps. Uh, back into the thing. But let's switch our input voltage off and bang.

And our input voltage is, uh, still. 11.3 But I've switched it off and let's see if it recovers. I'll switch the input voltage back on, turn off the output and it's recovered. No problems at all.

So we haven't killed this thing. If you switch off the input voltage, not at all. all right, it's time to get a bit drastic. Instead of just switching this input voltage off, let's actually short it out to ground and see what happens.

So our output voltage is at, uh, 12 volts. Let's force that on again. Bingo There we go. We've forced our output voltage to 12 volts with the external Supply and not only will I switch off the input voltage, but I'll disconnect the leads from the power supply.

There're the input leads and I will short them and we'll probably see an increasing current here. Who? Yeah? Bingo Woo, Oh yeah, there was some smoke somewhere. You better believe it. Let's actually, um, switch the input voltage back on and see if we killed the regulator.

Nope, it's still fine. It's survived that. No problems whatsoever. Bang.

We're forced our output voltage to 12 Vols and this time I'll probably hold it and see what happens. Oh yeah, look at that. th oh, there we go. What smoked? Check it out.

and the poor little sucker there never stood a chance. Little quarter watt resistor. so you actually saw it there. It was getting over 3 amps feeding back into that regulator.

All of that current must have gone through the Uh voltage regulator. There's nowhere in the Lt380 there's nowhere else for it to go. So let's uh, see if it's uh, if it's killed it or not. I've just disconnected the output and yeah, something's going on, but our current shunt resistor might have blown.

Let me check that and uh, see what's happening. Let's measure our current shunt resistor should be 1 ohm but what? 7.5k dead as a Dodo Okay, I've replaced the uh shunt uh with a short I didn't bother uh, putting a 1 ohm resistor back in and let's switch the power input. uh Supply back on and hey, no problems whatsoever. The regulator survived D just fine.

So much for blowing the thing that was over 3 amps if I recall from that meter. going back through the LT 380 and there's nothing wrong with it at all. It is a really quite a robust device. uh you? I'd probably have to take it to its extreme limits to actually see where it blows.

Well, it seems we may have actually done some damage to it. I Um re hooked it back up and tried to play around with it again and it was uh, it was kind of working but uh the out put was about 1.8 Vols or there about tracking 1.8 volt higher than the set pin voltage. So I replaced the regulator and um and now it's it's just fine. So um I'm not sure what happened there because it you know it was working okay to begin with but then seems to have died.

So bummer and there it is. It actually has a low of 3.86 Vols and it goes up to uh a high of uh 7 odd Vols and that should be 0 to 5. So that regulator unfortunately we have killed it w but that was a pretty uh extreme case though I mean uh, you know short In short in this input directly um here was a little bit extreme. so if we did actually have the diet in there, that would have shunted all of the current through the diet.

and if you got a big beefy you know 5 amp diet in there, it should actually handle that uh while actually preventing you know, like a maximum voltage of like a Volt across the Uh regulator there. So do we actually need a diode across here? Well unless you uh, somehow, uh, short out this or connect or this is connected through to your supply and you force your voltage to a much higher uh voltage which then uh can cause enough current to flow through the Uh device and into your low impedance uh Supply at the front end to actually kill the device, then the answer is no, not really, it doesn't. It seems to survive quite fine if you simply just, uh, switch it off like that. And of course, if you going, if you're feeding in, uh, big voltages from outside, then you've got other things to worry about.

like you know, uh, maximum voltage on your Um, on your Uh opamps and your other components that are supplied from this positive input voltage here. and well, if you wanted to solve that, then you'd have to have another regulator on the input here just to power all of your other stuff stuff. So you know I don't know as it as it stands, the answer is not really. and if you remember the internal block diagram of the Lt380 here it is.

It's got the Um Standard Npn series pass element here. so when you feed in a uh a voltage from the output and your input is either floating or it, uh, grounded grounded potential, which it, uh, certainly could be If this capacitor's discharged and you input a voltage in here and you've got significant bulk, uh, capacitance on your input or you're uh, clamping it deliberately. uh, clamping it low for some, um, overcurrent reason or something else, then, uh, you can get a reversed bias situation in your Um Npn transistor. Here you can, uh, reverse Vus your basem Junction and you can, uh, actually blow that series past transistor.

not to mention uh, any other Uh internal circuitry in there. And if we take a look at a Um Uh standard 78 Series 7805 voltage regulator, you'll note it's basically the Uh same thing. There's your Series Pass Element Npn right there and this is like a Darlington configuration on the output here. and this is specifically the Uh Texas Instruments data sheet for the 78 XX series and it specifically tells you: Reverse Bias protection.

Occasionally, the input voltage to the regulator can collapse faster than the output voltage. This can occur for example, when the input Supply is crowbar during an output over voltage condition. If the output voltage is greater than approximately 7 volts the emitter base Junction of the series pass element internal, or if you're using an external Um series pass transistor for extra current, it can break down, it can be damaged Uh. And and to prevent this, it's a standard industry practice to put a Um reversed biased Uh diode in there, which actually limits the voltage across the input and output differential.

on the Uh voltage regulator. It's the same for the 7805 or for the LM 380 and Uh Lt380. Sorry. So whether or not you actually uh, on this power supply design or your own power supply design, put that reverse bias diet.

it's up to you I've it I Was able to damage an LT 380 but then it. but it actually recovered after that. um, serious 3 amp? Uh, short in. So I'm not actually sure what happened to it after that? I'd have to do further testing more methodical to find out exactly um, how to actually break the thing, but it seemed actually to survive a fair amount.

And if you don't, uh short. uh V in, then uh, it survived 40 volts. No problems whatsoever. Whether or not your other circuitry connected to there would survive 40 volts.

Well, you know that's up to your design. Uh, you could limit it with um Xeno diode clamping or something like that. Or you could have a Xena diode clamping on the output as well. If your output, if you got a 0 to 10 volt um, bench power supply, you might put say a 12volt xener on the output.

Uh, so that you can't get uh, dangerously high voltages. So you're going to blow your Zena before you're going to uh, or the Zena should clamp Anyway, you might blow it. uh, depending on the capability of the supply you're hooking up. but it's really hard to cater for all types of scenarios.

If you're going, start applying uh, you know, large voltages to the output of your power supply. Well, you're eventually going to break something. So whether or not you put it in I don't think I'll have it in my design. Um, but if uh, people want to add it, that's up to them because configurations might be different or a better solution might be one of these Tvs's or a transient voltage suppressor.

They're also called trans Orbs and uh, all sorts of other trademark names like that from various manufacturers and they're basically a uh Xena Diode a high energy Xena diode to protect you against uh, overvoltage conditions and uh, they're They're really nice. So if you had say a 0 to 10vt uh power supply, you might say use a 12volt uh, one of these to not only uh, protect you from transient overvoltage conditions feeding back into your supply, but also uh, basically for uh, your negative protection as well. Because if we take a look at our circuit here this, D1 here this diode. Um, then if you replace that with a TVs instead of a shoty DI then you get uh The Best of Both Worlds high high voltage uh over protection and also reverse voltage protection as well.

Well worth using one of those. and if we take a look at a data sheet for a typical uh MPN transistor, in this case, we've only got a low power one. We've got a PN W22 also known as a two N22 which you're probably uh, familiar with. Now the thing we're concerned with here is the uh emitter base uh, breakdown voltage which they specify and you notice that it's not base emitter breakdown voltage, it's emitter base.

which means that it's actually uh, the the reverse voltage applied to the standard base emitter. um Junction for an Npn transistor. In this case, it's only six volts and as you uh, saw in the uh, LM 7805 uh data sheet, that was a very similar value as well. They said uh, you know, order of 7 Vol so that, uh, that's really going to be the killer uh inside the device.

Okay, one last thing. I've now got the regulator circuitry uh, separated from the rest of the circuitry. so I don't blow up the rest and I'm going to uh, wind the wick up on the output voltage. it's currently um 12 Vol I'll wind that up and I've got a fix resistor there actually with the 10 microamp through to give me a 5V uh out.

So there's really nothing else connected on the input side and I'm going to force the output side up. so let's switch that voltage off. There we go. and let's switch the output voltage on there.

It is. Okay, so we're feeding in our output voltage and let's wind that up, shall we? Let's wind it up. and uh, as you can see, it's still. you know there's there's really nothing flowing into that regulator.

So I'd be surprised if there's any damage happening in there at all. There's only a milliamp 20 volts. Ah, we're really feeding a lot. Well, 1 milliamp.

We're really feeding a lot of voltage into this uh output. uh pin. Let me tell you. 30 volts.

My Supply only goes up to 40 I think 40 is the maximum voltage for the LT 3080. Don't quote me on that, but let's wind it up and all the way to what my Supply is capable of. 41 volts. Okay, there you go.

Let's wind it back down, switch it off, and uh, let's switch the Uh power back on and see if we've done any damage at all. None whatsoever works just fine. It's a nice, robust little device. Nice I Like it catch you next time.