Hi. Now it's time to build up the constant current power supply which I've described in the previous two Uh episodes. I've got it up here on built on the breadboard and we'll take a look at it in depth and we'll see if this thing actually works I hope it does I expect it to Um and if not, well, I guess we'll have to fix it. I've got uh, current and voltage? uh pots here.

these are only cheap crap. uh single, uh turn ones I don't have uh, any free uh 10 turn ones to hand so these will do uh. good enough for testing. I've got my constant current load over here which you've seen described before that'll help us test it.

I've just got a resistor for a dummy load here which we can attach and do some testing. I've got the Um Lt380 mounted on a heat sink here, which is the same heat sink I plan to use in my final design. Uh, actually. and uh, let's go let's see if this sucker actually works.

And here's the circuit we'll be using today. It's pretty much identical to the Uh whiteboard one that I've done in the last two videos. I've added a couple of things I've added a limit lead on the output of the Uh current limit comparator down here. so when the output of the current limiter goes high and turns on the transistor and goes into current limit mode, it also lights an LED so we can visually uh see and if that happens really fast and you'll see, should see some flicker on the LED there or something like that.

um I've added a 2.5 volt voltage uh reference because a power supply is no power supply is only as good as its voltage reference because that's where additional uh noise will come from. That's where it gets its absolute Valu from I've used an Lt19 here slight Overkill in this instance, but uh in the final design um, it's like a 0.2% uh 2.5 Vol voltage reference but it will allow us to uh Drive uh the uh voltage reference input of a pick microcontroller or something like that which which we can use use for our digital panel metering and we'll have to go into that in some uh further video. but today you can just use an LM 336 or something like that 2.5 volt voltage reference. Um, we've got another dropper resistor in here to give our current range from 0 to 1 amp and our voltage here I've added uh, some gain here with these two resistors.

Um, so our gain should. um so our final voltage that multiplies um, that 2.5 by a value of uh, that has a gain of 2.2 So that will be our maximum voltage output and up here for our Uh current uh, shunt measurement I've um, actually matched these 10K resistors here. So I've picked ones out of my bin and I've physically measured them on the multimeter and actually matched them so they're within I Don't know I'm not sure what it was, but they're pretty close. sort of.

you know. Plusus two least significant digits or something. So much better than the Uh standard 1% tolerance, which is what these resistors uh marked as. And the opamps we'll be using today are: TLC 2272 You could use a quad Op amp which is the 2274.

This is just like a cheap, uh, a cheap sub 1 molt offset. uh, opamp. You could use like a, you know, a generic Jelly Bean Lm32 4 or something like that. but uh, realistically, when you're talking about the current shut measurement up here.

the voltage offset's probably going to be too, uh, too high for that. So you want to use a semi-precision type device here? unless you want to tweak these values and put pots in and tweak it, eh. gets nasty. We don't want to do that and we'll just go through the layout very quickly here.

I've got my uh, positive rail up the top here I've got my negative rail all along the bottom so all that bottom stuff is all ground. Um, this is my uh, voltage reference over here. This is my uh, constant current load the LM 334 here and uh, where that's that's our um, our current limiting uh transistor and uh, that's Ic1 and that's Ic2 and that's our current shunt resistor there. it's only a quar watt one.

1 Ohms I Don't have a 1 watt or 3 watt version I Have to go to J car quickly to pick up a higher wattage one to measure some higher power stuff. but uh, there's our output um, caps and uh, there's our cap which by passes the adjust pin on the Lm380 and well, that's about it. Really are bypass capacitors on the Ic's there input bypassing on the Lm308 and that's about all we've got our output LED here for the uh current limit so that should light up and well, let's power this thing up and see how she goes. All right, let's power this thing up.

I've got it powered from about 8 volts from an external Supply over here cuz we're only going for a 5.5 volts are there output range today, so let's flick it on and uh, see what? We get our meter over here. this is the output, uh, voltage. We got nothing. Oh, our current limit leads on.

Let's ah yeah, the current limit. The current limit's all the way down. that's why. So if we turn our current limit up, let's turn the current limit up to maximum so we don't muck around.

Our output voltage is 5.5 volts. Yeah, that's because the pot the voltage adjust pot is at maximum. it seems to be working I've got my uh, constant current load set to minimum down here so it shouldn't be doing anything much if anything. and what do we go down to? hey, we go down to 0 volts what do you know? And there we go.

that appears to be working just fine I Like it. I'm surprised that it does actually go down to zero there. I Think our constant current load could have something to do with that? So let's turn it up. I'm drawing 60 milliamps.

Let's say draw 100 milliamps out of this thing. There we go. 100 milliamps and uh, let's adjust that voltage down again and see what. See what happens here? We still got.

Oh no, we still. it's still going to allow us to go down to zero there. I'm going to disconnect this constant current load and without the load on there, it only goes down to 0.7 volts. And that's pretty much what I expected.

Perfect. And let's actually remove that Lm334 current source and see what we get at the minimum. There we go, it only goes down without that 1 milliamp current. Source it only goes down to 1.59 volts.

That's cuz there's no load at all on the output except the multimeter which is a 10 m ohm input impedance. Now let's check the output on the scope to see if it's uh, clean to see if it's oscillating or not. Uh, I've got set to 1 volt per division. We've got our maximum 5.5 Vols output and let's wind that down.

and uh, I don't see any resemblance of any oscillation at anything during that range. Once again, we've got, uh, just the minimum, uh, load there. There's the 0.7 Vol minimum. but let's switch that to AC cing on the input so that, uh, we can see wind it right down to 10 molts uh per Division And as you can see, it's nice and clean.

Okay, so this is the output noise here. as you can see: 10 MTS per division. No problems whatsoever if we wind it right up. of course the bandwidth of the scope.

it just gets noisier and noisier. And I haven't uh, probed it. you know it's a breadboard so we're just getting ballark sort of noise thing. Noise: Um, figures here and that was at that's at minimum voltage.

Okay, that's at minimum output voltage minimum load. Let's go: maximum output voltage minimum load. There it is there. Uh, no problems at all and it jumps around of course cuz it's AC coupling.

so you'd expect it to jump around until it settles through the AC coupling input cap and I like that. No complaints at all. It looks like it looks very clean as you'd expect from a linear voltage regulator with a Uh rated 40 microv volts. um output noise.

In fact, I don't have the gear to measure 40 microvolts output noise I'd have to roll my own amp and do all sorts of other things and ah, go too hard and if you actually want to see the difference, okay, I'm probing the output at the moment. Uh, we're at 10 ms per division. let me probe the ground. So I'm leaving the ground hooked up.

but I'm now probing the ground of the circuit as you can see virtually no difference as you'd expect with 40 microvolts output noise RMS Now let's do the same thing again at 250 milliamps cuz that's in theory, the limit of our quar wat resist resistor there will now be dissipating A4 Watt in that 1 ohm current Sens resistor I don't want to go over that until I get a bigger current sense resistor. So we're drawing 250 milliamps and uh, measuring the output noise. Again, let's go down to minimum. There you go.

even down at 0.0 This is our output voltage of course and and no problems whatsoever. And let's go right up to our 5.5 volts. No, it's it's perfect. It's working exactly as I expected in terms of Uh output voltage regulation.

I Like it. Okay, now we'll check our basic current limit. Uh, up until now we've had our current limit up on uh, maximum and we've had our well, let's set our way. We're going to test that is to set our output voltage to maximum.

We're getting now 5.5 volts out and let's uh, wind. We're drawing 250 milliamps constant current on the output and let's wind wind our pot down until we start. We should. If it enters current limit, we should see that lead turn on and this start to drop as soon as we reach the limit.

Turn it down, turn it down. come on. Oh there we go There we go. Just the point it starts to drop out.

Now if we measure the voltage on that pot, it should be 250 250 m volts because if you look at our circuit here our here we go If I get it in here. uh our current adjust pot down here to have our current limit at 250 mamps, we expect 250 MTS on that um pot there which sets the threshold where it goes into current limit mode. So if I tap the measure the center point of that pot. it should measure just either on or slightly over 250 MS 262 bang And when we turn it down it's got to 250 3 There's a bit of error there.

it started to go into current limit mode. got a little bit of error there, but basically it's working. So let's set this to 100 I'm not looking at the other one and we should be pretty close to 100 down here and we are. You know there's going to be some error here.

you know? 50 milliamps 55 These basically should uh should match and they do So Constant current limiting is working. Okay now what I'm doing is uh going to probe the output pin 7even here of our current sense amp and let's have a look at what that's doing when it enters current. uh output Uh Current Limit Mode I've got it set to 1 VT per division here. And of course, it's not doing anything at the moment because it's It's not actually current limiting.

But if we turn this pot down slightly, it'll enter Wa. There we go bang. It's just looks like it's jumped from there up to just under a volt 0.8 volts. And let's continue to turn that pot down and lower the current.

And yeah, as you lower it, it looks like it's going up. And that's why our lead doesn't turn on cuz there's not enough voltage to switch on the lead. Only when it gets to the you know, only when it gets to about 2 volts is it going to turn on our LED there. Now, clearly what you're seeing here is this: Although it's still acting like a comparator, it's doing the job of a comparator.

It's not strictly just going between one and zero in the output. That's because it's in the feedback loop here. via the Uh current limit transistor via the LT 3080 via the you know current s It's all happening all in real time and it's a linear. It's working in in a linear type Uh region now in the feedback configuration.

Now, if we actually disconnect the transistor from here and actually remove the loop there, the control: Loop then uh. we'll find that this Uh device will actually switch like a comparator as you'd expect when we adjust the control pod. so this is with the transistor in place, it jumps up to there and then it does. You know that's where it entered current Limit bang and then it's Works in a linear type region from then on.

and if we disconnect the base, that transistor gone. there we go. but we're still measuring the output bang. Our LED is on and it switches bang like a true comparator cuz there's no feedback happening there.

So clearly just trying to put the Uh LED on the output of that comparator is uh is not a good idea. It's not going to work. We need some sort of other mechanism to uh, keep that Led on when it enters constant current mode. Now you might think that the solution to this current limit lead is is easy.

Aha, we got a spare Op amp here. why not just uh, parallel these inputs up here. get rid of the Uh feedback loop there and put the lead straight on the output like that. Well, if you do that, it's going to be a problem and I'll show you why.

Now if we go into current limit mode here. So when it drops 250, watch the lead here and bang. See it flashed on for a second. So it it worked for Split Second.

But you continue, it's in current limit mode, but the lead's not on and it really only comes on when you get right down in the noise down in there. And the reason for that is because the input is so, uh, these inputs are so marginal. Of course, they're going to actually uh, be equal very close to equal because that's the idea of the current limit. uh loop is that it makes you know the output voltage of this exactly the same as the set current here by way of that Loop And if they're exactly like that, then, well, it's it's You know it's going to be hit or miss.

whether it works. it depends on slight offset voltages and things inside the opamp. It's bad design so we need another method and as usual, the solution is pretty simple. All we want to do is just, uh, change the margin on this input just a smidge.

So what we want to do is lower the voltage on this non-inverting input just a tiny little fraction uh compared to what it is to the loop over here so that when we, uh, adjust this uh pot down and it gets just on the current limit it's going to. there's going to be some margin on these pins and this lead is going to switch on. So this input, instead of being exactly the same as this input and having no margin, the non-inverting input will be slightly lower than this one up here. So when the pot gets just right on the margin, bang, it'll switch on now.

I've used a 10K and a 2 Meg 2 here that'll give us about 4.5 M volts uh margin at the full scale volt input and when you get lower, you know down at .1 Vols it's only 450 uh, microvolts or something like that. but it's just going to give it enough so let's try that out. I' I've added the Um I've added the 10K in there. coming back into the opamp.

I've added the extra Uh 2 Meg 2 going to ground there and well, let's try it. Where let's get down to our margin here and watch. This is our current limit. This is the output of our Um opamp of our current Loop Op amp and watch the Lead watch the lead.

it'll just switches on. we're not quite in current limit mode yet bang, so it's it's just on the margin. But check this out, it's not perfect. If we wind it down, it will actually go off because there's probably some noise on there which is causing us an issue and then it comes back on solid right down at the lower end of the scale.

So uh, we're going to have to fix that one as well. So let's replace that with a 470k and see if that's any better. Let's try it out. Yep, I like that 470k it is.

and of course you can take that uh current limit output off to a uh, digital. uh, pick digital input to your controller if you have one. But uh, of course if you have an intelligent controller and you're not using nut pots like we are here today, adjust the voltage of current and you're using a digital control like a pick or an appil or something else to actually dry it to generate. Uh, the Uh voltages to drive the voltage and current.

It knows what voltage and current is driving at and you would, of course, uh, read the output voltage back off and you'd read the output current. Uh, back off as well. Both of those. Then the microcontroller knows you don't need any of this current limit.

It knows that the output voltage has not matched what it's set, so therefore it must be incr current limit mode or vice versa. It just knows all of those values so it can figure it out itself quite easily. We just check that out. Lm334 is actually 1 milliamp and it is with our 68 ohm um, sense resistor there, so not a problem.

Let's just check that over the whole output voltage range. that's our output voltage there and it's uh, still sticking with 1 milliamp And let's go. let's drop it down and you know it's only going to to go down to about .9 before it. Yeah, there we go.

It's starting to starting to mck up now and the good thing is is that it's still drawing that half you remember from the data sheet from the previous uh blog. We uh thought that it would get down to um, about uh, it would still be drawing half a milliamp um at about 0.8 Vols or or thereabouts I think we said and I just wanted to point out that up until now. um, my out output capacitor here. Um, although the data sheet says 2.2 microfarads minimum to make it stable I'm only actually using one microfarad I'm not even using a ceramic um either as as recommended.

So it's because I don't actually have a um a you know, a nominal 2.2 microfarad ceramic uh to hand. So it's uh, just fine with that output capacitance and we can add some more later and stuff like that and we'll play around with that. but uh, no, it seems to be quite good. and um, let's actually remove the output capacitance all together and see what happens.

Okay, that's with no output capacitance at around about Uh 2 Vols And as you can see, it's a lot noisier than what we're getting before. And let me put on a47 bang and we'll just go down to our 10 ms per division. AC Coupled, that's with the Uh 47 microfarad bypass cap on there at Uh. By the way, this is at 0.2 amp load.

by the way, I am driving 200 million amps. so let's remove It and Bang There you go it does. That's with no output capacitance. It really, Uh does not like that at all as you'd expect for a low Dropout voltage regulator like this.

So um, it seems to be uh, stable. but you know, at high you'd have to tested it at higher current. but of course you wouldn't. Um, just as good design practice putting anything less than the recommended 2.2 microfarad ceramic output capacitance.

But even with 47 seems to work fine, it add a bit a quar of an hour. Okay, let's test the Uh Dropout voltage at uh, no load so there's no load on the output. The output voltage is at maximum at 5.5 Let's tweak it down to if we can to 5.5 just as a nice round number. Ah, it's a bit hard with these single turn pots.

Really? Dicky That's why you need a 10 turn pot. Let me tell you, any good lab power supply should have 10 turn pot. Ah I Breathe on that fart halfway across the room and that thing's going to change. Okay, now this is our input voltage.

This is our output voltage. Um, the green is the input voltage here on the scope. The uh, uh, yellow is the Um output voltage down there. the ground reference is right on the bottom graticule down there.

And let's wind the input down until we start even seeing. uh, well, we probably won't see it oscillate, but uh, until we start seeing that drop there. Oh, is that it? Yep, there we go. So who? So we're talking 1.25 volts or thereabouts Dropout voltage at no load.

And if you're curious to see the Dropout voltage at an amp here, this is our input voltage I'm probing that directly on the Uh input pin of the LT Uh 3080. so that's bypassing all the wiring on the breadboard and stuff like that. and I'm measuring the output uh directly on the pin as well. and that's still the output noise there.

So let's wind the input down and see where the output starts to drop. There we go around about 6.8 volts by the looks of it. Yep, about 6 H, 6.85 6.9 Vol So that's about 1.35 to 1.4 Vols Dropout voltage. And is that what we expect? Well let's have a look here at the Uh Dropout voltage minimum because the V control pin.

Remember it's not the quoted specs cuz we've tied the V control pin and VN pin together. and uh, sure enough, at an Aamp up here at Junction temperature, well, we're going to be above Uh 25, let's go up to 50. You know we're about 1.4 Vol. So yep, it, uh, pretty much agrees with the Uh data sheet.

as you'd expect, now, one thing we really want to check is the Power Up Performance of this because the Uh Power Up performance of any bench power supply is very important. You don't want it to overshoot if you got your output voltage set to 5 Vols You don't want a big spike like that and then it to you know level back down at 5. Vols You want it to ramp up nicely and then go rock steady at 5. Vols Now let's have a look here and try and predict what's actually going to happen here now.

Uh, the input voltage up here is pretty much it. Technically, it's going through an RC fi here with the current shunt resistor, but it's NE negligible really. So it's so the voltage is going to go straight through to power the Lt380 straight to the output. There's no, uh, there's no capacitance, um around our voltage adjust pot or anything like that, um, or our voltage reference So that's going to power straight up instantly.

Power the pot instantly. This Op Amp's going to turn on instantly and it's going to try and drive the Uh set pin of the Lt380 instantly assuming our constant current. Uh, control is set all the way to maximum of course. um now.

but we've got this massive 22 microfarad here with these uh with this 2K series resistance here total cuz remember that transistor is not going to uh, switch on during Um unless there's current limit so that basically does not exist. So uh, our turn on RC time constant is going to be there. these two series resistors plus this cap. We're going to see this thing uh, ramp up and um and go to the output.

but because that's actually in an active Uh feedback loop and we're actually driving this with an Op amp? um, it's not. It's uh, going to actually charge uh, faster than our uh, if, say, if we set our output voltage to a volt, then it's going to charge up much faster than you'd expect based on that RT RC time constant. So let's uh Power this thing up and um and actually capture that on the digital scope and see if that's confirmed. Now what I've got here is uh, the Y yellow Trace here is the Uh output uh voltage and I've set that to 5 volts and we're at 1 V per division as you can see, spot on 5 Vols and the green Trace here is actually the Um output from the Op amp here pin 7even that actually uh drives that and as you can see, it's significantly higher and the reason for that is, uh, fairly obvious.

Not only do we have the 10 microamps current flowing through there, which will give us a slight uh drop across there, but also these values here. the feedback resistor and the gain setting resistors here are fairly low value. uh compared to these. So there's going to be uh, some output current flowing through there and some drop across there.

but because the feedback tap is there, it, uh, takes care of any extra current flowing through there. that's why there's that difference. So uh, we're going to have a look at these and we're going to capture these two points. Uh, when the circuit poers on I've got it into no load, so this will just be switching on into no load with say, a 7 and 1 half or 8 volt input voltage.

So to do that, we're going to set our trigger point to about a volt. There, we're going to, um, have it on the Uh positive. Edge we're triggering off channel One, which is our output. uh, voltage.

So on the rising Edge as soon as it gets to a volt, we're going to uh, trigger that. Let's set it to 50 milliseconds per division or so. let's switch the voltage off. Let's uh, trigger that.

Put it into single shot mode and uh, let's give it a go. Here we go bang. I Switched the power on and you can see it ramp up. Oh, what's that down there? That's interesting and uh, you can see the output of the Op amp.

Uh, go up like that and then come back down when it starts to regulate like that and that's a very clean response. I Like it. There's no overshoot or anything there's We'll have a look at this thing down here later. But uh, basically, all we care about with Switch On Power Supply is that it ramps up cleanly and doesn't overshoot.

And this one doesn't overshoot at all. It ramps up like that, almost linearly. and then it, uh, it clamps into into voltage regulation. Like that, our opamp, uh response.

of course it's it's got to do things cuz it's an active uh loop there and it's trying to stabilize itself and it comes down like that. Now this looks kind of linear, but it's not. That's actually going to be the um, the uh, exponential uh curve caused by this RC here. But it's um, but you'll find that its time constant is going to be a lot faster than if, um, than uh, what you'd expect from uh, zero if you do the math from zero to 5.

Vols And why is it going to be quicker? Well, that's a good question is because it is not par. it's not going to be 5 volts here instantly. This opamp is going to be higher than that. and as you can see, the Opamp right there suddenly went Wham right up to there and then started to go up like that.

So that's going to affect. the output voltage is higher than 5 volts there, so it's going to charge that capacitor quicker than what you'd expect. and if we lower the output voltage I think we'll find that that will look even more linear because you're right down in this curve there. So let's give that a go.

All right. I've set my output voltage at a volt. so let's um, do five. Uh, let's do 200 MTS uh per division, so that will'll end up uh, ramping up to the same value there.

But I think you'll find it's going to be pretty linear. Let's capture that again. Single shot mode and Wh, We're ready to go here. We go Bang There It Is there and we can actually see that if we zoom in on that yeah, look that is that looks linear, but that is actually the start of the um, uh, exponential.

Um RC curve. It's because that active opamp there is driving it. uh, faster than a normal RC curve. Now this little bit here is is interesting.

What's that? That's 200, 400, 600, 700 molts. Aha, that 700 MTS Um happens to be the same as what our regulator is capable of going down to with no load. so there's some turn on mechanism there internally to the Lt380 device that, uh, that causes that sort of, uh, jump up there and then flat for a little bit. But if you followed that curve down there I think you'll find that that will intersect precisely down the bottom like that.

So it's got this little knee characteristic. uh, switch on like that. but it's not a problem because it doesn't overshoot cuz our output voltage is 1 volts and there's no over. There's a tiny little bit of overshoot in there, there's not.

You know, it's nothing to worry about at all, so it's nice and clean. Turn on? Really? I Don't mind that at all now. I've switched it back to 5 Vols output voltage here. Exactly what we're getting before and just to demonstrate.

Uh, we'll see if we can demonstrate that, uh, faster ramp up time due to the uh Op amp there. What I'm going to do is I'm going to temporarily move this uh sense point back to here and then power up, power it up and uh, see what we get. and I think you'll find that the time constant will be longer and match the math exactly for 1K and 20 2 microfarads. All right, let's give it a go.

I'm exactly the same time base. so I've got my power. Switched Off Let's single shot that and let's turn it on. Bang! There We go and you can see that the output of the opamp is just switched on instantly Like that, as you'd expect because there's no RC time constant in the feedback.

uh, loop anymore. so it just bang it due to opam action. This output here instantly becomes this input here. so it ramps up to there and bang 5 volts and this is your RC time constant.

but You' still got this little knee here due to the uh, that's probably due to the constant Uh current uh Source in there, perhaps? uh, something like that. So that's still causing that little knee. But as you can see, the RC time constant changed. That was so much fun.

Let's do it one more time backwards. We'll switch it off going into single shot mode. switch it back on and Bang! There We go much faster rise time still that exponential curve as you can see the slight curve in there and the Op Amp of course has to compensate for it all now. hence all that crap in there and you can see that the Uh output of the Op a Rings a little.

once it hits that uh point and it goes bang straight down. there's a little bit of undershoot and Recovery there and that's but I like it powers on cleanly as far as I'm concerned now. I Know what you're thinking: 22 Mike Do we need it that high? What if we change it to like a 100n? Uh, because we do need some capacitance on there just to lower the Uh output, uh, noise and we'll need it for our um current, uh limit as well as we'll uh, probably, uh, find out in some further testing. but uh, let's lower that to 100 and see what happens at Switch on.

So let's take out the 22 and we'll put in the 0.1 mik in there and uh, that should work. A treat. Let's see what happens when we switch it on. All right.

I'm going to keep the same uh time base there as we had before 20 milliseconds per division. Let's go into single shot mode. Let's switch this sucker on and uh I think it'll turn on? Well, it definitely will turn on much quicker. bang.

Look at that. Woohoo! Check out the ringing on the Op amp. It's trying to recover there, but the Uh switch on is actually uh, quite clean. Still very little overshoot there and that's 5V output.

but there is. You know it's I You know it's not as nice I don't like it I don't like all that uh ringing and uh and and all that decay on the Uh opamp like that I think I much preferred the Uh. It was a bit smoother with the Uh 22 mic, but that's 100n for you. Now it might be tempting to use that Uh mic there and have a faster switch on.

Who cares what the Op 's doing, the output, you know it's It's not too bad at all. Well, let's uh, try to see what happens in current limit mode and to capture that, we'll change our trigger to uh up uh, just below the 5V level there and we'll trigger on the negative slope because when it goes into constant current mode, the output voltage is going to fall and well, let's try that and see what we get. I've put the 22 Microfarad back in. So this is our normal Uh circuit configuration now.

All right, we'll see if we can capture it going into constant current mode. Now we're exactly the same before as before. The yellows the output voltage at 5 Vols and uh, the green is the output of our uh, our control Op amp. Now I'm going to turn the uh, the pot, uh, the current pot, uh down.

so that, uh, the well, let's go into single shot mode. Okay I set my trigger below there and I'm going to turn my current part down I've got no load, but it will actually switch on right down at the low end when it hits zero. Bang There it is and you can see the Um exponential uh discharge due to, you guessed it, um, this resistor R2 here because when it goes into constant current mode like that, Bang it uh switches on this transistor which then discharges this 22 microfarad cap through that 1K to ground and uh, that's taking about uh 10, 20, 30, 40, 50 milliseconds or so there to switch that off so we can actually tweak the value of uh that? 1K You wouldn't want it any higher, but uh, we could, uh, certainly, uh, lower that, um, almost down to well, effectively down to zero so that we can, uh, short short out the cap directly and uh, get that fast turn on, but it seems uh, very well behaved. uh in uh, current, um in in switch over to constant current mode so let's try that again, but this time I've got a 1V output.

so I've changed my volts per Division and my trigger uh point and we're got driving 100 milliamps uh into a load. So let's turn our constant current pot and bang. That's what we get, but you'll find I think that that wiggle in there was me, uh, turning the pot. uh, let's see if I can, uh, do it a bit quicker.

so let's try it again. Single shot bang There we go and let's do it slowly again. Here we go bang and yeah, that's me actually turning turning the pot there. so bang, that's as fast as I can turn it pretty much and you see the opamps are doing some business there.

but uh, the output, uh, just drops very smoothly and very cleanly. All right I've changed R2 there to 100 ohms dropped it by an order of magnitude and let's uh, see what we get. and once again, you can see the Op amp oscillating in there and if we actually Zoom right in on that, you can see that it is actually oscillating like that doing its thing. but that doesn't manifest itself on our output here.

our output is, uh, dropping very smoothly. Um, because we've got the 22 mic cap on there. but let's drop that uh, capacitor. as we mentioned.

uh, way back down to 100n and uh, see if we get a similar response. So we now have a 100 in there with 100 uh n in there where we 1V output with a 100 milliamp load and let's give that a go. Let's go into constant current mode, turn our pot bang and this is rather interesting. Look that the uh, the output voltage, the yellow Trace looks fairly clean, but when we zoom in, it goes up and you can see that there's noise following in there like that.

and I Think that it's possibly because we're in high res mode. So let's actually uh, switch off um, high res mode and we'll try that one more time. bang There We go. it's exactly the same, so that is a disadvantage.

You can see the rolling average done in uh, high res mode on the scope. Bit of a trap for young players. that one now. I Put my 22 microfarads back in and let's give that a go bang.

look at that see so it is much cleaner. much cleaner response with the 22 microfarads and you can see that live here too. With the Uh 100n there it is as I turn it down bang look at that terrible and with no capacitor. At All By the way, it's an absolute shocker.

You don't even want to go there. So I'll put that 22 mic back in bang clean as a whistle. Look at that and there you go. Not a problem and I've put it to AC coupling 10 ms per Division and that's the noise if I adjust it I'm adjusting the pot there so you see the AC that's in voltage mode and we'll switch to current mode.

Bang Exactly the same noise. All right. We'll measure this switch off uh, transient now I'll set up a current for Uh 250 milliamps. There we go.

We got 250 milliamps running on there and uh, this is our Um output. uh AC coupled 50 Ms per Division and the green Trace again is our Um is our control Opamp output. So let's uh, trigger that and just disconnect the load I Know this isn't ideal. normally you'd electronic switch this, but we'll just disconnect that real quick and Bam There it is and you can see how it's uh immediately Switched Off there.

There's a bit of high frequency stuff there, as you'd expect and then it Boop RS back down and then it kicks in and the Op Amp kicks in and controls it All and all settles back down nicely. but that's uh, 50 MTS per division. So um, that's only jumped 50 MTS there for uh A4 of an amp, uh, output current. and of course, that's with our 1 microfarad output capacitance.

Let's uh, change that. Let's actually add some extra capacitance in there I'll get an extra Uh 47 microfarads and I'll whack that on the output and uh, see what difference that makes here? We go. didn't even trigger there. we go.

looks like we have to lower our trigger there. Let's try it again and bang. There We go much smaller uh output transient this time cuz we've got greater output capacitance. but as I mentioned in the previous video, it's a bit of a tradeoff.

You can't have an infinite amount of output output capacitance cuz then it when it switches into constant current mode, all that energy from the output capacitors can be dumped into the load before it has time to regulate the current. So probably not a good idea. And we'll check the lob when we do the switch on transient. So we want to set our trigger signal below like that.

So because we expect it to drip down like that, 50 MTS per division. so let's set it down there and let's disconnect our load. And here we go. We're going to connect our load real quick and bang.

There We go. There's our switch on transient 100 Ms uh, around about two divisions and then you can see the Op amp recovering there and stabilizing. And of course that was with the 1 microfarad output capacitance. Let's uh, try it again with the 47 microfarad and uh oh, there we go.

We got a lot a whole bunch of noise this time, but the transients not as large as before. Let's try that again. There we go. So it's a a little bit smaller than before.

now. set my output to maximum here and it should be 5.5 Vols but it's not. And once again, another trap for young players. It depends on where you actually measure that voltage.

I Got the ground reference over here on this side of the load. but I'm actually uh, attaching the Uh I'm actually measuring the load voltage right on that, right on that resistor there. Now the problem with that is is that the burden voltage of my multimeter is in place there. um Plus or plus the Uh plus the voltage drop in the Uh wiring and the breadboard and stuff like that.

So if I disconnect that yeah, it's a little bit hot and I connect it straight on to the tab of my To220 device. I think we'll find it's exactly 5.5 Vols Bang There it is. So that drop is in your breadboard and uh, and your wiring and your burden voltage. Just be careful.

Okay, I've replaced the current shut resistor with just a jumper uh link to get us uh, greater output current capability. I've currently got it set to uh, just over an amp and there's our output noise. Not a not a problem at all. We're still getting 5 uh, 48 volts out maximum and I can wind that Wick up a bit.

Of course things are starting to warm up a tad now and uh, it is rated to uh this. but now we're going a bit over starting to see some funny business. Now Are we maybe the uh, thermal overloads kicking? Oh, something's kicking in? perhaps? Yeah, there we go. Our outputs dropped to to 2 volts.

it's dropped out Bang but there you go, it actually recovers at 1.2 amps. Not a problem I Like it and just for fun, we'll measure that uh heat sink temperature will be at the 1 amp. Uh, it's about 28 1/2 29 ambient here in the lab and let's uh, let's probe that case and see what it's at. Okay, it looks like it's settling at about 43 1/ 12 and the heat sink itself is at about Uh 41 1/ 12.

C So there you have it. There's some basic, really rough and ready uh tests. Not sort of really. Precision uh power supply measurements at all.

but I I think it passed with flying colors on the uh breadboard. It pretty much worked as expected. I'm happy with that. So I might have to get on to uh laying out the board and building this thing up.

um for the final project and uh, I'll eventually, uh, show that one. So um I guess it's to be continued and there might be some more uh, performance checks and other things on this. But anyway, hope you enjoyed that. See you next time.