Voltage Regulator performance at/near dropout?

[FenderBender]

Hello,

I don't really have the resources to actually experiment myself, but for a standard say 7805 regulator with a dropout voltage of 2.0V.

What happens if I apply 6V? Does the output drop to a regulated 4V? Does it go to 0V? Is it just really unstable?

Thanks.

[JackOfVA]

The regulator will deliver the source voltage minus a volt or two, but with a significant series  resistance so you have little to no regulation.

In other words, with a 7812 12V  regulator, applying 10V to the input will yield around 8 or 9V output open circuit but with high series resistance so that as you try to draw appreciable current from the output, the voltage will drop.

[FenderBender]

Thanks!

[madires]

You might like to watch:
 

It wasn't intended I guess, but you'll see what JackOfVA wrote :-)

[FenderBender]

Thanks again. I'm actually trying to make a powerpoint presentation on AC-> DC PSUs right now, that will eventually go on youtube. But I'm away from my lab and scope right now so I couldn't test it myself.

[c4757p]

I'm actually trying to make a powerpoint presentation on AC-> DC PSUs right now

Ah - then does it even matter? You're not supposed to do it, so don't do it. It's not even entirely unlikely for slightly different regulators to behave differently outside their specified region of operation. I'd just keep to the way they're supposed to perform.

(And the little bit of subversive punk asshole in me cringed at "you're not supposed to do it, so don't do it". OK, I amend this to "when using purpose-built components in a high reliability device".)

[FenderBender]

Well the idea is that I wanted to show them what happens if you don't follow the rules, I guess for curiosity's sake. But as you say, I guess its adequate just to say "it won't work. don't do it".

Just a question about fuses, why does it matter where you put a fuse in a circuit? I know the convention is to put the fuse on the 120/240V mains live side.  But I'm just thinking if current moves everywhere in a circuit at once, if you think about the marbles lined up in a tube analogy, why does it matter if the circuit is cut after a voltage regulator or on the mains plug? Is it more a safety thing? Am I being stupid?

[c4757p]

Depends on where the fuse goes. Live/neutral is a safety question - if you break neutral, than the circuit will be energized even though it appears to be shut down. Very bad. You could also conduct current to PE. More electrically, if you put the fuse after a voltage regulator, it won't do jack shit for an internal short in the regulator or filter capacitors or whatever. You want to protect as much as possible, with the safest possible condition after the fuse trips.

[c4757p]

Well the idea is that I wanted to show them what happens if you don't follow the rules, I guess for curiosity's sake.

Sometimes this is a fun question to answer, but I suspect an undervolted regulator won't really do anything interesting.

[IanB]

Just a question about fuses, why does it matter where you put a fuse in a circuit? I know the convention is to put the fuse on the 120/240V mains live side.  But I'm just thinking if current moves everywhere in a circuit at once, if you think about the marbles lined up in a tube analogy, why does it matter if the circuit is cut after a voltage regulator or on the mains plug? Is it more a safety thing? Am I being stupid?

The mains can deliver a lot of power if there is a fault on the mains side of the circuit. So you put a fuse on the mains input for safety to avoid heat, smoke or fire if anything goes wrong. Remember that the mains can easily deliver 1000 W into an appliance like a toaster when this is intended. It can just as easily deliver 1000 W into your faulty power supply when it is not intended.

[vlf3]

It's been said and shown, and if anyone has actually experimented with 78XX 1Amp series... however if you want to draw under 100mA for battery designs, the other version for 78LXX is the LP2950CZ it's a LDO, Low-Drop-Out regulator; the head-room is about 40mV... just thought to make the point. 

[FenderBender]

Ohhh alright. That makes sense. I guess I was just assuming the short/problem was with the load, not within the supply.

Makes sense.

[SeanB]

With regulators if you go into the headroom often some will happily oscillate under the right conditions, and will become very hot. You need just the right no name brands, the right voltage, the right load current and the right low value OHL caps though, and then it becomes a power oscillator and cooks itself. Replace the regulator and the caps and all is fine.

[FenderBender]

haha nice fun fact.

[SeanB]

Bitten me once or twice, worked fine until I had a low mains voltage and then smoke.

[madires]

The mains can deliver a lot of power if there is a fault on the mains side of the circuit. So you put a fuse on the mains input for safety to avoid heat, smoke or fire if anything goes wrong. Remember that the mains can easily deliver 1000 W into an appliance like a toaster when this is intended. It can just as easily deliver 1000 W into your faulty power supply when it is not intended.

Over here the standard circuit breaker used has16A. The older ones break at about 80A immediately, so you'll get 240V * 80A = 19.2kW for a few ms.

[JackOfVA]

I had some spare time today, so I wrote a quick program to step an HP6612C power supply over a voltage range and read the resulting output voltage with an Agilent 34410A voltmeter when applied to a fixed voltage regulator. (Uses EZGPIB to control the power supply and voltmeter over the IEEE-488 ports.)

To see if the regulators  behave the same way when the voltage starts in regulation and then decreases compared with being out of regulation and increasing, I ran the test with both a start low, ramp up and a start high, ramp low voltage. I've added arrows to the 5V plot to show which path belongs to the increasing and decreasing ramp. The difference between the two paths is the hysteresis. The hysteresis loop is so small in the 10 and 12V regulators that I did not add arrows.

Attached plots are for  5, 10 and 12V standard dropout regulators, in TO-220 or similar metal tab packages. All operate into a 100 ohm fixed resistive load.

A few things of note ...

(1) Note the hysteresis exhibited only by the 5V regulators. Just a trace of hysteresis can be seen on the higher output voltage parts, but it's quite distinctive for both 5V regulators.
(2) Output voltage dropout is quite evident with a sharp knee. (Data is taken in 100 mV steps).  Looking at the datapoints, the difference between being in regulation and dropping out of regulation is not more than a 200 or 300 mV.

Leaving the hysteresis area aside, the output voltage for a fixed resistance load looks to be nearly linear with respect to the input voltage. In other words, the regulator operating below the knee is close to a zener type sort of constant voltage drop. The fixed drop in the below threshold regime for TI TL780-12KCS 12V regulator is about 1.5V. This may well be different with a different load impedance, temperature, phase of the moon, etc.

Note that this data is only for one sample part of a limited number of devices and, more importantly, I can't think of a reason - good or bad - why one would intentionally operate a fixed regulator with an input  voltage below the regulation knee.

I didn't see signs of oscillation for any of the six regulators, but that's far from a conclusive test. My experience is that the low dropout regulators are much more prone to oscillate under out-of-specification conditions, such as drawing less than the minimum current. 

[FenderBender]

Thanks! Awesome plots and experiment!

[mariush]

If you want, I can test a LT1085 adjustable reg (set at 5v on a protoboard) for you. I may also have a MIC2941 around and some LD1117 adjustable.

[IanB]

Nice experiments. Now I wonder about the LM317 in comparison to those fixed regulators? The data sheet suggests the LM317 is "better" in certain ways than comparable fixed regulators. If you adjust an LM317 to 5 V how does it compare to a 7805?

[JackOfVA]

Nice experiments. Now I wonder about the LM317 in comparison to those fixed regulators? The data sheet suggests the LM317 is "better" in certain ways than comparable fixed regulators. If you adjust an LM317 to 5 V how does it compare to a 7805?

I have a few LM317's around but didn't have time  to lash up a test setup. If I have time tomorrow or Tuesday, I'll try it.

[JackOfVA]

NJM317F adjustable regulator set to 5.0V, with 100 ohm load.

No sign of the hysteresis seen with the two fixed 5V regulators.

[FenderBender]

Very interesting. Wonder why that is.

[mariush]

Here's LT1085 3a adjustable reg. ... the graph below is with about 110 ohm load (2 x 220 ohm 5% in parallel)

I've used 5% resistors so by default it stays at 5.083v
The power supply I used (the Tenma I posted here) has only one decimal after the dot, but I tried my best to get the measurement done in the middle of .2 and .3 , .7 and .8 etc etc   The output measured with 22000 count Uni-T UT61E

http://cds.linear.com/docs/en/datasheet/108345fg.pdf

Operates Down to 1V Dropout
Guaranteed Dropout Voltage at Multiple Current Levels
Line Regulation: 0.015%
Load Regulation: 0.1%

Looks like they didn't lie.

LibreOffice spreadsheet (.ods in .zip) attached if anyone cares...



 

[FenderBender]

Very linear. Oh the irony...

What happens if you apply a load? Does it go beserk?

[mariush]

I only tested with the 2 x 220 ohm resistors in parallel. I can do some more tests if you're really interested, but it will have to be much later... I'd rather get some sleep now, it's 4:50 am here.

[BravoV]

Very linear. Oh the irony...

What happens if you apply a load? Does it go beserk?

If its Linear LT108x series, the regulation and drop out basically spot on even at max load as stated in their datasheet. I've done an exact experiment like marius have done while ago on LT1083 at 7.5 Amps.