Regulated power supply

[D.M. Procida]

I have an adjustable PSU with a wide range of voltages (1.5-12V), which says "IC UNIVERSAL REGULATOR" on it. When set to 9V it produces 9.38V without load.

Does that sound like it's a regulated PSU?

Daniele

[fred]

In article , D.M. Procida writes

Yes

[Owain]

Probably, but what does it produce when loaded?

Owain

[geoff]

In message , D.M. Procida writes

A cheap and nasty psu and you expect it to be that accurate offload

Why not put a nominal load across it and try it under realistic conditions

[Roger Mills]

This may not help if you haven't got anything to compare it with, but ISTR that the dimensions of the output connector are different depending on whether or not it's regulated. Assuming that it's got a concentric-type cylindrical connector, I think that regulated supplies have a bigger hole in the middle - the rationale being that devices requiring a regulated supply have fatter central pins in their connectors, to prevent an unregulated supply from being connected.

I'm sure that someone will correct me if this is wrong!

But the fact that it says "IC UNIVERSAL REGULATOR" on it, and the fact that the no-load output is only 9.4v suggests to me that it *is* regulated. Unregulated 9v supplies often deliver 12 or 13 volts on no load.

[Harry Bloomfield]

D.M. Procida laid this down on his screen :

Not really, unless the thing (switch/control/?)you are setting the voltage with is out of calibration.

Apply a small percentage of the full load and measure the voltage - repeat with a full rated load and the voltage should be similar.

[mick]

Almost certainly. If it had been unregulated you would have probably measured well over 10v off load.

Just because the supply is regulated it doesn't mean that the voltage is accurate. What it means is that the voltage won't change (much) with varying load. In reality it will sag a bit, so 9v sounds very likely.

[D.M. Procida]

Thanks for all the various replies.

How different from the loaded output would the unloaded voltage need to be before one could reasonably judge such a PSU to be unregulated?

Daniele

[Bob Minchin]

It all depends on the design but I'd expect around 10% change when the load is varied from 10% to 100% of full load if it was an UN-regulated supply. A quality regulated supply would vary less than 1% over a similar range of loads.

Bob

[Dave]

A cheap and nasty volt meter will produce the same fault.

Without an accurate volt meter, this will tell the op nothing.

Dave

[Roger Mills]

Depends on the quality! I would expect a regulated supply to control the voltage within a couple of percent over its rated current range, whereas an unregulated supply may be several tens of percent over the nominal voltage at no load - but should be close to the nominal voltage at full load. I'm guessing a bit with the actual figures - but the thing is, they're an order of magnitude apart!

A regulated supply will have some sort of feed-back loop, to maintain a more-or-less constant voltage regardless of load - whereas an unregulated supply won't, so it's voltage will fall as the load increases.

[geoff]

Even cheap and nasty voltmeters are accurate to 1% nowadays

[geoff]

So he is ~5% off

a semi-regulated supply then ...

[Dave]

It did cross my mind, while typing that, to do a test with my cheap and nasty voltmeters in the morning, using my calibrated one. :-)

Dave

[terry]

,

So at one setting; 9.38 compared to 9 =3D a 4.22% variation! Would that everything else in this world be that precise? And 'regulated' most likely means that it will not vary more than a certain amount over a pre-designed and stated range of loads and voltage settings. In that case anything that does not vary by more than 5% is, for most practical purposes, pretty accurate! An amount of 0.38 volts 'out of wack' at 9 volts is unremarkable. Unless it is laboratory work; in which case the use of highly accurate measuring devices is indicated.

[mick]

It will. It will show whether the supply is regulated or not. Putting about 10% base load on then adding a further 50% load should give virtually no change in terminal voltage for a regulated supply. Only the repeat accuracy of the voltmeter is important.

A lot of cheap-n-nasty digital instruments are well within 1% tolerance now - particularly on DC voltage ranges where 0.5% or better is common (but you can't always trust what it says on the box, of course!).

[Ian White]

If it contains a regulator IC and it's also switchable between various voltages, that sounds like an LM713 (or similar) with a switched chain of divider resistors. If these resistors are selected from standard values with say 5% tolerance, there are bound to be small deviations from the nominal output voltages. Even so, the selected output voltage should be quite well regulated against changing loads... until it hits the limitations of the transformer and smoothing capacitor.

[Bob Minchin]

or even an LM317!

Bob

[Ian White]

Oops, sorry - my 317s and 723s are starting to interbreed in the component drawer.