Measuring Open Circuit Voltages with Sensing Resistors and a CRO

[houlahound]

I measured a bunch of open circuit voltages across a variety of power supplies using a CRO. the signal is just spurious.

Adding a resistor across the output of the same power supplies gives the expected trace. I used high and low impedance option which was 10 megaOhms on the CRO leads.

What is going on, in theory the open circuit voltage in this case max 12 volts should be measurable without a load resistor to draw a current??

Apologies if a dumb question.

 

[Tom.G]

A high open circuit voltage from a power supply is common. The reason depends on the type of supply.

If it is a low power AC 'wall wart' style these are usually just a small transformer and when you draw current theire is a voltage drop due to the internal resistance of the transformer.

If it is a low power DC 'wall wart' style these are usually just a small transformer, a diode, and a capacitor for filtering. they have the same problem as the AC 'wall wart' supply, internal resistance and additionally, at no load, the capacitor charges up to the transformer peak voltage. When a load is supplied, the capacitor voltage drops until the next power line voltage peak occurs to recharge it.

These small supplies are rated for output voltage at their rated load current.

(EDIT) added last sentence

 

[houlahound]

The power supplies were in fact inexpensive AC and DC constructed as you said. I also used a high quality 40 Amp, regulated 12 volt supply from my radio station.

The spurious reading was kind of noise on the mains hum. It was not really the output signal at all. Any random resistor placed across the output and the CRO put in parralel with the resistor fully cleaned up the signal and made the measurement sensible.

I am just wondering why it does that.

 

[Tom.G]

Hard to say for sure. What comes to mind as most likely is with no load on that 40A supply the output was pretty much disonnected from the internals. This would make your measurement about the same as putting the CRO probe on a length of wire that's not connected to anything, you would see stray pickup from the power lines.

 

[houlahound]

That's exactly what it looked like.

If I recall I have done this test with a digital multimeter and get the precise voltages expected.

 

[houlahound]

Due to the simple and quick nature of this little experiment I would like to see if anyone else can reproduce it.

 

[Tom.G]

It's probably dependent on the internal details of the particular power supply. Many have a resistance across the output so always have a slight load on them.

 

[houlahound]

You know I think that you solved it. The multimeter plugs into the output ports. The CRO leads kind of contact the outer conductor which may not be connected.

I am embarrassed if it is that simple.

 

[Tom.G]

All I can say is "Welcome to the crowd."
(It happens to all of us on occasion)

 

[houlahound]

Will attach my CRO leads to some banana plug leads that push an inch into the output port of the power supply to verify.

Really hoping on one hand it is a deeper more mysterious problem...but on the other hand.