Weird power supply signal

[sbo]

Hi,

I'm new to electronics and I recently bought my first DC power supply a Kungber SPS 3010

https://www.amazon.com/Kungber-Adjustable-Switching-Regulated-Adjustments/dp/B08DJ1FDXV/

When I set it to 10 volts (for example), I got the good voltage and I assume that if I connect it to an oscilloscope I will have a beautifull horizontal line at 10 volts with a little noise.

But look what i got  



Legend:

Channel 1 (Yellow) is on positive
Channel 2 (green) is on negative
Violet line is (Channel 1 - Channel 2)


Is it normal or a defect, or because of the bad quality of this power supply ?

Thank you in advance for your answers,
sbo

[WattsThat]

Incorrect hookup. Your probes are open circuit.

Probe tips on supply (+) , probe ground clips on (-)

That is normal 50Hz mains noise pickup from the unconnected probe connections.

PS: Do not attempt to connect your scope probes to mains to “see what it looks like”. That would not end well.

[RoGeorge]

GND banana plug for that power supply does not mean 0 volts.  GND there means is for the Earth grounding, or else said GND is the metallic case of the instrument.

To measure the output voltage (or to power your circuits) you need to connect only the + and the - plugs.  Earth ground (GND plug) is optional.  Read the user manual of your power supply, might be explained there how to connect it.

What you see on the oscilloscope is expected.

If you connect the oscilloscope probe normally, with the alligator clip to the - plug of the power supply, and the probe's hook/tip to the + plug of the power supply, you should see 10V stable, no square wave.

[perieanuo]

normal. more more more study needed about using oscilloscope, otherwise you'll fry him in 2 months testing some smps

[sbo]

Hi all,

I give you more precision. I'd like to have the same behaviour of when I take voltage measurement from my USB port (5V or 3.3V).



I exactly use the same technique, for the 3 measurements(technique discribe @ 21.45 in the following video) at the exception that I don't want to mesure the ripple or the noise from my new power supply.



If i got the same behaviour of 3.3V and 5V (USB) measurements

• I don't need to buy a differential probe
• I don't need to connect the ground of my probe to the negative slot of my power supply (which is not really safe, since ground is usually connected to the metal chassis ?)
• I don't need to use 2 slots of my oscilloscope with a math function

Thank you in advance for your answers,
sbo

[RoGeorge]

The poor man differential probing shown there might not always work because of various reasons, some reasons are shown in the next of the video, others are not shown, like for example saturating the analog input stage of the oscilloscope, recovery time, etc. 

The improvised differential probe works reasonably OK when all the instruments are properly earth grounded by their power plugs (at least), but you may never be sure.  Depending on what you measure, galvanic isolation (or not) is crucial.

For example, keep in mind that most of the oscilloscopes have the GND of the probe connected to the earth ground (the metallic enclosure of the instrument) and to the earth grounding wire of the power plug.  And more, all the GND probes are short circuited between each other, they are the same wire.

Therefore if you connect one probe alligator to a voltage and the alligator clip from a second channel to another voltage, your oscilloscopes GND will short-circuit the two voltages.

Another common mistake is to place the alligator ground of the oscilloscope in a random point of a circuit that is not galvanic isolated.  Such an attempt can blow your scope, your circuit to measure, or at best will trip some fuses.

Learn about galvanic isolation vs. earth grounding, that is a must know for anybody using mains powered instruments.



If you want to measure the ripple of the power supply, plug both the oscilloscope and the power supply in the same mains outlet, one that includes the 3rd wire for earth grounding, and that is properly grounded.

Then, use the measuring setup shown at the minute 32:00, use a power resistor as a load, do not use electronic active loads, and turn off any nearby electric apparatus (laptop/desktop, lights, white goods, wi-fi routers, mobile phones, fridges with motors, anything else that might produce spikes in the power lines or induce radio noises through the air).

If you are unlucky to live nearby a very powerful broadcasting radio station, or nearby some cell phone tower, you might also need a Faraday cage to shield the instruments against fields coming from outside the lab, but this kind of shielding will be melodramatic for measuring a switching power supply. 

It all depends of how precise do you want to measure.  The setup shown at minute 32:00 should be enough.

[sbo]

Hi RoGeorge,

Firstly many thanks for your time and answers.

If you look at the last picture down below, you will see that the third measure (Channel one yellow curve connected on +) has a ~30V peak to peak (not just ripple nor noise)
If you take your DC power supply and set it to 10V, do you have a nice horizontal line like the top and middle image ?



What I do not understand is why i got differents results between
- on one side, the 2 firsts measurements that are both from a USB port (Seasonic Power Supply, 3.3V & 5V) --> both straight lines
- and on the other side, the 3rd measurement from Kungber DC power supply --> why do I get waves here ?

Best regards,
sbo.

[RoGeorge]

The USB doesn't seem to be galvanically isolated (relative to your oscilloscope), while the power supply seems to be galvanically isolated (sometimes called a floating output power supply).  That is why the first 2 don't show any ripple.

The ripple you see in the 3rd screen is expected, such ripple is caused mainly by the Y capacitor inside the instruments, and because they are not properly earth grounded.

The common mode voltage (ripple) is exactly what a differential measurement with 2 oscilloscope channels (or with a dedicated differential probe) will try to cancel out.  The value of that ripple is not relevant, and varies wildly.  Try connecting the oscilloscope like at the minute 32:00, with shielded cables instead of the probes, and with the extra capacitors and the extra 50 ohms resistors shown in the diagram, and you'll see the common mode voltage ripple is not a constant.  What matters is the difference between the two signals, not the common mode ripple voltage.

Show the connections you did between the oscilloscope and the power supply while you were measuring.  Two probes have 4 end-connecting points.  Where did you connect each of the 4 wires while you were measuring?

[Jwillis]

Did you try measuring the voltage with the supply under load . Since the Kungber SPS 3010 is a SMPS the probes may not be enough "load " to activate the power supply . In some cases , Switching supplies can act weird without a load connected and your probes might be picking up background AC noise .