AC/DC Power Supply Unit

[sruthisreenivas2009@gmail]

How to improve AC input power measurement accuracy for DC output loads below 10 % of maximum output power of AC/DC power supply units?

 

[Easy peasy]

simply loop the AC wires thru the current sensor twice or more to up the current sensitivity ( obviously won't work with a resistive shunt )

 

[sruthisreenivas2009@gmail]

simply loop the AC wires thru the current sensor twice or more to up the current sensitivity ( obviously won't work with a resistive shunt )

Thank you for your reply. I was thinking with this very low load we will have more power loss (reactive power) than actual power. Then we may need to consider the power factor for the input power. I need to understand more on details like this. What all the factors will affect the accuracy of the measurement during this low loads below 10% of max output power. Let the max output power be 1320 W or 360 W.

 

[FvM]

The question isn't clear. Are you measuring input current or real input power? In the latter case you need a power meter, power analyzer or a digital oscilloscope able to calculate real power from voltage and current measurements.

Does the power supply perform power factor correction?

 

[sruthisreenivas2009@gmail]

The question isn't clear. Are you measuring input current or real input power? In the latter case you need a power meter, power analyzer or a digital oscilloscope able to calculate real power from voltage and current measurements.

Does the power supply perform power factor correction?

I need to measure the real input power and I am thinking of Power meter IC as an option. Since I need those values for later analysis. Yes, the power supply has power factor correction.
If we use power meter IC or power analyzer during low loads can we improve accuracy closer to the range of+/- 5%. Which option will be better? I don't know much about power analyzer.

 

[KlausST]

Hi,

As FvM said the question isn´t clear.
When you say "improve" we need to know the existing state. Regarding measurement type, sensors, schematics, actual accuracy...

If you want numbers:
I´ve built a somehow optimized meter for single phase 50Hz mains applications able to measure:
* typ: 230V AC (range +/-600V peak, single range, resistor divider)
* typ: 0..16A RMS (range +/-100A peak, single range, shunt)

Output is (digital via USB, max update rate: 20ms. Measures up to 30th overtone)
* V_RMS: resolution: 1mV, max zero error: +/- 30mV, user calibratable, drift: +/-5mV within 1 hour at room temperature
* I_RMS: resolution: 1mA, max zero error: +/- 20mA, user calibratable, drift: +/-5mA within 1 hour at room temperature
* P_eff: res: 1mW, max zero error: 80mW (gets calibrated with V and I)
* V_AVG: resolution: 1mV, max zero error: +/- 30mV, user calibratable, drift: +/-5mV within 1 hour at room temperature
* I_AVG: resolution: 1mA, max zero error: +/- 20mA, user calibratable, drift: +/-5mA within 1 hour at room temperature
All output values settle from 0% to 100% input within two updates (i.e. 40ms on a 20ms update rate. No typical time constant behaviour!)
Overall accuracy error should be better than 0.2%

Made using 24 bit ADC 2Ch simultaneous sampling ADC, microcontroller, advanced software (unique adaptive digital filter).

***
So give numbers like V, I and P ranges, nominal (expected) V, I and P, accuracy, update rate, informations about waveform, output format (analog, digital..)
...and we will be happy to give you detailed assistance.

Klaus