DCUU PM9913 single phase power meter 5kHz - All it secrets

[Kiriakos GR]

Back plate of the instrument, and wires connectivity.

Nothing found as wrong at the back plate.
The product designer did a good job, panel mount banana, AC input connector, along the AUX I/O port they all have a good symmetry.

[Kiriakos GR]

DCUU PM9913 this is based at a single metering chip product design.
And while the actual metering chip, this is in the market for two decades, DCUU indeed it did succeed to use all of it functionality.

At my PM9913 I did performed a significant amount of functional tests, so and I (first person worldwide), succeed to verify and or to better understand actual product design performance.

Test on Frequency counter detection
PM9913 lower frequency detection this starts from 20Hz. 
But higher frequency detection this is relative to input voltage level.
Input 26 ACV - frequency detection 20~50Hz
Input 30 ACV - frequency detection 20~60Hz
Input 45 ACV - frequency detection 20~100Hz
Input 80 ACV - frequency detection 20~200Hz
Input 110 ACV - frequency detection 20~300Hz
Input 150 ACV – estimated frequency detection 20~400Hz

Test on Frequency counter linearity
Input 20Hz - reading 20.0
Input 50Hz - reading 50.0
Input 60Hz - reading 60.1
Input 100Hz - reading 100.4
Input 200Hz - reading 201.9
Input 300Hz - reading 304.5
Input 320Hz - reading 325.1
Input 400Hz – not tested

Test on Power Factor detection
As it is proved at Frequency counter detection test, there is a requirement for a minimum input voltage for having best detection (Error Free).
For the power factor, input ACV this must be at 95 Volts, while actual minimum this is 90 Volts.
For my PM9913, Error Free PF detection, this is from 94 ACV and or higher than that.

Test on Voltage bandwidth 
PM9913 this is not using any voltage sensing transformer, it is using instead a voltage divider.
By design a resistors network in the form of voltage divider, this is not introducing any phase shift error.
And neither this seems causing any bandwidth limit.
Actual energy measuring chip this has highest voltage bandwidth limit at 1600Hz.
At my tests, I did supply AC voltage at a range of frequencies, and ERROR Free range this is from 20Hz up to 1200HZ (Class 0.5).
Just for fun I did tested from 1200Hz and above, and ultimate end of the road found at 1900Hz.

And therefore actual measuring bandwidth confirmed up to 2 kHz Max.
I am unaware of how or why, DCUU this wrote 5kHz as Max        

Test on Current sensitivity 
PM9913 this is using a sensing current transformer this been rated at 20A/10mA.
Long story in sort, DCUU engineers they are using only a portion of it entire range, and by doing so, they keep at a minimum any linear range error.
Transformer Phase shift error: lesser than 20 degree
Linearity error 0.2%
At my actually testing, PM9913 this is capable sensing from 0.20mA and higher than that.
But for best energy calculation, the input voltage must be at 90 ACV and ACI current input must be from 7.0mA or higher. 

Active testing Summary
A Single phase power meter, this is a combination of voltage meter and a current meter along a power calculator IC.
As is know from the world of multimeter, they are a portion of limitations, these be relative to measuring range per electrical parameter, for best accuracy.
DCUU specification of 5V as starting ACV this is verified, but this is valid if PM9913 this is used as this been a voltage meter.
For a proper power in Watts’s calculation at entire bandwidth 20Hz~400Hz, this is possible with an input of 150 ACV or higher.
For Europe and USA, PM9913 this is usable as power meter, with input voltage from 30 ACV and or higher than that.
Error Free Power factor detection at 94 ACV and above, and I will name that as a High precision Mode activation at 20Hz up to 200Hz bandwidth.

PM9913 this is able measuring as low as 1/3 of entire power consumption of single Neon bulb.
PM9913 voltage detection this is free of error at sine wave, square wave and triangle wave.
 
PM9913 when this measuring Stand-by power of pulsing power supply, Power Factor detection and Power measurement this is very unstable.
The solution is the use of PC logging software, so this to record the average measured power, within a long duration of product testing.

In conclusion, today that I am aware of most of strong and weak points of PM9913, regarding manufacturing quality along electronics performance, and by thinking also the factor of product price, I will say that my money in the form of an investment, they have not been wasted.
For my taste, DCUU this is an awkwardly managed company, product documentation this is extremely inaccurate and poor, website and due internet communication another major problem.
But their products they do work and deliver usable results, when they are operated at by Kiriakos Triantafillou discovered ACV and ACI electrical input levels.

 

[Andrew-88]

In summary, build quality it is at the edge of the limit of been called as acceptable.
When a factory it is small, when all participants are young and inexperienced, such quality control details they get overlooked without a penalty.
None of these irritations would be there, if production cost this was increased by five USD dollars.

This mini review and the results, it makes me wonder, of how many similar issues they might hiding at the most expensive single phase power meter?   

I had long time to read something that interesting.
Thanks mate !

[overvolt]

I will add in the list of good news, that this model has a linear power supply inside. 
(According to notes over the power socket)

This translates to a much longer product life cycle, in to my eyes.
And the measurement circuit, this has nothing to worry about any radiated interference.

[Andrew-88]

In such a complex instrument, it is possible PF LED window and Frequency LED window them to be few counts out.

But what about an actual comparison, of Watt with the VA window.
Does it measure correctly from low to high?