General > General Technical Chat
Mains Quality Measurement System
JoHr:
Hello Folks,
I´m doing a data acquisition project on qualifiing the mains parameter.
Its running since one year now. Exclusivly with frequency measurements.
You can find published data, multiple analysis and some variations of displaing data here ...
http://vanhinten.de/
Now I´d like to move forward and check more stuff. I did a quick shot
with an oscilloscope and a high voltage differential probe (HVDP) and logged for a few days
some voltage data. Its worth to log them continuosly as well.
An HVDP with ADC and Raspi backend is already designed to get rid of the oscilloscope and expensive probe.
For analysing voltage I´ll take a look at
* Vpp, RMS, Mean, StdDev, Crest, Rise/Falltime
* Harmonics (20)
Especially harmonics is a well indicator of when the mains get dirty by unregulated load.
Furthermore is planned to do a kind of software triggering to get each period which I´d like to pass through a mask test to get
kind of runts or interrupts in periods. Those events shall be logged with their timestamp and characteristics
as I did for frequency data.
What do you think would be worth to be logged as well? :-//
Main focus yet is logging data. Analysing them is another part.
* maybe three loggers at the same time on the three-phase current system?
* with measuring phase to each other
Thanks for hints and ideas
:popcorn: :-+
BR JoHr
Hawaka:
There is a lot of data that you can extract from just the voltage. Have a look at commercial power analyser, you'll get a lot of inspiration. Most of the time the manual is free to download so easy to see what they do. In addition have a look at EN50160 to see what are the parameters that matters for the grid supplier.
Check out https://www.a-eberle.de/, they have a lot of product that could interest you. Especially the PQ Box (https://www.a-eberle.de/produkte/the-mobile-tool-for-the-expert-pq-box-200/?lang=en). Dewesoft (https://dewesoft.com/) also comes to my mind and ZES Zimmer (https://www.zes.com/en). ZES also include the exact way in there manual on how they calculate each value, worth to have a look. A fluke 345 could also be interesting, but it's a simpler device.
Regarding data logging you could save each point of the measuring but that will very fast be a lot of data. The best would be to calculate everything for each period and just keep that. Then you can save the full data only on certain event.
I hope that helps ;)
JoHr:
--- Quote from: Hawaka on January 12, 2023, 09:16:48 pm ---Regarding data logging you could save each point of the measuring but that will very fast be a lot of data. The best would be to calculate everything for each period and just keep that. Then you can save the full data only on certain event.
--- End quote ---
Exactly around 1GSample on an analog sample feequency of 10kHz.
Second based data is sufficient. Interupts or runt will be worth to log those periods in detail. But I can decide this later on.
For the quickshot on logging harmonics which I did last christmas minute based logging was sufficient.
Additional comment ...
I took a look into my frequency logging. I got really every period with value and timestamp in a compressed json format dumped.
It makes ~1MB per hour. Peanuts ...
nightfire:
For stationary use, I had at my last employer Power Measurement (now Schneider) ION7600 and 7650 devices including ION Enterprise oin the software side- nice stuff, if you have the €$$$...
Otherwise: Whats of interest is also the earth leakage current in a big building. Some big devices have an own input for a current transformer, others are at least able (Janitza IIRC) to do some differential measurement approach across the phases.
JoHr:
The EN50160 is a good hint.
I´ve just had the EN61000 3-2 A/B/C/D in my mind.
But that includes voltage and current relations in absolute values and harmonics. I don´t want my own current consumption being mesured.
Navigation
[0] Message Index
[#] Next page
Go to full version