Effective values measurement

[naturephoenix]

Hello guys. I got AI card which takes samples every 50micro sec. My PLC cycle is 1ms which means I get every ms array of 20samples. I am measuring 3 voltages and 3 currents. Phase values. My goal is to have correct and quick as possible measurement. I am measuring frequency with PLL method and lets say speed from 0 to 50-60Hz is around 30ms. Filter for harmonic is included same as calculation of simetrical components in order to eliminate frequency deviation due to nesimetry.
Regarding voltage effective values I got some doubts and ideas as well:
1.Every ms Iam calculating maximum value, once I reach zero I got my Amplitude and after that I start calculating for next period. Effective value I calculate from Amplitude.
2.Since I am calculating frequency I can calculate Effective value every ms from the buffer of samples. Buffer size is related with frequency. In a case Iam measuring 50Hz, 10ms buffer(4000 samples) woukd be appropriate. This means I have to have predifined buffer for lowest possible frequency I want to measure. Lower frequency higher buffer.
3.Calculating maximum value every ms from the buffer of 20ms(8000samples) which should be enough for 25Hz frequencies and above. From Amplitude I can calculate Effective value

 

[Skogsgurra]

You have no distorsion? Neither in voltage nor in current?

Gunnar Englund

http://www.gke.org

--------------------------------------
Half full - Half empty? I don't mind. It's what in it that counts.

 

[itsmoked]

Thinking on paper here. If you're getting 1ms voltage readings and the shortest cycle time will be 16.6ms (60Hz) then each sample should have enough data to arrive at a frequency number. Take the first sample skip the next 8 samples and find the difference between each of the rest the cycles finding the remaining one with the least difference when summed. Extrapolate to refine the frequency. I might think of an improvement to this.

Keith Cress
kcress -

http://www.flaminsystems.com

 

[naturephoenix]

Thx guys on quick response. I will also think about more ideas too. Frequency is variable, I mean I am measuring it with PLL, It can be different than 60Hz.

I mean I should be able to measure at least from 20-75Hz

 

[naturephoenix]

Is it possible to calculate active power without knowing frequency and power factor?

 

[RyreInc]

Couldn't you just multiply each voltage and current sample together to get a power signal, and take the average over some period?

 

[IRstuff]

"Is it possible to calculate active power without knowing frequency and power factor"

You are already already measuring voltage and current, which means that you can detect zero crossings from that data alone. That gives you both period and phase, from which you can calculate power factor.

TTFN (ta ta for now)
I can do absolutely anything. I'm an expert!

https://www.youtube.com/watch?v=BKorP55Aqvg

faq731-376 forum1529 Entire Forum list

http://www.eng-tips.com/forumlist.cfm