True RMS measurement

EDAboard.com Forum Index -> Professional Hardware and Electronics Design -> True RMS measurement

Author

Message

techie

Joined: 05 Feb 2002
Posts: 850
Helped: 43
Location: Pakistan

03 Jun 2003 9:24 True RMS measurement

How to measure the True RMS voltage and current of 50/60Hz mains (sine wave with harmonics) using a successive approximation ADC and a microcontroller. What is the sampling rate required, ADC resolution, computation involved and technique used.

thanks for any suggestions

C-Man

Joined: 19 Jul 2001
Posts: 1235
Helped: 73

03 Jun 2003 9:43

There was an article in circuit cellular:
hxxp://www.chipcenter.com/circuitcellar/november99/c119dp1.htm

sourcecode can be found here:
fxp://ftp.circuitcellar.com/pub/Circuit_Cellar/CConline/1999/Nov99/Perkins.zip

best regards

Last edited by C-Man on 03 Jun 2003 9:55; edited 1 time in total

Google
AdSense
Google Adsense

03 Jun 2003 9:43 Ads

tjalps

Joined: 28 May 2001
Posts: 266
Helped: 2

03 Jun 2003 9:49

Hello!

Check at www.analog.com RMS . They have solution for you.

Regards!

billano786

Joined: 26 Apr 2001
Posts: 370

03 Jun 2003 10:08

AD536 from analog devices may help you.

techie

Joined: 05 Feb 2002
Posts: 850
Helped: 43
Location: Pakistan

03 Jun 2003 10:38

The AD7754 seems the best suited IC for this purpose. Thanks for pointing that out. This IC has all the front end required for the purpose, a PGA, an ADC and rms convertions etc. great design

tmk

Joined: 20 Dec 2003
Posts: 11

31 Dec 2003 9:40 Re: True RMS measurement

the True RMS chips are available at the market but all of them are expensive. you can realized this with a MCU.

if you want to work at 50-60 Hz you should sample the signal at the 100-120 Hz. but these restricts are theorical. I realized this projects at speaking frequency with Cypress PSoc.(www.cypressmicro.com) and in my opinion you should sample the signal at least 200 Hz. (for 50 Hz.) and sample count should be 128 at least.
the RMS calculation C sample code is below

total = 0;
for (i=0; i<SMP_CNT; i++)
{
total += (sample[i]*sample[i]);
}//for
RMS = sqrt(total / SMP_CNT);

you can use this code as you want.

tmk Wink

km

Joined: 26 May 2004
Posts: 32

27 May 2004 10:00 Re: True RMS measurement

hi tmk, I just saw your post and I have some related question to ask you.

tmk wrote:

if you want to work at 50-60 Hz you should sample the signal at the 100-120 Hz. but these restricts are theorical. I realized this projects at speaking frequency with Cypress PSoc.(www.cypressmicro.com) and in my opinion you should sample the signal at least 200 Hz. (for 50 Hz.) and sample count should be 128 at least.
the RMS calculation C sample code is below

total = 0;
for (i=0; i<SMP_CNT; i++)
{
total += (sample[i]*sample[i]);
}//for
RMS = sqrt(total / SMP_CNT);

I'm going to do a PIC project and this is the link to my project:
http://www.edaboard.com/ftopic78357.html

Do you have the C code for the frequency, peak-to-peak, average, and duty cycle calculations of an analogue waveforms.

My project is also going to measure these values for an analogue waveforms 50/60Hz mains using an ADC (TLC 548 8-bit resolution) and a PIC 16F628. What is the sampling rate required? ADC resolution? computation involved and technique used? Is it same as what you have said?