Power metering with Atmel M90E26

Go To Last Post
5 posts / 0 new
Author
Message
#1
  • 1
  • 2
  • 3
  • 4
  • 5
Total votes: 0

I am interested in using this chip to measure active and reactive power, rms values and phase between two sinusoidal signals. Unfortunately this IC is only specified for a narrow frequency band (50-60Hz), whereas my signals could probably be some higher (let's say 300 or 400 Hz).

Do you think it's possible, and could you estimate the error increase in case it is?

Otherwise, do you know a better suited (and possibly in the same magnitude of cost :-) ) chip?

 

Thanks in advance

  • 1
  • 2
  • 3
  • 4
  • 5
Total votes: 0

As this seems to be a "metering IC" for mains metering (have not looked at the datasheet) it might well be that at 300 - 400 Hz it will measure almost nothing.

I can not imagine that there are no graphs in the datasheet that will tell you how much suppression there will be at those frequencies.

 

As it is hard to make small (size wise) filters that are good in killing 300Hz but still leave 60Hz through without problems it might be that you need to add external filtering. In that case you might want to investigate in how that is done for the 50HZ and adapt for the 300hz and see what the result is.

 

  • 1
  • 2
  • 3
  • 4
  • 5
Total votes: 0

meslomp wrote:
 you might want to investigate in how that is done for the 50HZ 

The chip has a DSP to do all that:

 

 

It also has a fixed sampling rate - it doesn't seem to be adjustable at all.

 

http://www.microchip.com/wwwproducts/en/atm90e26

 

 

Last Edited: Thu. Sep 14, 2017 - 07:45 AM
  • 1
  • 2
  • 3
  • 4
  • 5
Total votes: 0

Hi guys, thanks for your interest.

I could not find any graph; filtering is made at the inner DSP level I guess. However the datasheet declares:

 

Sampling Frequency 8 kHz

 

(should be enough...)

 

Harmonic Metering (active and reactive) Bandwidth 1.1 kHz 

 

(to me seems suggesting that with some correction it could be feasible...)

 

1% total energy metering error limit;
V-harmonic <= 10% of fundamental;
I-harmonic<=40% of fundamental
...

 

Line Frequency

47.5-62.5 Hz Active energy metering
47.5-52.5 Hz Reactive energy metering

 

These lines could mean a tombstone on my idea, unless it is intended that you could only expect the stated precision within these limits...

For my purpose a 10% error after the available  inner calibration and / or by means of an external correction table would be acceptable.

 

Another perplexity comes from the zero crossing signal.  The DS says it is transformed in 5ms pulses; it means a D.C. at just 200Hz.

I can't figure wether such wide pulses are given out just for external use or if they are part of inner phase measurement mechanism..

I am not interested in any integrated energy value, but I do need phase.

Altough this chip mai purpose is energy measurement, the D.S. says:

 

APPLICATION
... With the measurement function, the M90E26 can also be used in power instru-
ments which need to measure voltage, current, etc. ...

 

I know, too many questions and little ideas but confused.. Perhaps an aswer may only come by playing with the developement board...indecision

 

 

 

 

 

  • 1
  • 2
  • 3
  • 4
  • 5
Total votes: 0

Your best bet, surely, is to contact Atmel/Microchip Support, and/or discuss with your distributor FAE ...