# measuring power factor

Discussion in 'Electronic Design' started by Pinchy, Sep 24, 2004.

1. ### PinchyGuest

I intend to design a device to measure the power factor without the
use of a microcontroller. In the design, the measured angle will be
given by a simple DC voltage, e.g. 30° = pi/6 rad = 0.523 V

A problem that occurs is calculating the cosine of this voltage. In a
discrete setup, I will use Taylor series and realise it with analog
multipliers and a few operational amplifiers.

Before starting this, I would like to know if a single IC exists that
can make this calculation at once, eg 0.523 V IN results in cos(0.523)
= 0.866 V OUT

Thanks for help

(designer has a degree of engineer in elektronics)

Geert

2. ### Spehro PefhanyGuest

Check these out:

http://www.analog.com/library/analogDialogue/bestof/pdf/06_3.pdf

not sure if either is still available...
(Maybe designer should learn to use microcontroller)

Best regards,
Spehro Pefhany

3. ### John FieldsGuest

---
Why on earth would you want to do that?
---
---
Sure.

The single IC is the ROM, a Read-Only-Memory with a cosine lookup
table burned into it.

5. ### Tim WescottGuest

Power factor can either be defined as pf = cos(theta_v - theta_i), where
theta_v and theta_i are the phases of the voltage and current,
respectively, and we assume that all waveforms are pure sinusoids.

I better definition in my mind is the one that goes:

pf = -----------------.
(V_rms)*(I_rms)

This definition accommodates all sorts of situations where the "pure
sinusoid" definition doesn't, like pulsed currents (even on a DC line),
harmonic currents from rectification, or anything else you can dream up.

If you are not bent on measuring the sign of the reactive vars, then you
can build your power factor meter with a couple of RMS chips and a few

And you won't have to mess with trig functions.

6. ### Ken SmithGuest

Why are you using such a messy way of doing it?

I suggest:

Run the voltage into an AGC

Run the current into a comparitor.

Use the output of the comparitor to sync. demod. the AGCed voltage.

7. ### Auxiliary UserGuest

New definition of "Single" today, John?

Hello!
Zuzej Maaya

8. ### Spehro PefhanyGuest

You could use an ADuC832 or something like that, and get single chip
with the above block diagram, eh?

Best regards,
Spehro Pefhany

9. ### Pooh BearGuest

Given the loads today on the AC network, It would seem foolish to measure
power factor by the old fashioned phase angle approach.

Your definition is much more appropriate.

Graham

10. ### Ken SmithGuest

Use one of the Cygnal micros and you get the ADC the table and the DAC all
in one chip. That would do it.

11. ### The PhantomGuest

You haven't told us what accuracy you require.

12. ### Tom SeimGuest

Yeah, a Microchip PIC micro w/ADC.

14. ### keithGuest

Because his professor asked the question, and he's been skipping class?

15. ### Tom SeimGuest

He asked if there existed a single IC that can do this, and there is.