Power meter - calculating cost

I'm looking at build one of those Kill A Watt meters. I know I could buy one but where is the fun in that. Going to be using an Arduino with a current sensor.

I would like to use it to work out the cost of a AC device which is plugged into the unit and run for a certain period. Say for example I plug my TV in and run it for 30 mins - how much could it cost?

Originally I thought it would be simple using P=VI but after some more reading up it looks like I will need to take the power factor into account.

A few questions:
1) I've got a sensor (**broken link removed**) which I will use to determine the current, I know the voltage - 220V in South Africa, what else would I need to work out the cost?
2) If I didn't take the power factor into account how inaccurate would the results be? How do I determine the PF is I should be using it?

How do these kill a watt meters work - do they take the PF into account? How do they do it?

Depends on the device, power factor with many types of loads is a nonissue.

Sceadwian said:

Depends on the device, power factor with many types of loads is a nonissue.

Click to expand...

THanks - I"m going to be using it to measure loads around the house. Fridge, fish tank lights, dishwasher, toaster, TV, computers etc.

I agree with Sceadwian. For most residential loads the power factor is a non issue. A Google will bring up several Arduino projects to do what you want to do. I may be wrong but your sensor while a good choice is low current as in I believe about 5 Amps max. That should be fine with 220 VAC mains.

Ron

Reloadron said:

I agree with Sceadwian. For most residential loads the power factor is a non issue. A Google will bring up several Arduino projects to do what you want to do. I may be wrong but your sensor while a good choice is low current as in I believe about 5 Amps max. That should be fine with 220 VAC mains.

Ron

Click to expand...

Great stuff. I got the 20A version.

Sceadwian said:

Depends on the device, power factor with many types of loads is a nonissue.

Click to expand...

It depends upon the accuracy you want. Resistive loads are no problem but for loads with lower PF such as motors or electronic devices, the power calculation could be off by 20% or more if you don't include the PF.

If you measure the voltage and current by average or peak you will have power factor problems. Even if you find the RMS V and RMS I and then multiply them together you will have errors.
Power factor problems will be taken care of if you over sample. Measure the V and I (at the same time) then multiply to find power, do this 100x60hz (50hz).

ronsimpson said:

If you measure the voltage and current by average or peak you will have power factor problems. Even if you find the RMS V and RMS I and then multiply them together you will have errors.
Power factor problems will be taken care of if you over sample. Measure the V and I (at the same time) then multiply to find power, do this 100x60hz (50hz).

Click to expand...

Thanks - Think I"ll do that.

ronsimpson said:

If you measure the voltage and current by average or peak you will have power factor problems. Even if you find the RMS V and RMS I and then multiply them together you will have errors.
Power factor problems will be taken care of if you over sample. Measure the V and I (at the same time) then multiply to find power, do this 100x60hz (50hz).

Click to expand...

You forgot to mention that you average those 100 values to get the RMS power for one cycle.