Understanding Electrical Motor Failure Condition Data

[AprilByrd]

Hi Guys,

I have some data on a small electric motor being run continuously for a year. The problem is I only have the Headings and numbers and no units or information on what devices were used to measure the data.

The variables measured were Time, Speed, Temperature, Power, Powerfactor, Current and Voltage.

The problem is that I don't know what Kind of "Power" the Power variable is measuring. Let's call it XPower

I've calculated InputPower= PFxVxI, and at certain points in time its less than the XPower hence XPower can't be output power. Someone talked about Consumed vs Supply power, but wouldn't that have the same problem because consumed power should be less than supplied power?

I also considered Apparent vs Real figures but again apparent is always more than real power?

Both the Input power and Xpower have average values around 380 (Watts?) but the calculated input power fluctuates around the value about 4 times as much as the XPower.

I'm no electrical engineer and I think I may be missing something obvious.

I would greatly appreciate anyone who knows where I'm going wrong, or if I have the right to not know what's going on!

 

[vk6kro]

Power would be the power consumed by the motor.

You can calculate this from the input voltage and current because you are given the power factor.

Power factor = real power / apparent power
where apparent power = voltage * current.

You have already done this and it doesn't agree with the figures given, so then you need to query it with whoever supplied the data, if it matters. It could be a printing error.

If it doesn't matter, don't worry about it. You aren't doing anything wrong.

 

[AprilByrd]

Thanks for the reply vk6kro,

So even if the measured power was in VA instead of Watts, it should still match the other calculated power using the power factor that was measured?

Thanks so much I will query the data.

 

[TurtleMeister]

The device used to measure the data was probably a power analyzer. I worked for an AC induction motor manufacturer many years ago and we used Magtrol for all our motor testing. It is normal to have fluctuations in the measured values over a period of several seconds. But the average over a longer period of time should be pretty stable.

 

[AprilByrd]

Hi Turtle Meister, thanks for your reply.

When you say that there are fluctuations, do you know what kind of degree could exist which would be counted as these fluctuations?

It is possible that a snap shot measurement was taken at the times listed and hence the values will jump around a bit but sometimes the difference between measured and calculated value is 380W and 422W. Do you think that is too high to be counted as a "fluctuation" of the measurement device?

Many Thanks

 

[TurtleMeister]

Yes, 42 watts difference seems extreme. I don't know if the update interval for the different measurements are synchronized. You may be able to go to the website of the power analyzer's manufacturer and get more info. I would go with the measured values instead of the calculated values.

 

[vk6kro]

Thanks for the reply vk6kro,

So even if the measured power was in VA instead of Watts, it should still match the other calculated power using the power factor that was measured?

Thanks so much I will query the data.

There is a strict relationship between power, Volt-Amps and power factor, so that you can calculate one of them given the other two.
If you are given all three, then you can check for consistency of the data by using two of them to calculate the third.

If they don't agree, then you can query the source to see if there might be printing errors.

There is another possible explanation for this. If the data was changing and someone was writing it down, then it could change while they are doing that and the data may not be consistent.
I have a power analyser that only shows one reading at a time and I have to push a button to get from, say, voltamps to power factor to power to KWH etc. So, the data could change while I am doing this.

 

[I_am_learning]

How did you measured the power, Full Load or No Load? That can surely make difference.
Also keep in view that in a course of a year the performance of motor may change due to deterioation of the core, but I am not much sure on 'by how much'.

 

[TurtleMeister]

There is another possible explanation for this. If the data was changing and someone was writing it down, then it could change while they are doing that and the data may not be consistent.

That is a good point. The test data is for a period of one year. At my company, this was called a "life test". The motor was run at no-load for a long period of time. An operator would come in at regular time intervals and simply record the readings. Normally when I would take readings for calibrations checks I would watch the readout for a period of several seconds and average out the fluctuations in my head. If the operator was careless, and did not do this, then the fluctuations would show up in the data.

Edit: correction. The life test motors were run under normal load, not no load. I seem to remember "some" motors being test at no load, but most were tested with their normal operating load. This was a long time ago, so my memory is a little faded. :)

 

[AprilByrd]

Thanks everyone for all the replies.

How did you measured the power, Full Load or No Load? That can surely make difference.

It was a constant 'normal operating' load.


OK so just in case anyone's interested in the result, turns out that yes it is measuring the same input power. Which explains why both the values sit around the average.

It doesn't really sit quite well though since as vk6kro said, there is a strict relationship between the other variables; so its going down to poor measurement technique.

Unfortunately i just have to accept the results as they are as two separate measurements of input power and i'll explain the huge possible mistakes in my report.

Thank you so much everyone for all your help :)