The power factor of SMPS power plant

[T.M.Haja Sahib]

A SMPS type float rectifier has been provided in a cell tower station (BTS). Its capacity is 50A (DC) and is working on 415V, 50 HZ AC supply.

The question is: Whether the power factor of the SMPS power plant depends on its load.

We had low power factor penalty when the total energy consumption is low during the last two bimonthly periods in this year. The average power factor is around 0.70. But when the energy consumption is high during bimonthly periods last year, the average power factor was around 0.98. At present, the instantaneous power factor as observed from the energy meter provided in the site is 0.97, but the average power factor is 0.67.

There is so far no power factor correction equipment in the site. I plan to connect a 3 phase 0.5 KVAR capacitor permanently across the power lines in an effort to improve the average power factor.

Is it effective?

Your inputs please.

 

[AdrianWint]

I think the problem here may be with the instrumentation.


I have had issues in the past whereby the metering was being confused by the discontinuous current waveform taken by the rectifier. When this is a significant proportion of the load the instrument mis-reads the discontinous sinewave and harmonics, interpreting these as as a phase shift, thus thinking that it is seeing a reactive component to the current when non is present.

I first came across this when looking at the power drawn by a resistive load fed via a thyristor type power controller - the instrumentation was indicating a significant inductive component to the load current when, by definition, none can be present in a resistive load. This was confirmed by connecting the same load direct to the same supply via the same meter but without the controller.

If this is happening then adding capacitance wont slove the problem, it may even make it worse by increasing the system voltage & turning the power factor leading.

Adrian

 

[T.M.Haja Sahib]

Adrian
Thanks for your reply.
I think the thyristor type power controller you discussed was not power factor corrected and so it fed back a distorted current/voltage into the incoming side measuring meter. This might have caused the meter to display erroneous reading. But in case of SMPS type power plant, the power factor is corrected and no distortion of power supply.

 

[Besoeker]

A SMPS type float rectifier has been provided in a cell tower station (BTS). Its capacity is 50A (DC) and is working on 415V, 50 HZ AC supply.

The question is: Whether the power factor of the SMPS power plant depends on its load.

We had low power factor penalty when the total energy consumption is low during the last two bimonthly periods in this year. The average power factor is around 0.70. But when the energy consumption is high during bimonthly periods last year, the average power factor was around 0.98. At present, the instantaneous power factor as observed from the energy meter provided in the site is 0.97, but the average power factor is 0.67.

There is so far no power factor correction equipment in the site. I plan to connect a 3 phase 0.5 KVAR capacitor permanently across the power lines in an effort to improve the average power factor.

Is it effective?

Your inputs please.

I'm with Adrian on this.
Power factor comes in two flavours. Displacement and distortion.
Capacitors can correct the former but not the latter.

Most SMPS units I've come across first rectify the supply and that results in non-linear current but usually not a displacement.
Non-linear current causes harmonic distortion.
Harmonics and capacitors don't make good bedfellows.

 

[T.M.Haja Sahib]

Bes
Then what is your solution for the present low power factor problem in the BTS?
Do not say:shut down the BTS.....

 

[Besoeker]

Bes
Then what is your solution for the present low power factor problem in the BTS?
Do not say:shut down the BTS.....

I'd first want to determine whether there is a problem with poor power factor or whether it is being incorrectly displayed.

Looking at supply voltage and current with an oscilloscope would help to determine what is causing poor power factor
and the nature of it. Armed with that information, I'm sure a solution to the problem could be found.

As a matter of interest, what is the battery nominal voltage and, if you have it, the Ah rating.

 

[T.M.Haja Sahib]

I'd first want to determine whether there is a problem with poor power factor or whether it is being incorrectly displayed.

Looking at supply voltage and current with an oscilloscope would help to determine what is causing poor power factor
and the nature of it. Armed with that information, I'm sure a solution to the problem could be found.

As a matter of interest, what is the battery nominal voltage and, if you have it, the Ah rating.

The riddle is the instantaneous power factor displayed by the energy meter is around 0.97 (may not be for 24 hours for sure), but the average power factor displayed by the same meter is 0.67.

I do not believe the scope display will reveal any thing, because capacitors installed to improve power factor in medium size telephone exchanges, where SMPS power plant is operating have not caused any problem and the power factor is maintained near unity.

The battery size in the present case is 400Ah and battery voltage is 50V.

 

[Besoeker]

I do not believe the scope display will reveal any thing, because capacitors installed to improve power factor in medium size telephone exchanges, where SMPS power plant is operating have not caused any problem and the power factor is maintained near unity.

Your opening posts asked about the relationship between the SMPS and power factor.
You got honest answers from Adrian and myself. Capacitors are not the way to go to correct the SMPS - if that's the cause of your problem.

The oscilloscope measurements I suggested would would readily reveal whether this is case.
And it's quick and simple to do.

You asked me for a solution and I told you how I would go about it based on your opening post
But, you are rejecting it. And that's fine. What's your solution if mine is rubbish?

The battery size in the present case is 400Ah and battery voltage is 50V.

So pretty small, then. About 2.5kW output rating and, I suppose, no certainty that it will run at rated power all the time.

What is the total power consumption of the exchange? How does that compare with the 2.5kW SMPS rating?

 

[T.M.Haja Sahib]

Bes
The SMPS power plant under discussion here is power factor corrected and so it does not distort the incoming power supply. So connecting capacitors across the incoming power lines do not creat any problem as evidenced in other working telephone exchanges with SMPS power plants.

I request your comment on this:
''The riddle is the instantaneous power factor displayed by the energy meter is around 0.97 (may not be for 24 hours for sure), but the average power factor displayed by the same meter is 0.67.''

My proposed solution is to connect a 0.5 or 1.0 KVAR capacitor permanently across the power lines to see whether it avoids the low power factor penalty by improving the average power factor from the present 0.67 t0 0.85 or better.

For bimonthly period ending December 2011, units consumed is 3220. Average Power factor was 0.99 and no low power factor penalty.

For bimonthly period ending February 2012, units consumed is 2950. Average Power factor was 0.98 and no low power factor penalty.

For April 2012, units consumed is 600. Average Power factor was 0.72 and there was low power factor penalty.

For June 2012, units consumed is 270. Average Power factor was 0.76 and there was low power factor penalty.

 

[Besoeker]

Bes
The SMPS power plant under discussion here is power factor corrected and so it does not distort the incoming power supply.

You mean the SMPS takes true sinusoidal input current?

I request your comment on this:
''The riddle is the instantaneous power factor displayed by the energy meter is around 0.97 (may not be for 24 hours for sure), but the average power factor displayed by the same meter is 0.67.''

Which is why I suggested puting a scope on it. That would give you independent real-time measurements.
But you keep telling me I don't understand things so you probably think it's a daft idea....

For bimonthly period ending December 2011, units consumed is 3220. Average Power factor was 0.99 and no low power factor penalty.

For bimonthly period ending February 2012, units consumed is 2950. Average Power factor was 0.98 and no low power factor penalty.

For April 2012, units consumed is 600. Average Power factor was 0.72 and there was low power factor penalty.

For June 2012, units consumed is 270. Average Power factor was 0.76 and there was low power factor penalty.

I assume that the units you mention are what most of us consider to be the kWh.
On that basis, average power consumption for the periods ending in November and December was a little over 2kW in both cases.
For April 2012 it's about 0.8kW and for June it's 0.375W. Average.

Why is the average power in June about one sixth of that in December? The answer to that may well explain your difference in power factor.

And do you really get penalised to any significant extent for poor power factor on a load of 375W?

 

[T.M.Haja Sahib]

You mean the SMPS takes true sinusoidal input current?

Yes, my lord. The SMPS power plant under discussion here, unlike a typical computer SMPS power supply, has additional circuitry incorporating inductance to smooth out peaked nature of current and limiting the harmonic content to present a sinusoidal current draw at unity power factor.

Which is why I suggested puting a scope on it. That would give you independent real-time measurements.
But you keep telling me I don't understand things so you probably think it's a daft idea....

The site is in a remote hilly area and it is easy for me to send a contractor to fix a capacitor of size 0.5 or 1.0 KVAR to see what happens next. But it is a different matter to do with a scope. If you give me more technical reasons for doing it, I shall be motivated to do so.

I assume that the units you mention are what most of us consider to be the kWh.
On that basis, average power consumption for the periods ending in November and December was a little over 2kW in both cases.
For April 2012 it's about 0.8kW and for June it's 0.375W. Average.

Why is the average power in June about one sixth of that in December? The answer to that may well explain your difference in power factor.

And do you really get penalised to any significant extent for poor power factor on a load of 375W?

Yes. The penalty amount is very small. It is only INR740. But still it will give rise to statutory Audit objection and it should be settled.

 

[Besoeker]

Yes, my lord. The SMPS power plant under discussion here, unlike a typical computer SMPS power supply, has additional circuitry incorporating inductance to smooth out peaked nature of current and limiting the harmonic content to present a sinusoidal current draw at unity power factor.

Well then, if it takes sinusoidal current at unity power factor, it isn't the cause of you poor power factor is it?
So it's a bit odd that you asked this question in the first place:

The question is: Whether the power factor of the SMPS power plant depends on its load.

The site is in a remote hilly area and it is easy for me to send a contractor to fix a capacitor of size 0.5 or 1.0 KVAR to see what happens next.

You didn't answer my question.
Why is the average power in June about one sixth of that in December?
And at 0.375kW, even the 0.5kVAr is too much if its to correct lagging load current. At 0.375kW you will get leading PF. And that was average kW. Excursions to a lower than average will result in the leading PF getting worse.

What are the other loads at the exchange besides the SMPS?



But it is a different matter to do with a scope. If you give me more technical reasons for doing it, I shall be motivated to do so.

Yes. The penalty amount is very small. It is only INR740. But still it will give rise to statutory Audit objection and it should be settled.[/QUOTE]

 

[wiigelec]

Put a power quality recording data analyzer on the system for a month so you can see what the meter sees...

http://dranetz.com/portable/powervisa

 

[T.M.Haja Sahib]

Well then, if it takes sinusoidal current at unity power factor, it isn't the cause of you poor power factor is it?
So it's a bit odd that you asked this question in the first place:

I asked the question because it seems that the power factor of the SMPS power plant depends on the load. So your question may be wrong.

Why is the average power in June about one sixth of that in December?

The SMPS power plant had six modules of each 25A capacity operating during bimonthly periods when the average power factor was around 0.98 and no low power factor penalty. The cell tower in charge informed me that due to faults four modules were removed and only two modules were operating after those no low power factor penalty bimonthly periods.

And at 0.375kW, even the 0.5kVAr is too much if its to correct lagging load current. At 0.375kW you will get leading PF. And that was average kW. Excursions to a lower than average will result in the leading PF getting worse.

I want to draw your kind attention to the following again.

The riddle is the instantaneous power factor displayed by the energy meter is around 0.97 (may not be for 24 hours for sure), but the average power factor displayed by the same meter is 0.67.

May be at some time in the day or night the power factor may become very low so that though the instantaneous power factor is high most of the time, the average power factor becomes low. My plan is to connect a 0.5 or 1.0 KVAR capacitor permanently across the incoming power lines to offset this effect so that the average power factor is also high enough.

What are the other loads at the exchange besides the SMPS?

Two nos each 100w florescent tube light fittings. Two nos 1.5 TR air-conditioners which are not operating for the past one year.

 

[Besoeker]

I asked the question because it seems that the power factor of the SMPS power plant depends on the load. So your question may be wrong.

"to present a sinusoidal current draw at unity power factor" is what you said.
Either it does or it doesn't. That could be simply ascertained by sticking a scope on the supply voltage and current.

I want to draw your kind attention to the following again.
"The riddle is the instantaneous power factor displayed by the energy meter is around 0.97 (may not be for 24 hours for sure), but the average power factor displayed by the same meter is 0.67."

May be at some time in the day or night the power factor may become very low so that though the instantaneous power factor is high most of the time, the average power factor becomes low. My plan is to connect a 0.5 or 1.0 KVAR capacitor permanently across the incoming power lines to offset this effect so that the average power factor is also high enough.

I want to draw your kind attention to the following again.
If you have 0.375kW at 0.76 pf then even 0.5kVAr will result in a leading pf which not usually a good thing especially if the supply isn't particularly robust and it might not be given that site is remote.

 

[T.M.Haja Sahib]

..... not usually a good thing especially if the supply isn't particularly robust .....

I have no idea why the supply need be robust particularly for a leading power factor. I request an explanation.

 

[Besoeker]

I have no idea why the supply need be robust particularly for a leading power factor. I request an explanation.

It can result in the voltage being higher than you'd ideally want.
You need to draw the phasor diagrams to see why.

 

[T.M.Haja Sahib]

It can result in the voltage being higher than you'd ideally want.

I think it will happen for robust supply. For weak supply leading power factor may compensate for the voltage drop and the voltage may be more or less stabilized and so leading power factor load may be good for weak supply.

 

[Besoeker]

I think it will happen for robust supply. For weak supply leading power factor may compensate for the voltage drop and the voltage may be more or less stabilized and so leading power factor load may be good for weak supply.

I disagree.
But if you want to try that rather than undertaking the simple checks I suggested first, fine.

 

[T.M.Haja Sahib]

...... undertaking the simple checks I suggested first,

I would like to hear more of your valuable comments just like those of other members here to execute the easiest one.