Power Factor Correction Help 230V AC

Discussion in 'Power Electronics' started by cells, Jun 21, 2017.

1. cells Thread Starter Active Member

Oct 19, 2008
34
1
Hi Everyone,

Before I start, a quick question. Is there anywhere I can buy an off the shelf power factor correction devise sized anywhere between 60 - 80 micro farads? If not I will have to attempt to build one myself. So here goes...

I have a slush machine which is primarily a compressor/motor. My power meter shows a power factor of 0.6 its true power is about 1100 watts and its drawing about 8 amps of current. This is for the UK with 230 volts AC and 50Hz

I would like to correct the power factor towards 1 so I can add more slush machines to the same line without having to upgrade the line which would be very difficult to do as it would require removing the flooring. If my calculations are correct I need an 80 micro farad capacitor to bring the power factor towards unity. Can one of you please confirm I have the capacitor size correct?

I was thinking of wiring the capacitor to a plug with a bleed resistor inside the plug.
A bit like in this youtube video. You can skip to 1 min into the video to see what I mean

Unlike in the video which has multiple capacitors I was thinking just have 1 x 80 uf Capacitor and the load both on the extension lead and that connected to the mains.

I have a few questions.

What size should the bleed resistor be? Is there a certain specific type of resistor I should be using? I have found the following website which has a 52,000 ohm 3 watt metal oxide power resistor. Would that be the correct type and sizing?
https://www.rapidonline.com/truohm-3w-metal-oxide-resistors-5-66471
An online capacitor discharge calculator says it would take ~10 seconds to discharge from 300 volts to 30 volts using a 52,000 ohm resistor. Would 8 seconds be considered sufficiently safe or should I size a faster bleed resistor like 22,000 ohm which the online calculator says would take ~4 second to discharge from 300v to 30v.

Can I put the bleed resistor inside the plug or will it heat up too much and melt/burn or otherwise cause a problem if installed inside the plug?

Is there anything I have missed or not considered?

Also to note I believe there is a start capacitor inside the slush machine already, will there be any problem or interference or something I am not aware of if I correct for power factor on the line?

Also what would happen if the mains line was disconnected for a few seconds and then reconnected. Would the capacitor discharge rapidly into the mains causing a spark or any problems?

Will this power factor correction interfere with the RCD or trip it in any way. (my guess is no but its best to ask)

Any help and advise greatly appreciated
Many Thanks
Kaya

2. Dodgydave AAC Fanatic!

Jun 22, 2012
8,146
1,410
Go for 40uF capacitors in parallel or one 80uF, and use a 22- 47k bleed resistor across the capacitor terminals. It wont effect the Rcd as this detects Earth leakage current.

3. tsan Member

Sep 6, 2014
90
16
There are energy saving devices on the market that seems to be compensating capacitors. They are said to be a scam (as energy savers or reducing electricity bill) but on your case these units might work. An example
http://kvarenergyproducts.com/
Kvar etc. similar devices are also available on Ebay.

I think that slush machine have an intermittent load. If you target power factor of one when the machine is running, you have an over compensation situation when it is not running with full load. Overcompensation is usually avoided for various reason like resonances etc. Perhaps good target would be something like 0.9 instead of one. Or more intelligent system that switches the capacitors when required. Have you checked that there is no harmonics on the voltage? Harmonics causes additional current flow on the compensation capacitors.

Have you checked that the cable/wiring including installation conditions doesn't allow one size bigger fuse or circuit breaker? Sometimes wire size is bigger than the circuit breaker size suggests.

Even 4 seconds is not safe. If someone disconnects the plug it can easily happen that he/she touches with the plug to his/hers other hand and the plug terminals have a high voltage. More permanent installation is needed if safety is important (or a lot of tape so that it is not possible to disconnect the plug). Or perhaps a box where there is relay with 230 VAC coil. The relay would connect the capacitors to the circuit only when AC supply is connected. Relay works in few hundred milliseconds so the capacitor would be quite quickly disconnected from the plug when the plug is disconnected.

Compensation capacitors in industrial three phase system are known to create over voltage transients. I have never seen these "energy savers" but it seems to be that some of them have built-in over voltage protection.

4. tcmtech Distinguished Member

Nov 4, 2013
2,808
2,705
You can cheat your amp draw a little using this method but technically you are still limited to only drawing about 80% of whatever your wiring and overcurrent protection for that circuit are rated at.

Ideally you would be ahead of things to put a smaller PFC capacitor in each unit right across the motor or compressor rather than externally outside each unit with one larger one.

Personally I rarely use PF compensation on anything but when I do I aim for about .95 - 1 at full load or whatever near that the most common size of capacitor I can get gets me to.

Also overcompensating a bit on a local circuit during different loading conditions really doesn't bother much being more than likely there are other inductive loads on the main system that will use it.

5. cells Thread Starter Active Member

Oct 19, 2008
34
1
Thank you both.

TSAN, I have no knowledge at all about what harmonics are so I have no idea if it is a problem or how I would go about solving it. Any more info would be appreciated. Yes I was aware that the machine lowers its power use after the product is fully frozen it appears the motor doesn't work as hard so it draws less real power and less amps (roughly 800 watts real power and 6 amps) once the product is frozen. I did consider using a lower farad capacitor so that on average its closer to unity but I figured the line is only near its limit during the higher power ramp ups so it would be best to size the capacitor for when the machine is at its peak needs of 1.1kw (real) 8amp

Regarding the power factor devices off ebay etc I did consider them but when I watched a few youtube tear down videos they all had small capacitors of about 3-5 uf so I would need far too many of those

Regarding the safety. We rarely ever unplug the machine. If for some reason we lose line power (which very rarely happens) then its quite likely the capacitor will just discharge into the slush machine almost instantly. The only scenario I can imagine is if the machine is off and we lose the line then the capacitor will have a charge on it hence the discharge resistor. So long as in this scenario it is not disconnected in the first few seconds the discharge resistor should lower the voltage sufficiently to be safe. I think I may add another 3 watt discharge resistor on the extension cable to double the discharge speed.

I believe the compressor has a start cap on it, I will check tomorrow to see what rating it is.
I will also check to see if there is the space inside the machine to put the capacitor in the machine across the compressor or across the initial power line just behind the fuse. That way it might negate the need for a bleed resistor. Only small concern is that the machine vibrates due to the compressor and its hot inside due to the heat pump effect and this might not be an ideal environment for a capacitor but hopefully within limits

I've ordered a couple of 80 uf caps so I will see how this goes

6. tcmtech Distinguished Member

Nov 4, 2013
2,808
2,705
If its a common metal can type they can take temperatures well over 200F without issue. They are the same type of capacitors used in HID lighting ballasts which can be in extreme temperature environments your mixer motors would burn up in after a short while

As for mounting most often I attach them in special application work with common silicone. It holds them very well even in high vibration applications. Just put enough on to give them a good bond and let it cure for a day.

7. tsan Member

Sep 6, 2014
90
16
Most likely there will be no problem at all with harmonics but information about harmonics is for example on
http://www.aerovox.com/Products/PowerFactorCorrectionCapacitors.aspx
and there is a pdf file
http://www.aerovox.com/Portals/0/PDFs/Technical Notes/AN101312 PFC Guide.pdf
There is a chapter "Understanding harmonics".

Those on a small plastic box really seem to be almost nothing. Some of the units installed in USA (based only on Youtube video) really seem to compensate reactive current. Perhaps for that reason there are services that a person come to measure reactive current to dimension the correction.

My concern was for the situation that the capacitor is connected to an extension cord (like on the video on your first mail) that is easily available. Then someone could just disconnect the capacitor box plug in order to use that socket to charge a phone or something.

8. cells Thread Starter Active Member

Oct 19, 2008
34
1
Thanks. silicone idea seems interesting it should help absorb some of the vibration. I was thinking just attach it to a study part of the machine using cable ties. The silicone might cause local heating where it effectively insulates the capacitor. So I will have a think about the two and which might be best.

My concern was for the situation that the capacitor is connected to an extension cord (like on the video on your first mail) that is easily available. Then someone could just disconnect the capacitor box plug in order to use that socket to charge a phone or something.

Yes that makes sense and I did not even consider that risk since I would not do that but as you note its a risk that someone else might do that. If I do go with the extension cord idea I will make sure the capacitor plug on the extension cord is wound up with duck tape with a note to not remove. After these discussions here I feel putting the cap inside the machine would be the safest option that way in any event where the power is removed or the machine unplugged the capacitor should near instantly discharge into the compressor motor.

9. tcmtech Distinguished Member

Nov 4, 2013
2,808
2,705
The capacitors themselves don't give off enough heat to ever be an issue.

#12 likes this.
10. cells Thread Starter Active Member

Oct 19, 2008
34
1
I had a good look inside the slush machine today and there was already an 88 uf capacitor connected to the compressor.
Looking at google for the model number I believe it is a electrolytic start capacitor and wiki seems to suggest these types of capacitor just work for a few seconds until the compressor is up to speed before a relay or some other type of switch disconnects the capacitor from the motor. I'm sure the manufacturer must know what they are doing but why would they use an 88 uf start cap if they could have used a run cap instead and produced a machine with a good power factor.

I must admit I don't feel very confident about this experiment but since I have purchased the run capacitors I might as well give it a go. I will probably take the whole machine outside and run it off an extension cable and let it run for a few hours so as to minimise any risks of fire and to make sure its safe.

11. tcmtech Distinguished Member

Nov 4, 2013
2,808
2,705
They don't care about power factor. Its just one more item that adds to manufacturing costs they don't esee a return on investment for adding. Especially given it's a lower power machine and the end user is not likely being billed for power factor issues anyway.

To me this whole thread strongly implies you are trying to find a way to cheat your electrical codes and well defined total power draw limits you are allowed to have on one circuit because you are too cheap to either do the proper supply upsizing or run an extension cord from a separate circuit or relocate some of the load to another location on different circuit.

12. cells Thread Starter Active Member

Oct 19, 2008
34
1

Well yes I have already said that running a new line would be costly as it would involve ripping out tiles and replacing the flooring.
As a guess it would probably cost in excess of £3,000 to do that, while trying to fix the power factor might cost less than £50

13. cells Thread Starter Active Member

Oct 19, 2008
34
1
Good news. It seems to work absolutely fine.
It takes the machine amps down from approx 7.5 amps to 4.5 amps which is a significant improvement.

I was concerned that the bleed resistor would heat up too much in the plug but that seems fine too. I went with a 47k ohm resistor so it should discharge to a safe voltage in about 7 seconds. I may in future move the capacitor to inside he machine across the motor itself in which case the bleed resistor would not be necessary.

ronv likes this.
14. cells Thread Starter Active Member

Oct 19, 2008
34
1
Assuming my meter is correct, it shows that the capacitor itself draws ~6.3 amps of current (at 240v AC) which was a surprise to me. I did not even think or consider that it would be drawing current when not in use.

Putting the capacitor inside the machine to only operate when the main compressor motor is on would be a very good idea. As the setup currently stands the slush machine + power factor correction setup draws 4.5 amps when the slush machine is on and 6.3 amps when the slush machine is off. That seems a silly design so more incentive to put the capacitor inside the machine so it is only on when the machine is on. Perhaps I could just turn the whole setup on/off at the mains but that would not be a neat a solution.

Just adding this as it could be useful for future reference to someone who does a google search for a similar problem

15. tcmtech Distinguished Member

Nov 4, 2013
2,808
2,705
It's reactive apparent power not real wattage power so it has no real play in the overall power usage your utility meter measures so the ~6.3 amps it appears to be drawing is just canceling out ~6.3 amps of inductive apparent power load somewhere else in the system.

But, yes. Ideally each power correction capacitor should be sized for each compressor and connected inside the machines at the compressor not externally as you have it now.

16. #12 Expert

Nov 30, 2010
18,076
9,686
Been there, done that. 80uf across my refrigerator compressor changed the amps from 5 A to 3 A (in a 125 VAC system).
Then I thought about the contacts which power the compressor.
If they close at the peak of your power line voltage, you have 339 volts firing into an empty capacitor and the thermostat contacts weld together.
Think. Are you creating a reliability problem?

17. cells Thread Starter Active Member

Oct 19, 2008
34
1

Can you expand on this?

For these machines there is no thermostat they work by engaging a switch on/off depending on the thickness of the slush.

I have put it behind a switch so the capacitor is only engaged when the compressor and cooling fan are engaged. So they are all on or all off. I think this is quite safe as the cap has a large compressor and cooling fan to dump its charge into. Maybe at peak voltages the cap has 3 joules of stored energy and the compressor + cooling fan use about 1000 watts so they can take the discharge.

My one other concern or bit I do not quire understand is how (if at all) this addition of the capacitor impacts on the compressor/fan-motor in terms of starting or stopping. When the switch is disengaged I wonder if there is some sort of interplay between the cap and motors. The motors do they temporarily become generators? Imagine by chance I close the switch at the 0 volt part of the AC cycle. Well the compressor and fan motor still have momentum so do they charge up the cap at that point? Does this mean the compressor and fan motor effectively have a load placed on them so slow down faster? Likewise on start up does the cap give the compressor and fan motor a boost?

18. cells Thread Starter Active Member

Oct 19, 2008
34
1
I had the following observation which I can not explain. Maybe one of you will understand what is going on here.

Its for a different machine I was testing. Imagine the compressor plus there is a fan to cool the condenser. The fan says on it 25/110 watts which is confusing I think the fan draws 110 watts but is rated for 25 watts (is that air watts?) anyway that bits not important

The bit that I do not understand is that when I took the fan out of the circuit somehow the total power used went up? That was very confusing and I cant explain it or understand it

With the fan in the circuit
670 watts 5.5 Amps 0.49 Power Factor 242 volts

Without the fan in the circuit.
710 watts 5.3 Amps 0.56 Power Factor 242 volts

As you can see the amps went down, which was expected, but the power usage went up even though I removed 110 watt load
The fan motor is an inductive load which can be seen as the power factor improved without the fan in the circuit

I thought maybe the fan and compressor were wired in series but that makes no sense they are both 230 volt AC so you could not put them in series. I am at a loss as to what the hell is going on. Take out a 110 watt load from a circuit and somehow it draws 40 watts more power?

PS there was no power factor correction capacitor etc on that machine.

19. tsan Member

Sep 6, 2014
90
16
It is possible. For this reason I mentioned avoiding over compensation earlier. But at that time, the cap was to be installed to extension cord outside the machine and as tcmtech wrote, there is no over compensation issue in that case. This site has more information,

http://www.electrical-installation.org/enwiki/Power_factor_correction_of_induction_motors

From that, page:
How self-excitation of an induction motor can be avoided
When a capacitor bank is connected to the terminals of an induction motor, it is important to check that the size of the bank is less than that at which self-excitation can occur

More info is on the same page.

20. #12 Expert

Nov 30, 2010
18,076
9,686
So a switch connects the compressor to the power. Is that switch so different from a thermostat that current does not flow through its contacts? And when that current does not flow, it can not weld the contacts together during a start surge into the capacitor because current doesn't flow through the switch contacts. I don't believe that is true.
You stop the fan and the machine uses more power. Hmmm...do you think the pressures increase and the compressor works harder for lack of air flow?