question

durval avatar image
durval asked

Trying to size a Victron Phoenix for a bad power-factor load, how to proceed?

I'm trying to size a Victron Phoenix inverter to power a refrigerator. My power meter indicates it consumes around 700W peak (immediately after the compressor turns on, lasting less than 2s) then stabilizes at around 70W, *but* with a power factor of only 0.60.

I have two questions:

1) To what power should I size the inverter?

2) Would adding an appropriate capacitor in parallel to the compressor motor terminals (as explained here: https://youtu.be/qf_ZLIYT0Ro) allow me to use a smaller inverter?

Thanks in advance.

EDIT: just keeping everyone posted, I followed the instructions on the video above to calculate and add a capacitor in parallel to the refrigerator, and was able to correct its power factor from 0.60 to 0.98; as a consequence, the VA went way down and is practically identical to the Watts value now, and this made it much easier to size and buy my Victron inverter (and also allowed me to buy a much smaller one than I would have to). Thanks to everyone who contributed.

MultiPlus Quattro Inverter ChargerPhoenix Inverterpower factor
2 comments
2 |3000
Toggle Comment visibility. Current Visibility: Viewable by all users

Up to 8 attachments (including images) can be used with a maximum of 190.8 MiB each and 286.6 MiB total.

Matthias Lange - DE avatar image Matthias Lange - DE ♦ commented ·
A capacitor would maybe increase the power factor but it would not lower the inrush current.
0 Likes 0 ·
durval avatar image durval Matthias Lange - DE ♦ commented ·

AFAIK reducing the power factor, in an inverter, means reducing both the starting ("inrush") current and the running current. This seems to be confirmed by the video I linked in my original post.

0 Likes 0 ·
2 Answers
Mike Dorsett avatar image
Mike Dorsett answered ·

minimum size of inverter would need to be 1200W; you calculate this as the peak load current x the voltage- i.e. the VA rating. The inverter does have some peak capacity, but stressing it repeatedly to this limit is not a good idea. I would not try to compensate the refrigerator. BTW, is the power factor 0.6 during start, or 0.6 during run?

1 comment
2 |3000
Toggle Comment visibility. Current Visibility: Viewable by all users

Up to 8 attachments (including images) can be used with a maximum of 190.8 MiB each and 286.6 MiB total.

durval avatar image durval commented ·

@Mike Dorsett:


> minimum size of inverter would need to be 1200W; you calculate this as the peak load current x the voltage- i.e. the VA rating.

You mean, 700 / 0.6 = ~1167, so we round it up to 1200W, correct?


> The inverter does have some peak capacity, but stressing it repeatedly to this limit is not a good idea.

I see on the datasheet the Phoenix 12/1200 has a 2000W peak power, so I think the remaining capacity should be more enough for the remaining stuff I plan on eventually powering up with it (my notebook which takes about 100W max).


> I would not try to compensate the refrigerator.

Why not?


> BTW, is the power factor 0.6 during start, or 0.6 during run?

Both. My power meter shows a starting PF of around 0.7, but it trends slowly to 0.6, reaching it after about 5 minutes running.


0 Likes 0 ·
rslifkin avatar image
rslifkin answered ·

I've got an icemaker on my boat that I've seen a brief (fraction of a second) spike to ~1800 watts, then runs at ~150 watts. I haven't checked the power factor on it though. It runs fine from my Multiplus compact 12/2000/80. Only issue is on startup. The compressor sometimes stutters slightly on startup and the AC output voltage from the inverter dips a bit (you can see a flicker in a lightbulb to go along with the stutter).

However, the issue isn't the inverter. It's the batteries. The sudden load spike is big enough that the battery voltage dips significantly, as the lightbulb test shows the stutter on the 12v side as well. If I'm underway with both engine alternators feeding power to the battery bank it starts more nicely. Solar doesn't help much, as the MPPTs can't ramp up fast enough. As battery SoC gets lower, the problem does get slightly worse, but not much.

I've been debating whether a large capacitor at the 12v input of the Multiplus might help carry the system through the compressor start better. That, or it's time to investigate a soft start for the icemaker compressor.

7 comments
2 |3000
Toggle Comment visibility. Current Visibility: Viewable by all users

Up to 8 attachments (including images) can be used with a maximum of 190.8 MiB each and 286.6 MiB total.

durval avatar image durval commented ·

> However, the issue isn't the inverter. It's the batteries. The sudden load spike is big enough that the battery voltage dips significantly, as the lightbulb test shows the stutter on the 12v side as well.

Can you please tell us what batteries (type/number/capacity) you have, as well as your other loads besides the icemaker?


> I haven't checked the power factor on it though.

It's cheap and easy to measure, shown immediately by my power meter: https://www.amazon.com/Poniie-PN2000-Electricity-Electrical-Consumption/dp/B0777H8MS8 (I believe the more common Kill-A-Watt meter would also show it, have to use the Poniie because I use 220V).


> I've been debating whether a large capacitor at the 12v input of the Multiplus might help carry the system through the compressor start better. That, or it's time to investigate a soft start for the icemaker compressor.

As instructed by the video I linked, I think it's better to try and put the capacitor directly in parallel with the icemaker motor terminals. But I wouldn't just try and guess the capacitance, as the video shows it has to be calculated (I've read elsewhere that an improper capacitance can actually make matters worse).



1 Like 1 ·
rslifkin avatar image rslifkin durval commented ·
I'd definitely talk to the icemaker manufacturer before making a change to the unit, as they're known to have pretty good support. And I do have a Kill-A-Watt sitting around, I just haven't had a chance to plug the icemaker into it yet (and I don't think power factor is the root of my issue anyway).


In my case, the startup stutter is unchanged whether there are other loads running or if it's just the icemaker. Batteries are 2x Fullriver DC400-6 in series, so 415ah at 12v (AGM). The compressor always starts on inverter power, just with a slight stutter, so I haven't really worried about the issue so far. It's never failed to start, just doesn't start quite as nicely as it does on shore or generator power, and the voltage flicker at start is mildly annoying.


My thought on the capacitor on the inverter input is that it seems like the battery voltage dip is what causes the whole issue when that large shock load comes on. So anything that helps carry the DC side of the system through that surge would likely help.

0 Likes 0 ·
durval avatar image durval rslifkin commented ·

Yep, if the manufacturer is available and responsive, it's always better to talk to it first.

Re: the kill-a-watt, it's important to know the PFC to properly calculate the capacitor (see below).

Thanks for the info on your batteries and load, very illuminating. Me, I'm going for 3 x 100Ah 12V LiFePO4 batteries in parallel, I expect to have no such issues given the much larger discharge rate they support.

Re: the capacitor, I used to think like you but power-factor-wise I learned that's not how it works. What happens when you have an inductive load is that it turns the current senoid out-of-phase with the voltage senoid so it wastes energy (as high current with low voltage and vice versa means little useful work). You have to calculate the capacitor rather precisely in order to counteract the effect of the indutance, too much or too little and you'd just remain out-of-phase (or get even more out-of-phase than you were to begin with). NB: IANAE, so I might well be wrong on that, but I tried and corroborated it rather carefully.

0 Likes 0 ·
rslifkin avatar image rslifkin durval commented ·

For a capacitor intended to correct power factor, I agree. But on the DC side it would just be to reduce the voltage dip during a very brief high load. The inverter's capability and the power factor of the compressor don't seem to be an issue in my application, as it runs nicely once it starts and the inverter has never thrown an overload warning during compressor start. It's just an issue of voltage sag on the DC side from what I can tell. Basically, the DC voltage sags rapidly and the inverter can't compensate fast enough, so the AC voltage dips as well. Throwing a 1000 watt load at the same inverter and battery bank pulls the battery voltage down a bit, but it doesn't change as suddenly, so the inverter and AC voltage remain happy and the load is handled just fine. Even if that 1000w load is a vacuum cleaner or other motor load (without the huge startup spike of the compressor).


In your setup with LiFePO4 batteries I expect voltage sag will be much less of a concern as long as you don't exceed the momentary discharge rating of the batteries on startup.

0 Likes 0 ·
durval avatar image durval rslifkin commented ·

Hi @rslifkin, thanks for your continued responses.


You are right, I missed the part where you mentioned adding the capacitor to the *DC* side of the inverter. As it's DC and not AC, there would be no current nor voltage phases, then of course the concept of power factor doesn't apply. But then, I'm not sure you would be able to procure a capacitor with enough capacitance to act as a reserve for the kind of amperage we would be talking about.


OTOH, I took the liberty of looking up your batteries and found their datasheet here: https://www.fullriverbattery.com/product/dc400-6/


I was impressed with their listed CCA, 1500A ought to be more than enough to get you the current necessary for your icemaker with no voltage sag. I would bet that either they are degraded (hopefully not) or that something else is amiss, ie cables connecting them to the inverter being too long and/or too thin, terminals not well crimped/corroded/dirtied/loose, etc. If I were on your shoes, I would double check all that before trying the capacitor.


In the name of double checking everything (ie "eating my own dogfood"), I just rechecked the specs on the LiFePO4 batteries I'm planning to buy, and saw they support "only" 100A discharge current (each), no doubt a limitation of their embedded BMS. But as I will have 3 of them in parallel, 3*100=300A * 12V = 3600W should be more than enough even for inrush/starting current/power of anything I'm planning to use.

0 Likes 0 ·
rslifkin avatar image rslifkin durval commented ·
My batteries are only a few months old, and the system will handle a few minutes of 100+ DC amps without issue or enough voltage sag to be any concern. It seems like the issue is mostly around there being enough sag that happens too quickly for everything to compensate when the compressor kicks on. But being that the compressor does start reliably, I'm not all that concerned. I'll see what the icemaker manufacturer says about a soft start, as that's really the only equipment that has any issue.
0 Likes 0 ·
Mike Dorsett avatar image Mike Dorsett commented ·
This is one major problem with running large appliances on a 12V inverter - thre just isn't enough headroom to cushion the inverter against the sudden large current demand. A Capacitor on the inverter dc would be huge - not worth trying.

This problem disappears with 24 and 48V inverters - to a great extent. input current is 1/2 or 1/4, and the voltage is high enough that <1V drop does not affect the inverter performance. I'm running a 2.1kW electric kettle on my 24V solar system, and the lights hardly dim at all when the battery is hit with a 95A load.

Mike.


0 Likes 0 ·