Inverter Size and Possible Transfer Switch

[Franziska]

Btw. Simpler than switches are sometimes plugs.

We have our 220V system setup in such a way that it can be fed either by plugging shore power or genset or inverter into it, but never more than one of these sources.

[donradcliffe]

Let me muddy the water a bit. Buying an inverter twice the size that you need will mean that you will have waste twice the power--the power loss on standby and and the inefficiencies at low power are significant.

I ran a microwave for years on a modified sine wave inverter--it worked, but I finally sold it because I didn't have a genset and I only used it while I was motoring. All those amp hours have to be put back, and for heating things 5 gallons of propane lasts a long time.

In the RV I put in a 600 watt inverter that runs the CPAP, TV, laptops, etc, with much lower losses than a 2000 watt unit, and it runs 24/7 when we are off the grid. The RV had an automatic transfer switch between the genset and shore power, and I added another DIY one to energize one string of 120v outlets from the inverter.

[Dockhead]

Originally Posted by donradcliffe

Let me muddy the water a bit. Buying an inverter twice the size that you need will mean that you will have waste twice the power--the power loss on standby and and the inefficiencies at low power are significant.

I ran a microwave for years on a modified sine wave inverter--it worked, but I finally sold it because I didn't have a genset and I only used it while I was motoring. All those amp hours have to be put back, and for heating things 5 gallons of propane lasts a long time.

In the RV I put in a 600 watt inverter that runs the CPAP, TV, laptops, etc, with much lower losses than a 2000 watt unit, and it runs 24/7 when we are off the grid. The RV had an automatic transfer switch between the genset and shore power, and I added another DIY one to energize one string of 120v outlets from the inverter.

I agree that way oversizing the inverter will waste some power.



But I say again -- the nominal number the Victron inverters are sold under are NOT their continuous rated capacity. The Victron 2000 is only rated for 1000 watts at 65C, and I don't think that's ambient. Even at 40C it's only 1450 watts, only slightly more than the OP will require. The Victron 2000 is definitely not oversized for this use case.

[mitiempo]

Originally Posted by Dockhead

.........the 1600 for 1200 watts...

Victron's 1600 watt inverter is a European model - 230 volts output. They do not have a US model with this wattage output.

The 1200 is the only Victron inverter under 2000 watts available with 120 volt output for North America.

The op is in Texas.

[Dockhead]

Originally Posted by s/v Jedi

The real power vs apparent power differs by the power factor and this will differ for every load. In these tables they have taken a power factor of 0.8 (3000 x 0.8 = 2400) but if you would power a pure resistive load then the power factor could be 1.0

The real power rating indeed has more to do with heat dissipation capacity than with how many amps. I know I wrote continuously before but that isn’t true because we don’t use that much power. But it’s been 15 minutes or even a bit more. Also, induction cooktops, ovens and microwaves cycle power so that’s not continuous either.

All this is certainly true, and I think probably underlines the point that you should have decent overhead in the inverter and not plan on running it flat out at its theoretical maximum power.


My own experience is just one data point, but perhaps a useful real life data point for the OP. With my unit, to not experience Temp faults, I have to keep the loads, even pure resistive loads, down to somewhat less than the rated capacity at ambient temperature, in watts with power factor, which is not all that much more than half the nominal power in VA.


The amount of power wasted for a somewhat larger unit is pretty small; the amount of aggravation saved is large. Also, as an electronics engineer, I'm sure you'll agree that running this kind of equipment flat out so that Temp and Overload circuits get activated all the time is not conducive to long life.

[Dockhead]

Originally Posted by mitiempo

Victron's 1600 watt inverter is a European model - 230 volts output. They do not have a US model with this wattage output.

The 1200 is the only Victron inverter under 2000 watts available with 120 volt output for North America.

The op is in Texas.

And that's definitely too small.

[Dockhead]

By the way, concerning how much power is wasted by an inverter too large for the load -- has anyone actually seen efficiency curves? We have been assuming maximum efficiency is at full load, and less and less as the unit is partially loaded.


I didn't know this myself, but this turns out to be false. In fact the opposite is true -- maximum efficiency of the newer Victron inverters occurs at low loads of around 10%:






https://www.victronenergy.com/upload...efficiency.pdf


And this guy:


https://www.photovoltaikforum.com/th...richtern-wirk/


based on actual bench measurements found the same thing -- the Multiplus 3000 is MORE efficient than the Phoenix 375 already from 171 watts, which is only 6% of the 3000's capacity.



So thanks to this thread, I learned something myself. From the point of view of efficiency, the OP would be much better off with actually the 3000, and even better off with the 5000. The bigger the better.


Unfortunately the bigger inverters are more costly and more bulky. But certainly he shouldn't go smaller than the 2000.

[s/v Jedi]

In the past I have had temp faults which were caused by low DC voltage at the inverter, resulting in higher currents to compensate to get the same power, resulting in higher than normal temperatures.

Shorter DC cables and battery banks with higher voltage are the cure for that.

[noelex 77]

Modern inverters are quite different to older models. When striving for maximium efficiency it was important with older models to keep the inverter size as small as possible for the load concerned. This is no longer true.

Unfortunately, these days it is more complicated and each individual model needs to be looked at separately.

There are two factors: the standby current and the voltage conversion efficiency.

The standby current varies substantially between models. While this standby current tends to rise with larger models, this is not always true. Sometimes larger, more sophisticated inverters can have a lower standby current than smaller models. This is primarily because they incorporate more advanced systems that will power down circuitry when no load is detected. You will see from the table Dockhead has posted that there is almost no difference between the standby current for Victron inverters between the 800w and 2000w models. From here there is a significant jump to the 3000w model. This unit would be best avoided if the 2000w model can supply enough power. However, these figures do not always reflect reality. Sophisticated inverters often have adjustable settings for these power down features. The manufacturers often quote the standby current at the most aggressive power down setting. In practice these aggressive settings do not work well, with the inverter inadvertantly powering down with low loads such as phone chargers. The other issue is if you regularly leave these type of devices plugged in they can defeat the power settings.

The next issue is voltage conversion efficiency. This varies substantially between models and across the power spectrum. Manufacturers generally only quote the maximium conversion efficiency, which only gives consumers one data point to compare. Larger, more expensive inverters often employ more sophisticated transformers, power supplies and ciruitry so effiency is sometimes worse for smaller inverters than larger inverters even when running moderate loads.

Ideally manufacturers would list the standby current for each of the possible power saving settings and a graph showing the efficiency across the power spectrum. Unfortunately, this detail is rarely published or at least it is very difficult to find. Using the few select details the manufacturer publishes such as lowest standby current and maximium power effeciency point is perhaps better than nothing, but using these single data points is not ideal and can be misleading.

These details are generally unimportant to boats using mainly generator power, but on a boat relying on alternative energy (solar, wind etc) these details become significant. One option to consider is having an easy and convenient system to turn the inverter off when not in use. Eliminating the standby power drain can make a difference. This option would not suit many boats, as they leave AC chargers etc plugged in and rely on these to work, but boats that are concerned about power consumption have probably eliminated or at least drastically reduced these devices in favour of a DC alternative.

[s/v Jedi]

We avoid zero load power losses by simply keeping the system switched off when we don’t use it, instead of relying on “search, detect” features etc. I am not a fan of these features.

We also have added losses from transformers so for longer duration but smaller loads, like a fan at night, we use small, per cabin, inverters or DC powered options.

[Pelagic]

[QUOTE=Dockhead;3495435]

The amount of power wasted for a somewhat larger unit is pretty small; the amount of aggravation saved is large.QUOTE]

I would liken this to a watermaker.
Never heard anyone say that they installed one too big !

Sooner or later when SOC is full and charging available via solar or alternator, you will find extra uses like a small aircon for a siesta.

As Jedi and Nolex said, turn it off when not needed and on usually during meal prep, when phones are charged

[Dockhead]

Originally Posted by s/v Jedi

In the past I have had temp faults which were caused by low DC voltage at the inverter, resulting in higher currents to compensate to get the same power, resulting in higher than normal temperatures.

Shorter DC cables and battery banks with higher voltage are the cure for that.

Good advice, but this is not my particular problem. I don't run large continuous loads off the batteries in any case. The problems are observed when running loads off the school bus alternator, and here the cable is fat and less than 2 meters long, and system voltage typically more than 27 volts even under load.


I think the problem is that the load vs temperature rating is not for ambient temperature, but heat sink temperature (something about that in the documents). Run a couple thousand watts for a longer period of time and the temperature gets up higher than that at which the unit delivers 2000 watts. I think that's all there is to it, at least in my case.


This doesn't cause me any problems; after years with this system I'm used to it and know how to manage the loads. But if I ever redo the system (or have a new boat), I'll have a large lithium battery bank and much more inverter capacity. A pair of 3000's or maybe a 5000, at the very least.


Especially now that I know (as of today) that these units are MORE efficient at partial load, than they are when you load them up.

[s/v Jedi]

Originally Posted by Dockhead

Good advice, but this is not my particular problem. I don't run large continuous loads off the batteries in any case. The problems are observed when running loads off the school bus alternator, and here the cable is fat and less than 2 meters long, and system voltage typically more than 27 volts even under load.

I think the problem is that the load vs temperature rating is not for ambient temperature, but heat sink temperature (something about that in the documents). Run a couple thousand watts for a longer period of time and the temperature gets up higher than that at which the unit delivers 2000 watts. I think that's all there is to it, at least in my case.

This doesn't cause me any problems; after years with this system I'm used to it and know how to manage the loads. But if I ever redo the system (or have a new boat), I'll have a large lithium battery bank and much more inverter capacity. A pair of 3000's or maybe a 5000, at the very least.

Especially now that I know (as of today) that these units are MORE efficient at partial load, than they are when you load them up.

With 27V measured at the inverter, voltage is good so not the problem. So in that case a temperature overload means the avg inverter load is higher than the heat dissipation capacity of the inverter. Two main things to check: does the fan run and displace enough air (clean fan?) and what is the temperature of the heatsink, i.e. is the internal temperature sensor operating correctly.

The temperature in the specifications is the ambient temperature.

[Dockhead]

Originally Posted by s/v Jedi

With 27V measured at the inverter, voltage is good so not the problem. So in that case a temperature overload means the avg inverter load is higher than the heat dissipation capacity of the inverter. Two main things to check: does the fan run and displace enough air (clean fan?) and what is the temperature of the heatsink, i.e. is the internal temperature sensor operating correctly.

The temperature in the specifications is the ambient temperature.

Good questions; thanks.


I can hear the fan running but I'll check to be sure everything is clean.


I'll see if I can check the heat sink temp -- don't know if it's accessible.


Nevertheless, I'm not sure it's a fault. It's not far from the rated max load for given temperature where I start to have problems.

[rslifkin]

When I picked out an inverter, I went for the Multiplus 2000 over the 3000 because of 1 big reason: the idle power use on the 2000 is half of what the bigger inverter needs, meaning I suffer half as much for leaving it on 24/7. And I figured I didn't need the extra capacity anyway.

Biggest load I've ever run on the thing is probably ~1100 watts (vaccuming while the ice maker was running), but any load that big is only ever for a few minutes at a time, so heat is unlikely to be limiting in most weather (my inverter is in the forward bilge area which stays fairly cool). Heat hasn't been an issue when charging in bulk at the full rated 80 amps either, and that's occurred for longer periods of time (and the unit shows 1100+ watts input, so there's plenty of power being converted while charging).