How much power does an Inverter use just sitting there idling?

Joe Samra · Apr 25, 2022

Hedges said:
Tim is using an AC/DC clamp ammeter (hall effect), can read DC current.

Joe has an AC only clamp ammeter (current transformer)

Possibly, "No batteries" has something to do with undesired behavior. Can you connect a batch of car batteries?
Then observe any difference in AC current, and read battery current with the AC scale you have (reads AC ripple but not DC.)

Tim - besides AC current in AC grid connection as requested, check AC current in battery cable (with and without grid connected).
I would expect that to be low under no load, but who knows. Likely has ripple when charging.
I find that when driving a large AC load from batteries, although battery DC (average) current matches load wattage, there is a large AC ripple. Capacitors can't buffer 60 Hz, only the higher frequencies of switcher. At least with my low-frequency inverter.

Joe - does your friend also have batteryless setup?
Without AC input, just PV, at least average current has to come from PV panels. Possibly peak current at top of sine wave. But I would guess a high frequency inverter would draw down capacitors to deliver peak current, at least in the batteryless case.

But I still don't know where 2A would be going. That's too much current draw on AC input when there isn't that much current on AC output.

This one?

Customized Solar Power System,Growatt Hybrid System,Split Phase PV System - Swtsolarpv.com

China supplier of solar system - off grid solar system,lithium battery storage system,solar power station offered by Sweet Power. Any questions,pls do not hesitate to contact us!

www.swtsolarpv.com

"PV & Grid Power combine if PV energy is insufficient for load."
"

AC CHARGER
AC Charge Current 80A
AC Input Voltage 230VAC
Frequency Range 50/60HZ

"

A simple UPS function can be implemented with a relay - either pass grid through to load, or disconnect from grid and use inverter.
Much more difficult is to combine PV and Grid on AC side, requires synchronized AC waveform and protection against backfeed to grid. Grid-tie hybrid inverters to this.
To accomplish the function more simply, a separate battery charger can be used, an "on-line UPS" or "double-conversion". AC is converted to DC, then DC to AC.

That still shouldn't explain 460W consumption, but possibly the battery charger is using significant inductance of a transformer to block AC current. 80A at 50V = 4000W. 2A at 230V = 460W, 11% of full-load current.

No-load current of a transformer is a thing. Some references say should be under 5%.
Probably more cheaply built and undersized will allow more.

https://cr4.globalspec.com/thread/106496/High-No-Load-Current-on-Transformers

I'm presently delving into transformer core characteristics, and there is a LOT of science and engineering detail in there. After metal is worked into the shape desired, it has to be annealed to restore desirable characteristics. This can be done without magnetic field, or with field in longitudinal or transverse direction. Those produce considerable differences in behavior, good for various applications. Power transformers are just one of many.

Likely, most of that 2A and 460VA is reactive power, not real power. But if your utility meter doesn't distinguish, that doesn't do you any good. (It also means you wouldn't be able to spin the meter backwards like we do with grid interactive inverters for net metering.)

I ran a test of a transformer type battery charger.
On 10A (12V) setting, it draws 0.00A at 120Vrms.
On 55A (12V) "start" setting, it hums moderately loudly and draws 1.18A at 120 Vrms.
That's zero W delivered on 660W setting, 142VA reactive power drawn, 21% of full-load current.

I suspect the GroWatt has a cheap undersized transformer for the battery charger (not a high frequency switcher, and not a larger heavier transformer.) At 50 Hz vs. 60 Hz the transformer goes further into saturation, also at 240V vs. 220V.

View attachment 92402 View attachment 92403

If this is what's going on, maybe best to resell it to someone who would only use it for solar, or downstream of their generator. They can run the generator briefly for larger loads, then turn off and run smaller loads from battery.

Alternatively, disconnect from grid, and get a different battery charger without such no-load draw. But it must properly regulate for the batteries you're using, either regulated voltage or have a "maintain" function that is OK to leave plugged in continuously.

Hello Hedges, my friend has 4 Amaron batteries installed in his system, he conducted the test i required with batteries breaker turned off and AC output breaker too.
I installed in my panel a modular AC contactor to pass through AC grid to my house without passing by the inverter, i guess it is a normal behavior for growatt to withdraw 2 amps from grid without batteries since @timselectric , me and my friend had the same results, im going to wait for my batteries to arrive, until then i only use the inverter for small loads when needed (not the refrigerator because upon starting its surge is way strong for 4 x 655W panels to handle so the inverter trips and indicates PV power insufficient).
Thank you so much for your help, i really appreciate it.

Joe Samra · Apr 25, 2022

timselectric said:
Yes, I'm getting the same as you. I cycled through all modes. With solar, battery, and load, disconnected.
Anywhere from 2 to 2.6 amps , on grid input at idle.
It appears that the inverter is using power to monitor all inputs, whether or not they are connected to anything.
At least, this is my best guess.

@Hedges
When charging or discharging, I could read Anywhere from 3 to 4 AC amps on battery cable. No AC volts, of course. And, the DC current level doesn't affect the AC current level. So, it's just a flat ripple from the inverter.

Thank you man, i guess the best option is to have batteries installed, PV and grid power without batteries did cost me a big electric bill, thanks again.

Hedges · Apr 25, 2022

Joe Samra said:
Thank you man, i guess the best option is to have batteries installed, PV and grid power without batteries did cost me a big electric bill, thanks again.

I think so. Still not optimum for nighttime use, because it will cycle batteries and then connect to grid if low. You might do better adding an automatic transfer switch downstream from GroWatt so if grid is up at night, loads draw from grid. During the day, loads powered by PV supplemented by AC through GroWatt's battery charger.

Selling it and getting one that operates more efficiently might be better.

For PV without batteries, best arrangement is net metering, backfeeding all power harvested into your house wiring, and into the grid if your loads are less. That only works if utility allows this and credits you for spinning meter backwards. I think your present meter spins meter forwards regardless of whether you d draw power or deliver power (based on it spinning forward for "reactive" power which is 50% draw, 50% deliver.)

Next best is zero export. A current transformer around utility feed measures current (and phase, draw vs. deliver, is determined by also monitoring voltage.) The zero-export inverter ramps production up and down to deliver up to 100% of available solar or total consumed by house, whichever is less, without exporting to grid.

Sounds like your GroWatt can be used with batteries to keep power on when grid goes down. And, to recharge from generator if you run that, allowing you to turn off generator sooner and keep running from inverter. Needs to be configured so grid input is disconnected from inverter (hopefully internally) so the 2A draw stops when battery is full. And remains disconnected until battery gets low, so you don't have that draw all night. Batteries should supply starting surge for motors.

For such use, you need batteries able to cycle many times. Lithium has claimed 2000 to 6000 cycles. Various lead-acid a few hundred. In both cases, I think cost of battery divided by all lifetime use exceeds cost of just buying power from utility grid. Exception being some low-cost recycled or DIY lithium batteries.

My inverter, SMA Sunny Island, doesn't work as an on-line UPS like yours apparently does, separate battery charger and always running inverter. It has a relay inside to connect loads to the grid. It consumes 25W (compare to 460W) for one 6kW inverter. My lead-acid batteries are kept at float voltage day and night when grid is up. If grid goes down, after a brief glitch it disconnects from grid and powers loads from battery. PV can be DC coupled to battery with separate SCC, but mine has PV AC coupled with grid-tie PV inverters. If grid is up they backfeed (I have net metering). If grid goes down, they feed my loads and the battery inverter which charges battery or draws from battery depending on AC power draw.

There should be some low-cost grid-tied inverters with zero export. They use a high frequency switcher to convert AC to battery or battery to AC without that large (transformer?) current draw of the GroWatt. Optimal configuration provides battery/PV backup to critical loads, and delivers up to 100% of PV backfed to other loads in the house.

GLC · Apr 25, 2022

I have 1 -4000 watt 110 inverter that uses 6 watts at idle and I have 1- 6000 watt 110/220 inverter that uses 45 watts at idle.

skylink · Jun 10, 2022

Joe Samra said:
Thank you man, i guess the best option is to have batteries installed, PV and grid power without batteries did cost me a big electric bill, thanks again.

Hi, You may see 2amps or so but that is apparent power, https://www.allaboutcircuits.com/te...ent/chpt-11/true-reactive-and-apparent-power/ However the true power used is around 60w really very little,

fblevins1 · Aug 18, 2022

I think at the end of the day, the best way to limit how much your inverter cost to just sit there with the lights on is to be realistic about your average load on any given day. If you know you average about 500 watts of load, then a 1 KW inverter is just fine and will consume less than a super whammadyne 5 KW inverter with bells and whistles.

wade0000 said:
As information, I just did a test of the inverter decent load versus almost no load:

Inverter:
250W load (confirmed thru measurement) = net of 29W for Inverter itself
2W load just charging a wireless speaker = 21W " "

Thanks for all the help. I want to put in an automatic switch DC to AC grid when the batteries get too low... so I'll have to add some kind of alert on that trigger so I can put in a DC switch to turn off the inverter.

Thanks too everyone who responded. Best forum. Saved a huge amount of time and grief with your expertise previous on cheap Chinese DC breaker issue.

The ATS that I use, quietly does it's job without any bells or whistles. When the batteries drop below 12.3 volts it switches over to grid, but empirically the inverter doesn't seem to kill off the battery at night. I think the reason for that is under load (say....200 watts) when the battery hits 12.3 it switches over to grid but then the load is reduced on the battery and it's voltage creeps back up to around 12.5 volts even with inverter on, giving you enough juice to power to inverter for the rest of the night without fear of your DOD going below 12.3 volts (or whatever the lower limit is for the battery and voltage system you use. Get the ATS first, wire it in and then just watch what your battery does and you may find you don't need to worry about inverter killing off your battery once the grid kicks in and your load is reduced, but then again...if you are good at wiring in switches then hell...go for it, sounds like fun.

Rconce01 · May 16, 2024

Hedges said:
It may be the inverter is drawing power to charge battery when AC is available.
Check battery current if you can. Otherwise, check battery voltage without/without AC from grid turned on.
If it is charging from grid, you may be able to program different behavior or battery voltage.

It is also possible the inverter presents a reactive load, is drawing current and VA but now power Watts. Some utility kWh meter might report power consumed even if it was all "reactive" power, zero real power.
You would need a device able to measure Power Factor or VA vs. W to tell. "Kill-a-watt" is one.

If the inverter presents an inductive load carrying 2A (which seems too high to me), adding a suitable capacitor could reduce that. That would be 23 microfarads if I did the math correctly. But it's not something I've ever tried, and first look for parts indicates it is not financially practical.

https://www.grainger.com/product/WEG-Power-Factor-Correction-Capacitor-16Y122

If the system is continuously consuming 460W, or causing an incorrect billing, that is obviously not good.

I was reading this thread and it's very helpful. I'm in the same situation we're I get a 2+ amp load in the ac input circuit with no load on the inverter and no battery.

What is reactive load? If I measure 2amps on the clamp on meter isn't it consumption 2 amps and my meter registers it?

Hedges · May 16, 2024

Reactive load is a load with capacitance or inductance. Current is a function of derivative (change) in voltage (for capacitor) or is the integral of voltage over time (for inductor) rather than the voltage at that instant.

2Arms at 230Vrms is 460VA.
If that 2A is positive during the fraction of second the 230V is positive, it is 460W.
If that 2A is negative during the fraction of a second the 230V is positive, it is -460W (e.g. backfed from grid tie PV inverter).

If that 2A is positive, hits peak of 2A x sqrt(2) at the instant voltage crosses from positive to negative (so half the time current is positive, voltage is positive, half the time current is positive voltage is negative), then it is 0W. An inductive or capacitive reactive load.

During second half of one voltage phase, energy is being put into the inductor which stores it. During first half of next voltage phase, energy is being removed from inductor, returned to the grid.

A transformer is two coupled inductors, with some wire resistance. Under no load, waveform looks much like above (but usually distorted, sized to save money it is far from "ideal".) It is shifted slightly, with power dissipation due to resistance.

When there is a load on secondary, current increases and phase shifts so much of the additional apparent power VA is delivered to the output, and is real power W.

The transformers I've put a meter and oscilloscope on draw about 2% maybe 5% of rated current under no-load. The current waveform looks more like a Hershey's Kiss than a sine wave, shooting up at the peak. That is approaching "saturation" where inductor almost fails to do its job. Tolerable on grid, not what I want driven by inverter.

But You're in Florida, I see Red/Black/White.
Is yours 230V, or 115V?
If 230V would expect zero current in white.
I see black/white/green going into conduit so maybe 115V.
115V x 2A = 230VA vs. 50W spec is a bit more reasonable.

I don't know if your inverter has a transformer based battery charger, or if this is rectifier/capacitor front end (which also has poor PF, higher Irms than expected for a given power draw.) In either case, if just a reactive load on the grid, not a problem. If real power dissipation, then it is costing you in utility bill.

Better inverters (utility interactive, able to feed into grid) would be PF corrected, using their bidirectional switch-mode power supplies to make sine wave current.

HRTKD · May 16, 2024

The white-ish wire that the clamp on meter is attached to looks rather sketchy.

Rconce01 · May 16, 2024

HRTKD said:
The white-ish wire that the clamp on meter is attached to looks rather sketchy.

The whole thing is sketchy because I'm sketchy!!

Rconce01 · May 16, 2024

Hedges said:
Reactive load is a load with capacitance or inductance. Current is a function of derivative (change) in voltage (for capacitor) or is the integral of voltage over time (for inductor) rather than the voltage at that instant.

2Arms at 230Vrms is 460VA.
If that 2A is positive during the fraction of second the 230V is positive, it is 460W.
If that 2A is negative during the fraction of a second the 230V is positive, it is -460W (e.g. backfed from grid tie PV inverter).

If that 2A is positive, hits peak of 2A x sqrt(2) at the instant voltage crosses from positive to negative (so half the time current is positive, voltage is positive, half the time current is positive voltage is negative), then it is 0W. An inductive or capacitive reactive load.

View attachment 215760

During second half of one voltage phase, energy is being put into the inductor which stores it. During first half of next voltage phase, energy is being removed from inductor, returned to the grid.

A transformer is two coupled inductors, with some wire resistance. Under no load, waveform looks much like above (but usually distorted, sized to save money it is far from "ideal".) It is shifted slightly, with power dissipation due to resistance.

When there is a load on secondary, current increases and phase shifts so much of the additional apparent power VA is delivered to the output, and is real power W.

The transformers I've put a meter and oscilloscope on draw about 2% maybe 5% of rated current under no-load. The current waveform looks more like a Hershey's Kiss than a sine wave, shooting up at the peak. That is approaching "saturation" where inductor almost fails to do its job. Tolerable on grid, not what I want driven by inverter.

View attachment 215762

But You're in Florida, I see Red/Black/White.
Is yours 230V, or 115V?
If 230V would expect zero current in white.
I see black/white/green going into conduit so maybe 115V.
115V x 2A = 230VA vs. 50W spec is a bit more reasonable.

I don't know if your inverter has a transformer based battery charger, or if this is rectifier/capacitor front end (which also has poor PF, higher Irms than expected for a given power draw.) In either case, if just a reactive load on the grid, not a problem. If real power dissipation, then it is costing you in utility bill.

Better inverters (utility interactive, able to feed into grid) would be PF corrected, using their bidirectional switch-mode power supplies to make sine wave current.

Wow that's an amazing explanation that's well over my head. I will however look into it more to try to learn the detailed explanation you went through. This is a 220v setup with no neutral. Therefore no split phase. The black wire and white wire are the ones feeding the line into the inverter. I am consuming those 2amps on 240v however I've only measured the white wire I'll measure the black one later this evening. When I feed the input of the inverter 120v instead of 240v the 2 amps drop to about 1amp.

It's pretty cool I can use the inverter for both voltages just not at the same time.

I have one of those smart digital meters from the utility company so I'm going to turn off most of the loads in the house and just leave the inverter connected to the 220v and see what the meter registers. I appreciate your expertise and feedback. I think the one day I left it hooked up it didn't pull those levels of current based on the usage chart I saw from the utility company. I actually got some unusually low consumption until 12 noon. Typically that the amount that only my fridge consumes all night.

I added some screenshots with no load attached the interface of the inverter says 14-17VA and 0% load..

Hedges · May 16, 2024

Capacitance is like pulling on a bungee cord.
Inductance is like swinging a bowling ball.
Resistance is like dragging a rake through through loose sand.

Complex loads including bungee cord pulling the rake, or a weight pulling it.

Current in white wire will be equal and opposite to current in black wire at any instant.
To see if energy is being dissipated, or just stored in the inverter for 4 milliseconds then returned to the grid, you need simultaneous measurement of voltage and current, at a sample rate of 1 kHz or faster. I do that with an oscilloscope and voltage + current probes. Kill-A-Watt meter should do that with its ADC and displays "Power Factor" as well as watts.

With your utility meter feeding just the inverter, no load direct or through the inverter, after enough hours you should see the difference between "Real" power measured by utility meter and "Apparent" power indicated by your ammeter x voltage.

If I'm going to feed a typical transformer from an inverter, I like to apply 120Vrms to 240V windings. It is much closer to ideal that way. Fed by grid, don't worry about it.

Rconce01 · May 16, 2024

Hedges said:
Capacitance is like pulling on a bungee cord.
Inductance is like swinging a bowling ball.
Resistance is like dragging a rake through through loose sand.

Complex loads including bungee cord pulling the rake, or a weight pulling it.

Current in white wire will be equal and opposite to current in black wire at any instant.
To see if energy is being dissipated, or just stored in the inverter for 4 milliseconds then returned to the grid, you need simultaneous measurement of voltage and current, at a sample rate of 1 kHz or faster. I do that with an oscilloscope and voltage + current probes. Kill-A-Watt meter should do that with its ADC and displays "Power Factor" as well as watts.

With your utility meter feeding just the inverter, no load direct or through the inverter, after enough hours you should see the difference between "Real" power measured by utility meter and "Apparent" power indicated by your ammeter x voltage.

If I'm going to feed a typical transformer from an inverter, I like to apply 120Vrms to 240V windings. It is much closer to ideal that way. Fed by grid, don't worry about it.

I understood all the way up to this comment. "If I'm going to feed a typical transformer from an inverter, I like to apply 120Vrms to 240V windings. It is much closer to ideal that way. Fed by grid, don't worry about it."

I'm not understanding how this is relevant? I'm feeding a 240 v mini split off this inverter not a transformer..

The only transformers I know about are the step up and step down and the auto transformers that creat 120v split phase out of single phase 240v . I'm not using any of those though.

Hedges · May 16, 2024

The transformer issues aren't relevant to you in that case.

I think most minisplits in the US have diode/capacitor front end, not "Power Factor Correction" circuits. They will require higher wattage inverter because they draw higher current than their wattage rating would suggest. It draws current looking like clipped top of sine waves.

I think someone said Mitsubishi is one that is PF corrected. In Europe, many/most are.

Mitsubishi mini splits off-grid?

I was wondering if anyone was using 24k or 36k Mitsubishi mini splits off-grid? I was told today by an HVAC guy it wouldn't work, but Daiken or Fujitsu would. I've read multiple other threads here, but didn't see anyone mention that they had Mits and were completely off-grid. Thanks in advance!

diysolarforum.com

Rconce01 · May 16, 2024

Hedges said:
The transformer issues aren't relevant to you in that case.

I think most minisplits in the US have diode/capacitor front end, not "Power Factor Correction" circuits. They will require higher wattage inverter because they draw higher current than their wattage rating would suggest. It draws current looking like clipped top of sine waves.

I think someone said Mitsubishi is one that is PF corrected. In Europe, many/most are.

Mitsubishi mini splits off-grid?

I was wondering if anyone was using 24k or 36k Mitsubishi mini splits off-grid? I was told today by an HVAC guy it wouldn't work, but Daiken or Fujitsu would. I've read multiple other threads here, but didn't see anyone mention that they had Mits and were completely off-grid. Thanks in advance!

diysolarforum.com

Interesting I didn't know that mine is cooper and hunter. It's really efficient most of the time is pulling 2-4 amps on 240v. It's a 24000 but unit.

I'm going to hook the inverter back up and see what the meter registers.

Rconce01 · May 16, 2024

So I got this response and image from the seller. I'm not sure if I should believe it. They show a much more sophisticated device and explained that the power factor is .08 under no load. I'm not sure if that believable.. I did find the manual references power factor at greater than .99.

Hedges · May 16, 2024

His current measurement is in line with yours, and 45W is a reasonable no-load power.
That 0.08 PF could be correct.

Now the question is, under what conditions is power factor > 0.99?

I would have expected higher no-load power factor from a device with such a figure quoted, because it would be a PF-corrected design. Is this LF (transformer) or HF (light weight, transformerless)?

Rconce01 · May 16, 2024

Hedges said:
His current measurement is in line with yours, and 45W is a reasonable no-load power.
That 0.08 PF could be correct.

Now the question is, under what conditions is power factor > 0.99?

I would have expected higher no-load power factor from a device with such a figure quoted, because it would be a PF-corrected design. Is this LF (transformer) or HF (light weight, transformerless)?

I haven't opened it up but it's lightweight definitely not heavy enough to have a 6.5kw transformer.

My main argument to them is why don't they state the power factor under no load?

With these smart meters they have installed in the US now I'm wondering if they are better at charging the reactive power instead of just the actual power.

I hooked up the AC load on the inverter and on the hook on amp meter its 2.4 amps out of the inverter output. And 3.8 amps inverter input. This tells me that the apparent amperage draw of the inverter is down to 1.5 amps under load from the 2.2 amps I was reading under no load.

AC CHARGER
AC Charge Current	80A
AC Input Voltage	230VAC
Frequency Range	50/60HZ

How much power does an Inverter use just sitting there idling?

Medical Electronics Engineer

Medical Electronics Engineer

I See Electromagnetic Fields!

Solar Enthusiast

New Member

New Member

New Member

Attachments

I See Electromagnetic Fields!

Boondocker

New Member

New Member

Attachments

I See Electromagnetic Fields!

New Member

I See Electromagnetic Fields!

New Member

New Member

Attachments

I See Electromagnetic Fields!

New Member

Similar threads