RV Air Conditioner tripping inverter after start

[Checkthisout]

It’s strange though, if I run an induction cooktop that’s double the power, the voltage doesn’t float that much. Only with the air conditioner

It's current demand isn't constantly changing.

If the voltage is too low or drooping too much under sudden demand then the target speed isn't met causing it to go full output. Once the voltage picks back up its output is now too high and dramatically slows its speed. Seesawing back and forth until it settles.

Maybe.

 

[Sirtate]

It's current demand isn't constantly changing.

If the voltage is too low or drooping too much under sudden demand then the target speed isn't met causing it to go full output. Once the voltage picks back up its output is now too high and dramatically slows its speed. Seesawing back and forth until it settles.

Maybe.

I agree...

The two voltage regulators (A/C compressor 'throttle' and the Inverter) are doing a 'Push-me/Pull-you' dance, neither is fast enough (or slow enough) to allow the other to settle in...

I've seen an inverter generator set to 'Eco' mode chase a variable rate compressor fridge that way... (Fridge rattling the beer bottles, and Genny gassing up and down)... turning off Eco mode fixed it...

Maybe there's a way to keep one side of the equation from trying to be more efficient? (run A/C the old fashioned way; all-on or off)

 

[Stever73]

I agree...

The two voltage regulators (A/C compressor 'throttle' and the Inverter) are doing a 'Push-me/Pull-you' dance, neither is fast enough (or slow enough) to allow the other to settle in...

I've seen an inverter generator set to 'Eco' mode chase a variable rate compressor fridge that way... (Fridge rattling the beer bottles, and Genny gassing up and down)... turning off Eco mode fixed it...

Maybe there's a way to keep one side of the equation from trying to be more efficient? (run A/C the old fashioned way; all-on or off

Checkthisout and Sirtate, thank you for helping me to sort this out.
I agree with you, that is probably what is happening, since the A/C has a soft start function it might be an inverter type. I’m not sure if it’s possible to run it in on/off mode, don’t think so.
Apart from changing the inverter or the A/C, is there any component that could be added to their connection, to help them settle?

 

[Checkthisout]

Checkthisout and Sirtate, thank you for helping me to sort this out.
I agree with you, that is probably what is happening, since the A/C has a soft start function it might be an inverter type. I’m not sure if it’s possible to run it in on/off mode, don’t think so.
Apart from changing the inverter or the A/C, is there any component that could be added to their connection, to help them settle?

Sorry, I don't know.

Not a "stab" at you but better quality gens and inverters that do a better job at mimicking grid power are what's needed.

 

[RCinFLA]

You could give more info on particulars of air conditioner. When you stated, "problem with an inverter powered air conditioning unit", does that mean it is a variable speed compressor inverter-air conditioner, like a mini-split, or mean you are powering a general air conditioner from a battery powered inverter?

Two possible issues that come to mind is related to air conditioner AC load power factor or continuously varying air conditioner load due to it being a variable speed three-phase compressor air conditioner being subjected to varying AC input voltage.

If you are in Europe, and the unit is a 230vac input mini-split with variable speed compressor, I would expect the air conditioner has a power factor correcting power supply due to European regulations on power factor.

If you have a meter, like a Kill-A-Watt meter, you might want to check its power factor and fluxuating power draw rate of change.

If mini-split is variable speed, its controller may adjust to compensate for AC input voltage fluxuations. In this case you may get an interaction between varying AC load and battery powered inverter's reaction to compensate AC output voltage regulation for the variable load.

When the battery powered inverter is running near its maximum output load it may not be well behaved when subjected to rapidly varying loads.

This feedback control on inverter's output AC voltage regulation may get unstable, having an interaction between inverter-air conditioner variable AC current loading to compensate for fluxuating AC input voltage and battery powered inverter's ability to compensate to keep its AC output voltage stable with rapidly varying output AC load.

The two units feed on each other's compensation feedback, causing instability on AC voltage regulation. Root cause of issue is battery powered inverter's inability to maintain AC output voltage regulation when subjected to rapidly varying AC output load current. Cure the AC voltage instability and it will stop the mini-split from creating a rapidly varying AC load.

If inverter has enough power capability overhead left, you might be able to add additional constant load to get it to stabilize its output AC voltage regulation again. I would not count on this delicate balance to be maintained as mini-split changes it loading due to normal operation.

 

[Stever73]

…

Hi,
I live in Europe and I’m powering an A/C with an inverter. I’m not sure if the A/C is inverter technology. When it’s running I don’t hear it varying its speed (I have an inverter A/C at home that do that).
It has a “soft start” function that is supposed to suppress surges at startup but I don’t know if that matters.
The specs of the inverter should be more than adequate for running the A/C, and the battery too (8s 300Ah 300A 28v LiFePO4)

Here’s the specs for the Epever inverter and specs and a scheme found on the instructions of the A/C unit, I hope this helps.


I’m looking for a way to correct the problem, any advice is welcome.
Thank you very much

 

[RCinFLA]

It has a “soft start” function that is supposed to suppress surges at startup but I don’t know if that matters.

Soft starter would not be used for a variable speed mini-split.

Soft starter will help initial startup surge on a single phase induction motor compressor.

Even with a soft starter, the battery powered inverter may not be able to maintain constant AC output voltage stable regulation when it is loaded near its maximum output loading capability, when load has an inductive power factor.

If it is a single phase induction motor, power factor may come into play. If it is a single phase induction motor you can improve its power factor with proper sized parallel capacitor to compensate for compressor's inductive load. You will need to know power factor and run current to compute proper sized compensation capacitor.

A soft starter can get damaged with a parallel power factor compensation capacitor on its output side feeding compressor. Compensation capacitor would have to be placed after the power contactor relay and before the soft starter. You should also put a 15k ohm 1 watt power resistor across capacitor to bleed its charge when off.

Without knowing actual power factor, a guess for a 12k btu single phase induction motor compressor would be about a 25 uF 400vac foil run capacitor.

You can just try placing a 25 uF 400vac run capacitor on inverter output while air conditioner is running to see if it cures the AC voltage instability before going to the trouble of actually wiring it into the air conditioner. Expect a small spark when you connect capacitor and make sure you discharge capacitor after removing it.

 

[Stever73]

Thank you sir. Your help is very much appreciated indeed!

Compensation capacitor would have to be placed after the power contactor relay and before the soft starter.

Looking at the scheme, is this the point where I should add the capacitor? (Marked in red)

Without knowing actual power factor, a guess for a 12k btu single phase induction motor compressor would be about a 25 uF 400vac foil run capacitor.

The A/C has a rated cooling capacity of 2000W. It should be 6820btu. Can you please advise me about the correct size capacitor?

I‘ve tried to calculate the power factor of the compressor but couldn’t find a way to determine it. The only device comparable to a wattmeter that I own is an Wi-Fi socket that has its app and shows watts, voltage and current but doing W / (V x A) gives me always 1 or 0,99 so I think it’s not correct.

If you can give me an advice on how to measure the power factor I would be grateful!
I have a multimeter and a small oscilloscope..

 

[Vigo]

Because PF is so time-sensitive, you could probably view it on a scope if monitoring volts and amps simultaneuously, but i dont know if it would be easy to turn the visual into a PF number. It may be enough to establish whether or not it’s a problem regardless of the number.

Other than that, cheapest thing is to get a power meter which measures it. Ive bought one for $15 and one for $40. Whether they are accurate.. unknown.

 

[12VoltInstalls]

cheapest thing is to get a power meter which measures it. Ive bought one for $15 and one for $40. Whether they are accurate.. unknown

Probably accurate enough. I have two because I wanted to know. One monitors a gerrilla micro inverter’s out put, the other records any supplemental 120VAC that comes back in. It records roughly the expectations on a very cloudy day.

It’s interesting to see how the powerco ‘pads’ transmission. Like this morning: I was surprised grid was running 126.xV rather than 120V but it was 38*F. Other times of the day are usually 120-ish. Evenings are typically 124.xV and that is the demand period.

 

[Vigo]

What i was referring to was the accuracy of the power factor reading. I have no way to verify it. I havent compared my ac power meters to multimeter readings but i have done so on most of the solar charge controllers ive ever had and THOSE have frequently been off. Since those are controlling batteries i consider them a lot more important to verify. The ac power meters cant do anything worse than mislead you if they are ianccurate.

 

[12VoltInstalls]

What i was referring to was the accuracy of the power factor reading

No, I got that

Probably accurate enough

power factor is a simple thing but sometimes it’s difficult to understand when the numbers skew so far.
The $15 meters are handy for Wh because when the numbers are generally consistent day to day and then go somewhere unexpected it can alert you to problems- plus you can see results over time.

 

[RCinFLA]

You do not want to overcorrect power factor and get too close to a power factor of 1. That is resonance and can increase voltage ringing when unit shuts off potentially damaging capacitors.

Problem is power factor on compressor changes with mechanical load. A regular air conditioner typically ranges from about 50% to 95% load on compressor over its operating conditions depending on temp/humidity conditions.

Attached at bottom of reply is an induction motor power factor graph of my 4 ton central air conditioner compressor. At upper right corner of graph is the capacitor value to compensate for power factor. I am using a 50 uF shunt capacitor which prevents over correction at lighter load on compressor. It actually drops the measured compressor current down by about 16%. This is direct benefit to battery powered AC inverter loading.

I use a MicroAir EasyStart 368 softstarter and the PF compensation capacitor is between air conditioner contactor and soft starter. I also placed compensation capacitor on a 20A 24vac coil relay out of caution to not discharge cap through soft starter when unit turns off. Also have a 27K five-watt bleed resistor across compensation capacitor to discharge compensation cap when off and a power NTC surge suppressing low value thermistor resistor in series with compensation capacitor to avoid excessive capacitor surge current on relay contacts when AC power is engaged.

The capacitor value depends on size of air conditioner and its power factor

Power factor = Real power / apparent power = real watts / (voltage * current). There are two forms of poor power factor. An induction motor has a phase shift between its voltage and current. A power supply or other electronic device can use a simple rectifier-filter capacitor that causes current to only happen at short high current peaks at sinewave voltage peaks. Some mini-split A/C, especially smaller BTU 120vac units, have poor power factor due to a simple rectifier-filter power supply for their three-phase inverter circuit that supplies power for variable speed compressor.





A HF inverter does not handle poor power factor loads as well as a LF inverter. The period of reverse current for inductive loads pushes current back into a HF inverter high voltage DC filter capacitor and drives the feedback output voltage regulation control on battery to HV DC converter into potential instability. A true bi-directional LF inverter will push the reverse current period back to battery.

 

[RCinFLA]

It’s strange though, if I run an induction cooktop that’s double the power, the voltage doesn’t float that much. Only with the air conditioner

An induction cook top is not an inductive load on AC source.

Induction cook top is a high frequency inverter feeding effectively a transformer primary side driving a single short circuit secondary turn which is bottom of pot.

The high frequency inverter runs on high voltage DC current.

More than likely it has a simple rectifier-filter capacitor from AC input for the HV DC to run its inverter, although European regulations may make units sold in Europe have power factor correcting power supply.

 

[GooseF16]

Induction cooktops are misleading.. they act like resistive loads.. so they work well in this setting.. they have a rectifier that drives DC conversion into coils at a higher frequency, in a resonance type circuit.. much like a voltage transformer, so it appears resistive and ramps up that way.. the compressor in the AC is of course a different animal as voltage fluctuates on load.. they need a more stable punch/voltage.. Thinking when on mains, you get direct pass thru, but when on inverter, it lacks the punch to hold 230v with its limits being open as said above the great explanations..