RV Air Conditioner tripping inverter after start

[Stever73]

Hello everyone,

I have a problem with an inverter powered air conditioning unit that happens after the A/C started and I can’t sort it out.

The inverter is an all-in-one charger/solar controller/inverter Epever UP3000-M3322 served by 8x280Ah 3,2V LiFePO4 batteries and I want to power an air conditioning unit Kronings A2000-50 EU that has a rated continuous input power of 690W. The inverter output specs are 2400W continuous, 3000W 15min., 4800W 5sec., 6000W Max surge power.

The unit runs smooth when 230v mains are connected, however when I run the inverter from batteries, the air conditioner starts shaking, making noise, and the output voltage of the inverter starts tripping.

When running the air conditioner with the inverter output set to 220V/50Hz as soon as it starts cooling it also starts shaking continuously.
When running the air conditioner with the inverter output set to 230V/50Hz as soon as it starts cooling it also starts shaking intermittently.

I’ve noticed on the inverter display that the voltage output when the A/C is cooling is not stable, it floats between 210 and 240 V. I’ve attached a video showing the display. No errors are displayed on the inverter or on the A/C unit.

The inverter powers smoothly all the other appliances like refrigerator, induction cooktop, heater, tv. I don’t have any problem with other appliances, and the output voltage stays stable. Tested up to 4000W constant load.

I hope that someone on this forum can give me an advice on how to solve this. Thank you very much

 

[Bluedog225]

Following.

 

[Denver]

this is interesting

 

[HRTKD]

My guess is that it's your battery. It could be the wiring between the battery and the inverter. Ripple effect could be an issue, but I would defer to one of the other forum members on that topic.

What is the BMS rating on your battery? What type of wire and length are you using between the battery and inverter? When running the test, have you checked every connection and wire for heat? Any connection or wire that is hot to the touch, or is hotter than anything else, is a red flag.

 

[Stever73]

Hi Jim, thanks for the answer.
The 8s 24v 150A BMS is in control of the battery but the inverter current actually doesn’t flow thru it, I’ve setup the on/off switch of the inverter to pass by a solid state relay that is powered from the BMS, so if the BMS detects anything’s wrong, the inverter will be safely shut off.
Wires are solderer type 35mm2 and the length from the battery to the inverter is less than a meter. There’s a 500A shunt on the negative and a 250A ceramic fuse on the positive, between the battery and the inverter. I’ve tested the setup for about half an hour at 2000W with hand and infrared thermometer, trying to find any hotspot in wires and junctions but they all stayed cool. I’ve made the junctions with commercial grade terminals, crimped with an appropriate tool and then filled ‘em with solder tin.
The AC main line is 4mm2 from the inverter to the junction box and then 2,5mm2 to the A/C. I’ve also tested the A/C with 2,5mm2 wires directly connected to the inverter with same result.

 

[RCinFLA]

Hi Jim, thanks for the answer.
The 8s 24v 150A BMS is in control of the battery but the inverter current actually doesn’t flow thru it, I’ve setup the on/off switch of the inverter to pass by a solid state relay that is powered from the BMS, so if the BMS detects anything’s wrong, the inverter will be safely shut off.

That is likely a bad idea. You have defeated battery inverter overcharging protection since solid state relay MOSFET body diode will pass charge current with a diode drop even if you turn off SSR by BMS cell overvoltage sensing. Most DC solid state relays are not bi-directional switches.

If inverter/charger shuts down when SSR opens then you are okay but if it stays on due to grid or PV power you could overcharge batteries.

Check the voltage drop across SSR during startup surge of compressor.

 

[Bluedog225]

I can’t think of anything that would make it start shaking but for some ripple or fluctuation. Do you have a clamp meter?

 

[Stever73]

That is likely a bad idea. You have defeated battery inverter overcharging protection since solid state relay MOSFET body diode will pass charge current with a diode drop even if you turn off SSR by BMS cell overvoltage sensing. Most DC solid state relays are not bi-directional switches.

If inverter/charger shuts down when SSR opens then you are okay but if it stays on due to grid or PV power you could overcharge batteries.

Check the voltage drop across SSR during startup surge of compressor.

The SSR turns on and off the inverter/charger thru its on/off switch, it’s not on the power line. And inverter/charger has its own overcharge/over discharge protection, voltage based. I’m not sure if I understood well what you meant.

I will check the SSR voltage drop at startup though, thanks for the hint

 

[Stever73]

I can’t think of anything that would make it start shaking but for some ripple or fluctuation. Do you have a clamp meter?

Yes I do, but it’s not a professional one. How could I check ripple?

 

[Stever73]

---

 

[12VoltInstalls]

Ripple effect could be an issue, but I would defer to one of the other forum members on that topic.

Me too

can’t think of anything that would make it start shaking but for some ripple or fluctuation. Do you have a clamp meter?

know nothing of ripple from any engineering perspective - but -

A refrigerator on modified sine wave would do a shimmy/shake/shudder and a crt computer monitor I fried on a modified sine wave inverter would ‘wow’ light intensity/colors. A dual-wheel bench grinder I saw not start but oscillate back and forth.
Anyway that’s what crossed my mind.

 

[Bluedog225]

Way over my head as well. Sorry. I was thinking if the voltage or amperage was dancing, you’d be closer to an answer.

Hoping one of the smart guys will be along.

 

[Al D]

Hello,

I looked up the specs on your UP3000-M3322. When the output is set to 230V, the allowable fluctuations are -10% to +5%. That corresponds to 207V - 242V. That's pretty much what you described seeing.

As the inverter air conditioner compressor is unique to other compressors in that it is driven by varying power, it's probable that the output of your UP3000 isn't stable enough to keep the compressor running smoothly, hence the shaking.

 

[Checkthisout]

Is your inverter pure or modified sine wave?

 

[Stever73]

It’s pure sine wave

 

[Stever73]

Hello,

I looked up the specs on your UP3000-M3322. When the output is set to 230V, the allowable fluctuations are -10% to +5%. That corresponds to 207V - 242V. That's pretty much what you described seeing.

As the inverter air conditioner compressor is unique to other compressors in that it is driven by varying power, it's probable that the output of your UP3000 isn't stable enough to keep the compressor running smoothly, hence the shaking.

View attachment 103896

When I set the output at 220v the A/C shakes more than if I set output at 230v, it should be less as the fluctuations should be smaller. How can it be?
Now I got my hands on an oscilloscope, not sure if i will be able to sort it out

 

[No Idea]

I ran a small chest freezer for 2 years no issues 230vac then it sat unplugged unused for a few months. Finally got around to buying one of those power outlet adapters that you can put a clamp meter on to meassure current. Plugged it into a mains extension cord, turned it on to measure inrush current, and the compressor just about shook itself out its mounts. The whole freezer was jumping. Turned it off after a few seconds thinking the compressor was done. Checked everything no issue. Turned it back on - ran quiet and smooth as normal. Tried the adapter for the clamp meter no issues second time. Never did figure out for sure but the only noticeable thing was the adapter for the clamp meter is a very loose fit in the extension cord outlet. All I can think of is I didnt jamb the adapter in hard enough the first time and the loose contact caused a voltage drop and the compressor never started properly. So long story short i think low voltage.

AI D seems to be on to something with suggesting the acceptable output tolerance of your inverter. My cyberpower 1000va UPS has options to narrow the voltage tolerance as well as the voltage output. Mine is an AVR/UPS.

 

[Al D]

This is from your AC unit's manual:

 

[Al D]

This is from your AC unit's manual:

View attachment 104195

This explains why you can't run it on 220 even though it's more stable. It seems the device needs a stable 230V, which doesn't appear to be possible with your current setup.

 

[Stever73]

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