AirCon : Inverter vs Non-Inverter questions

General Discussions
1

We’re schmelting!

So most info on Inverter vs Non-Inverter AC units favours the prior on energy consumption and silence and and and more. Except cost.

I am curious though on how much more energy efficient the Inverter units are. Struggle to find clarity, feel like a kid on a tricycle in gravel with Tom Waits laughing at me…

So, if anyone can weigh in I’d be thrilled.

Basic facts:
Area is 39.5 cubic metres. (3.4w x 4.3L x 2.7h)
Insulation okay…curtains rather than doors
South Facing, two windows, non-glazed.
1 person laptop office.

Thinking a 9000 BTU unit will do…

So…*how much more energy will a non-inverter consume vs inverter, I mean roughly?
% or per hour rough guide…

Also
Installation instructions are vague…
Do these units go into a 15A plug point or is it “a new breaker in DB situation”

Sorry if these are dof questions, just juggling a lot right now and hoping for some info from folks who are in the know

Kind regards
c.

2

Depending on who you ask, you will be given some number between 30% and 70%. I think it is much closer to the lower end, and this paper seems to concur, placing it at “about 35%”.

As I understand it there are two factors behind this.

The first and most prominent is that your single-speed compressor has a very high start peak (as much as 500% of its nominal power), and there are higher losses associated with that kind of inrush.

The second has to do with the energy needed to create a state change in water, in this case specifically the water vapour in the air. The energy needed to cool water down is linear up to the point where it changes state (from gas to liquid), when the energy requirement shoots up. An intelligent air conditioner can avoid cooling below the dew point, which affects the relative humidity inside the room and requires a lot of energy. It is unclear whether the average split-AC even has a dew-point sensor, but in some hvac systems it is a component.

(That is also why you don’t crank the AC down to the lowest it will go, if that is below the dew point, the AC might spend a lot of time essentially turning air into water without making the room any cooler).

Yes, for a floor space of about 15 squares, 9kBTU is sufficient.

This depends. When you sell the house, the sparky will tell you in no uncertain terms that this does not comply, and you must immediately rectify the situation at the cost of thousands to install a new breaker.

My understanding is that this is not necessary. First, if we consult our copy of SANS-10142, that classifies an air conditioner as a “fixed appliance”, that’s a device that cannot be moved without using tools.

Selection_206

You are allowed to have mixed circuits up to 20A, that is, fixed appliances can share a plugs circuit:

Selection_207

6.15.4.3 essentially requires an earth leakage, and 6.16.3.2.3 is about cooking appliances. For air conditioning and heating, if the device needs more than 16A, then it must have a dedicated circuit:

Selection_208

But a 9kBTU air conditioner is going to run at more like 3 to 4 Ampere.

You must however make it possible for the AC to be disconnected (for servicing), which means either the plug needs to be accessible to the technician, or you need to install an isolator:

Selection_216

And this disconnector should be within 1.5 meters of the outdoor unit, in a waterproof box probably:

Selection_217

And finally, you can’t have it on a multiplug adapter. You can however replace the socket with one of those dual-socket outlets, and designate one of them as the permanent power of the AC:

Selection_218

That’s my understanding of how it works, but I am not a sparky, and if I am wrong, then I will accept information to the contrary.

2 Likes
3

Thanks so much Plonkster!

4

I would be wary of trusting the findings in that paper - it seems that they are trying to compare units with different gasses, which is a can of worms in itself. They will have different properties depending on the environment. Then the units also seem to be different actual capacities and both undersized for the application (it looks like they never satisfy - they are running full out all the time).

But overall, for a 9000btu aircon, the price difference between inverter and non-inverter is so small that I would take the inverter just because the on-off cycling of the non-inverter is so annoying.

1 Like
5

To be honest, my gut feeling is that the difference cannot be all that much. If you cornered me on the street two days ago, I’d have said “20%… maybe”. Let me layout my thought process.

Ultimately, both units move the same amount of heat from inside the room to the outside. They do the same amount of work. So there is no difference in energy here.

One unit keeps a more level temperature, while the other has slightly wider temperature swings, but I still expect that the same average temperature holds, and hence we’re still moving the same amount of energy from inside to outside. So still no inherent advantage I can think of.

The COP of the mechanical part, how well it works, this (too) I would expect to be very similar. Newer inverter units may have a better COP, simply because they are newer, not necessarily because of the way it is driven. There may be a slight efficiency advantage in driving the pump at a slower speed, but then I would also expect that for the single-speed unit, the pump is optimised to work optimally at that speed.

That leaves only the last two, which is the startup transient (which lasts less than a second), and optimisations related to the dew point (which most of these units don’t even consider).

So… supposedly up to a third of the energy is in the startup transient? I am skeptical. It could be, but I would not accept that just because marketing types told me so.

1 Like
6

I am so bad with keeping records of the stuff I test to satisfy my curiosity. Way back when I started my solar journey I did this test in my own home and was surprised to see how well these Inverter aircons fared against conventional aircons. I had a conventional 9000 BTU in my Daughters room and had an 12000 btu inverter aircon installed in my own room and another 9000 btu inverter aircon installed in my office.

I monitored it through a Sonoff POW and logged the data to Home Assistant. Wiped the image later to install a new instance of home assistant and never backed the data up.

My findings in short.

1: 12000 BTU did the best over a 24 H period, mainly because it was oversized for the room and went into idle mode sooner than the 9000 btu in the office.
2: 9000 BTU in the office out performed the 9000 BTU conventional marginally in that specific area because of the fact that the ambient temp in the area only dropped enough to allow the aircon to work less after 22H00 in the evenings during summer.

Points to take into consideration.

1: Upsize if you can choose between a 9000 and 12000. Upsizing to an 18000 for example will hold no additional benefit as its idle current is just too high to ensure any savings at all.
2: In exstream warm days, even the inverter aircon will remain at full running current till the ambient temps drop low enough to allow for any reduced power consumption.

I have also seen that the full load current rating on the older conventional 9000btu unit was 1150wh compared to the full load current rating of the inverter 9000btu sitting at around 950wh.

Looking at that rating in your comparison, you might find that the constant running current in the same brand might differ slightly between inverter and non inverter models. There you might save 100-200 wh in normal conditions with an additional saving in the cooler parts of the day when these inverters reduces speed and just idle away at 250-320 wh in the early hours of the morning. In Phalaborwa from about 1 in the morning.

All together they are more efficient, how much will depend on a few factors and might vary from season to season.

EDIT: The reason I would install an inverter unit is the demand on my inverter as well as battery bank. So for solar installs, take the inverter.

3 Likes
7

Thanks for the responses.

I’m guessing I might struggle to run an AC unit off our small 48|12|13 Multiplus & Pylon US2000C?

But, thus far, I’ve managed to run a desktop computer, music, laptop, WiFi, a fridge and washing machine all at the same time, albeit midday, whilst completely off grid…so I feel plucky.

But let’s say with ESS will my system manage and will there be much difference between a 9000 vs 12 000 Btu unit (830W vs 1020W), ie “will my inverter be screaming at me non stop?”

(apologies, it’s been a while since I’ve had a close look at solar systems)

8

I cannot quite make out what model Multiplus that is (48/12/13?), but suffice it to say that a 3kVA inverter cannot start a 12000btu non-inverter AC. I’ve tried… well not me, the wife, inadvertently. In my previous house the rule was “ACs off before load shedding”, and she forgot and I didn’t know it was on. It was probably borderline, but the Multi died in a spluttering/flickering overload when the York unit we had tried to start.

A 5kVA might be able to do it.

As @_a_a_a said, in this case, pay the little more extra for the inverter AC, then you don’t have to worry about this.

You can run 80% of the nominal rating of your inverter constantly and it should be fine. Over that, it will eventually overheat and switch off, unless ambient is very low, or you add extra airflow.

9

I have the 5Kva.

During really hot nights we run a 9000 & 12000 non inverter simultaneously, but when the grid is off I suppose my Multi will also not be happy with me.

Will VRM be fast enough to capture those cold cranking amps when I start up one of them? I would like to see what it spikes to.

10

Nope. You have about a 1% chance of catching that spike when you set the log interval to 1 minute.

Your best bet is a current clamp with a max-hold feature.

11

Sorry, 48 | 1200 | 13

Yes, we’re looking at a inverter AC.

Thanks

12

Using NodeRED, when the panels can carry the load, inverter is set to run at 4.8kw under ESS.
Thought 5kw will be pushing it a bit.
Now and then I get a Overload Warning.
And yes, extra airflow was added.

Ps. This ties back to me setting inverter max watts under ESS at what the panels can do. Being Cpt and bad weather weeks.

13

Then running an AC at around 900W plus all your other loads, on a sweltering summer’s day, is probably looking for trouble. It should work, until the unit starts to run hot.

It reminds me of running my old 3kVA with the dishwasher (which has a 2000W draw when heating water). At night, it is alright (other loads are low, temperatures are low). During the day, it would raise temperature warnings and if you didn’t pay attention, it would eventually switch off.

I could get a load of dishes done by opening the door of the machine, waiting for the Multi to cool down, and then continuing, but it was definitely not ideal :slight_smile:

1 Like
14

Saw mine derate to like 1600w on warm days where it was installed at the time.
Lost a lot of sunshine.
Made me replace it with the 5kva.

15

Sorry, is this whilst off-grid or still grid tied (ESS)?

So for me at 1200VA that’s 960W…the 12000 btu unit in question runs (cooling) at 1020W…meaning no-no.

16

While off-grid. While grid tied, it will only run at 80% anyway, and the rest will come from the grid.

1 Like
17

That’s actually not bad. I assume that is not the maximum power value. I just looked at the Daikin R32 unit I installed this week, spec sheet says nominal is about 1kW, but peak is 1.4kW.

Also interesting, is that it is really a 3300W unit (cooling) at 1kW electrical use (so COP of 3.3), and 3.3kW (at 3.412BTU/h per watt) is about 11200 BTU :slight_smile: So cheating a little…

18

Yeah, it sometimes also counts in your favour - an effect of having these binned sizes - 12, 18, 24… kBTU. I’ve seen some where the 24k is actually 27k and the 30k is 28.5k…

19

I think the big difference is when you also have a variable expansion valve (usually in the form of an expansion valve + throttle). Then when the pump can run at fractional speed, it’s the same as increasing the size of the condenser/evaporator coil relative to the aircon size. Then you can run with a smaller delta temperature and because your maximum theoretical EER function has Tdelta below the line, the EER goes up.

Basically when the temperature outside is relatively close to what you want inside (or below), the EER of the inverter will be better.

The other thing is that the inverter pump will draw power proportional to the pressure difference (the variable expansion valve also helps here). Non-inverter will draw about the pump rating in apparent power, with the power factor varying based on pressure ratio. But the extent to which the PF can vary is quite narrow, so it draws quite close to the rated power, even when it does not have to do much work. Similar to a lawn mower using ~50-60% of it’s rated power even when it’s free-running.

1 Like
20

This is what I love about this forum. You filled in another gap in my understanding. Thanks man!

Do you think a variable expansion valve would be enough to reach 20% or even 30% less consumption? Maybe if Tdelta is always small? :slight_smile: