Mitsubishi Electric and Daikin power consumption

[motoguy128]

On a air cooled scroll chiller for example (A CGAM, since that's what I just bought recently), it's a multicompressor, multistage fan arrangement with a VFD on the 1st fan stage so it may be different 100% load (all 6 compressors running) it's only 10.3 EER, 75% = 14.4EER, 50% 17.8EER, 25% 20.3 EER. Rotary chillers are similar. Fixed speed Centrifugals rely on mass flow and have fairly constant effciencies.

My 2 Infinity unloading scrolls are more effceint at cooler temps on low stage. They are about equal at 85F, and high stage has an advantage over 95F. In heating mode I think 1st stage has a slight advantage over 60F (yah) 50-55 it's a bout equal. 40-50 slight advantage to high stage, below 40F 2nd stage has a clar advantage. But in most climates when properly sized, you're already in 2nd stage below 35F most all the time.

Multi or vairable does give you an advantage of longer cycles. I'm not sure if that's factored in to SEER ratings. I don't think it is, but the higher lower ambient effceincies are. I think effciencies otherwise are calculated at steady state.

[gschra11]

True, you have oversized ID and OD coils and low ambient air helping you out. But you are also operating a huge compressor at a fraction of its design capacity. That's lots of moving parts for very little cooling. Most motors, compressors etc that I have seen do not get the same efficiency when operating at low load. Also, at a certain point, massively oversized coils begin to work against you not for you do they not?

I don't know this for a definite fact, if somebody can prove me right or wrong with manufacturers data, that would be terrific.

But in my mind, if the unit gets 18 SEER at 100%, I would only expect maybe 14 SEER at 25%

To an extent, I would agree with you.

By virtue of their need to cool, compressors have a certain minimum RPM that they must maintain in order to ensure that they get adequate lubrication and motor cooling, just like any other ECM motor or VFD-driven motor. If the system doesn't require that minimum amount of pumping, then it must either unload or cycle. I don't think that all of them are this way, but my Daikin RXYMQ48PVJU (VRV-SIII) has an unloader. If it runs below the minimum compressor speed, it unloads. This reduces capacity but not power (OK, it reduces power slightly but it is very slight at that low speed) somewhere around the 25%-30% loaded range. It also gives me a COP of about 1 if the compressor is running unloaded trying to hit capacities below 15%.

I have four units attached to my system. When the 1st starts, the system pulls about 4 amps in cooling. When the 2nd starts, it pulls about 4.5. When the third gets going, it will run at 6.5. When all 4 run, the system pulls about 11. Clearly, this isn't a linear curve. The point, though, is that the system runs best and most efficiently when it is around 50-80% capacity. It doesn't like really low loads, nor does it like full load. Conveniently, the weather tends to cooperate in this regard.

I know its not manufacturer's data, but a submeter on an actual system installation is probably a bit more realistic anyways.

Just my observations to-date.

[LKJoel]

To an extent, I would agree with you.

By virtue of their need to cool, compressors have a certain minimum RPM that they must maintain in order to ensure that they get adequate lubrication and motor cooling, just like any other ECM motor or VFD-driven motor. If the system doesn't require that minimum amount of pumping, then it must either unload or cycle. I don't think that all of them are this way, but my Daikin RXYMQ48PVJU (VRV-SIII) has an unloader. If it runs below the minimum compressor speed, it unloads. This reduces capacity but not power (OK, it reduces power slightly but it is very slight at that low speed) somewhere around the 25%-30% loaded range. It also gives me a COP of about 1 if the compressor is running unloaded trying to hit capacities below 15%.

I have four units attached to my system. When the 1st starts, the system pulls about 4 amps in cooling. When the 2nd starts, it pulls about 4.5. When the third gets going, it will run at 6.5. When all 4 run, the system pulls about 11. Clearly, this isn't a linear curve. The point, though, is that the system runs best and most efficiently when it is around 50-80% capacity. It doesn't like really low loads, nor does it like full load. Conveniently, the weather tends to cooperate in this regard.

I know its not manufacturer's data, but a submeter on an actual system installation is probably a bit more realistic anyways.

Just my observations to-date.

I had kind of limited my mind to a variable speed compressor and VS motors when considering this scenario. And have seen before that %25 of full load does not equal %25 of full load amp draw
It's an interesting discussion to be had though

[ACPM]

You do realize that efficiency is not constant as capacity changes. Your MS will be most efficient at 100% output, and begin to decrease as output decreases. You reach a cross over point where the losses involved with starting and stoping are less than the losses involved with running at really low inputs.

You could always go for multiple smaller units if it was of such a concern

Thank you for your comments. Efficiency in an actual application is difficult to define and measure but I believe in a lab with a set airflow and defined humidity and internal and external temperatures, it is practically constant if you consider the coefficient of input power v. cool air produced.

[ACPM]

...and your point is. That's normal. All building through out hte world see low sensible load conditions at nighttime. However in your climate up north, latent load shouldn't be an issue except for a few weeks in winter.

I'll repeat it again. The ONLY way you can precisely control building load with a 10:1 ratio is hydronics using chilled or hot water on smaller scale (below 6 tons). End of story, case closed, no point in digging further. Now, at 6 tons, you could install 2 zoned inverter driven split systems (like Carrier Greenspeed) and each only runs at abotu 30% capacity and you get 15% between 2 systems.

Even with chilled water, the chiller will still only turn down to about 75% capacity and the boiler typcially 5:1, but you can use a storage tank in either case to prevent short cycling.

In you climate, actually, it's possible to do radiant floor and wall cooling. If you have high mass floors, here also a way to prevent short cycling. Then you just need a dehumidifier or HRV for dehumidification.


Finally on the point of a "cold blast". If the equipemnt and ductwork or with mini splits, the indoor units are sized properly or located correctly then there should never be a cold blast from the equipment. In my home even on high stage in heating or cooling, the home never feels drafty. In cooling mode you can feel the cool dry air a little if the system is in a dehumidificaiton on demand mode (lower airflow). But it's actually refreshing and not cold at all.


A properly sized and installed HVAC system is silent, heats and cools evenly, even at low load conditions, and never short cycles, unless it's a small home with out much mass in the building materials. Some homes will short cycle in low load no matter what you do. The space jsut doesn't have enouhg volume or building materials in proportion to it's peak heat loss ang gain. Remember, a smaller home have more surface area to volume and mass ratio. So they have more temperature and load diversity. A larger home with plaster, stucco, tile, hardwood, etc, can take a hour to warm up or cool off even 1F on a fairly hot or cold day. That means when the system comes on, even with no load, it can take almost 10 minutes to raise or lower the temp 1F, which a good controller will maintain if it can stay under 4 cycles per hour maximum.

First thanks for such a detailed reply.

A water based system is not an option as this is an old building with ceramic walls and marble floors but more smaller air units is. The issue is will larger units consume more power to over produce cool air in comparison with smaller units which have a lower min cool air production output but reach 75% of the larger max output.

The specification calls for zoning so only the rooms in use will have the ducts open in the evening. The "cold blast" concern is that even though the conduct will be sized accordingly (to max airflow with all outlets open) and the internal unit will reduce the airflow to a min, there is a case when all the cool air could be directed at just one room which actually needs only 50% of the larger machine minimum output.

I'm edging towards slightly under powering the installation. We can forget the night time areas with impulsion and return motorised vents. Also, we can specify that with extreme external temperatures, one zone will have to be closed off so that the rest has enough cool air when the machine can only produce 70-80% of peak at or beyond the max outdoor temperature. In return for this, a smaller machine will consume around 140kWh less per year based on the new ESSER rating.

[ACPM]

On a air cooled scroll chiller for example (A CGAM, since that's what I just bought recently), it's a multicompressor, multistage fan arrangement with a VFD on the 1st fan stage so it may be different 100% load (all 6 compressors running) it's only 10.3 EER, 75% = 14.4EER, 50% 17.8EER, 25% 20.3 EER. Rotary chillers are similar. Fixed speed Centrifugals rely on mass flow and have fairly constant effciencies.

My 2 Infinity unloading scrolls are more effceint at cooler temps on low stage. They are about equal at 85F, and high stage has an advantage over 95F. In heating mode I think 1st stage has a slight advantage over 60F (yah) 50-55 it's a bout equal. 40-50 slight advantage to high stage, below 40F 2nd stage has a clar advantage. But in most climates when properly sized, you're already in 2nd stage below 35F most all the time.

Multi or vairable does give you an advantage of longer cycles. I'm not sure if that's factored in to SEER ratings. I don't think it is, but the higher lower ambient effceincies are. I think effciencies otherwise are calculated at steady state.

I would love to know where you get ERR breakdowns by load. This data is not released normally in Europe, or at least Mitsubishi, LG and Fujitsu dont give it.

I believe SEER is based on constant state load of 100% @ 35 degrees, 74% @ 30, 47% @ 25 and 21% @ 20 with an internal temp of 27degrees (dry) /19 (wet). There is no weighting for the technology used to obtain the required load.

[ACPM]

I had kind of limited my mind to a variable speed compressor and VS motors when considering this scenario. And have seen before that %25 of full load does not equal %25 of full load amp draw
It's an interesting discussion to be had though

VRV systems are clearly very different in terms of input power usage.

The daikin split concealed unit databooks suggest that the most efficient "seasonal units" have input power coefficients of 0.33-0.40, that is 30-40% of the nominal power rating is consumed at the lower cooling limit (around 40% of the max cooling load). The graphics they show are linear, that is input power rises linearly from the min cooling capacity though the the nominal used for EER to the maximum for a constant airflow and temperature.

I assume that cycling begins below that.

[ga-hvac-tech]

First thanks for such a detailed reply.

A water based system is not an option as this is an old building with ceramic walls and marble floors but more smaller air units is. The issue is will larger units consume more power to over produce cool air in comparison with smaller units which have a lower min cool air production output but reach 75% of the larger max output.

The specification calls for zoning so only the rooms in use will have the ducts open in the evening. The "cold blast" concern is that even though the conduct will be sized accordingly (to max airflow with all outlets open) and the internal unit will reduce the airflow to a min, there is a case when all the cool air could be directed at just one room which actually needs only 50% of the larger machine minimum output.

I'm edging towards slightly under powering the installation. We can forget the night time areas with impulsion and return motorised vents. Also, we can specify that with extreme external temperatures, one zone will have to be closed off so that the rest has enough cool air when the machine can only produce 70-80% of peak at or beyond the max outdoor temperature. In return for this, a smaller machine will consume around 140kWh less per year based on the new ESSER rating.

Not trying to be a 'know-it-all'... however this is the reason a zoned system can have smaller equipment. MAX load only happens on the few most extreme days... and generally if the system is not operated with setbacks (day/nite temps), this is a small issue.

IMO folks who do residential zoning know the things you described...