Convert SEER to EER with monthly coefficient 5

[treddie]

Hi all.

I wanted to tag this question onto a similar thread, but the site would not let me.

Anyway...As far as I understand it, EER and SEER are different in that SEER is an attempt to give an annual EER, so that a yearly average across seasons is possible. That being the case, it seems that there should be graphs that represent some "SEER-coefficients" based on month and geographic location that can be multiplied by SEER to get an accurate EER. Obviously, the published conversion, EER = -.02*SEER^2 + 1.12*SEER (and its inverse, which is a quadratic) leaves much to be desired as far as accuracy is concerned.

Do such tables exist? Or is it something that has never been considered all that relevant?

Thanks for any answers!

 

[Drazen]

i am not sure what do you want to accomplish.

both seer and eer are obtained by measurements, graphs can only serve to give better visualization.

relation between eer and seer much depends on equipment construction. some statistical trending does not appear to have too much practical usefulness in such situation.

some manufacturers claim that their opponents are intentionally making products which have decent or even very good eer for referent conditions, but as referent conditions occur only few days in a year, such results do not contribute to energy efficiency.

that was the main argument to introduce seer at all, and it is expected that eer will go to history soon enough.

 

[treddie]

Semi-bummer. I was building a tiny app that was meant to calculate energy costs of running an A/C in any month of the year and it would be nice to stick with EER for that. But if you only know SEER, then there is no accurate way to convert back to EER, unless there were some handy coefficients that could weight SEER, in that regard.

 

[Drazen]

assuming that is not your field of expertise, I am not sure whether you understand what stands behind eer and seer.

eer cannot be used in any way for year-round calculations as eer is valid only for specified referent conditions, outdoor temperature in first place. it is not statistic average, which is main problem of eer, and that is why seer is introduced to give weighted seasonal average. so for your plan, it is last thing your would want to do, to convert seer to eer.

yet, seer will not be too helpful for some accurate estimates as you simply need to know assumptions, like operating schedules over the whole period of estimate.

playing with simple numbers will give you nothing useful. that is very similar to, for instance, average fuel consumption advertized by car manufacturers many years ago. they often used referent conditions like 2-3000 rpm, flat road, 3rd or 4th transmission gear, which are ideal conditions, with minimal consumption. such number is useless for practical driver, especially if he lives in mountainous area or area with high urban traffic etc. nowadays car manufacturers offer "combined" consumption figure, which is something similar to seer.

 

[DRWeig]

That would be a tough one. I had a quick thought of programming a simple building into one of the energy simulation programs, then extracting the equivalent of an SEER by month.

However, it depends on more than just weather and sunshine. Building orientation, percent and type of glass, insulation levels, ventilation levels, internal loads, operating hours, and other factors play into the load that is put on the machine. That, plus outdoor temperature all play into efficiency.

Sorry! It's not a trivial problem.

Best to you,

Goober Dave

Haven't see the forum policies? Do so now: Forum Policies

 

[treddie]

Sorry for the delay in getting back to everybody's responses. What prompted my OP in the first place was my central air system biting the dust in the middle of a heat wave, and needing to find an efficient interim solution involving a portable and a window unit for a two story home. I have been installing the units and had to return one unit due to missing parts, along with figuring out how to modify what I have for my particular setup, AND keep the window alarm sensors working.

"assuming that is not your field of expertise"

You are correct. Although I did come up with a solution that was a fair approximation (+- $30/month).

"Sorry! It's not a trivial problem."

I am going to do a full differential equation solution just for the heck of it. I started it some years ago and almost had it completed (as far as nailing down the heat load and the CFM-rate required to replace it), but things got in the way of finishing it. That will give me an accurate idea of the amount of heat that has to be replaced at any given outside temperature, and predicting a month in advance can be based on historical outside temp averages for my town. That will be the alternative to my current solution which was based on entering a representative KW-Hrs from some previous year's month, and subtracting a total "base" KW-Hrs (No A/C or blower running) from it to get to an estimate of power used for A/C and blower. That solution takes an accuracy hit, in that the base value assumes that the sum of everything else in the home is a constant value, which it is not.

Beyond that, the only way to refine the accuracy for the specific A/C compressor will be to weight the SEER of the A/C based on some historical data about my town. That might end up being an approximating curve built by trial and error.

"I had a quick thought of programming a simple building into one of the energy simulation programs, then extracting the equivalent of an SEER by month."

That's a cool idea!........No pun intended.

 

[MintJulep]

The way to do it would be using Coefficient Of Performance.

However COP is not a constant - it varies with condensing temperature and to a lesser degree evaporator load.

And you'll likely have a difficult time getting the data from the manufacturer.

SEER and EER are developed at a specific set of conditions that likely rarely occurs wherever you live, and are useless for what you are trying to do.

 

[treddie]

"The way to do it would be using Coefficient Of Performance. However COP is not a constant - it varies with condensing temperature and to a lesser degree evaporator load. And you'll likely have a difficult time getting the data from the manufacturer."

Well, then there is no solution other than an approximation, since as accurate as you can be with a differential equation solution for the "front end" of the problem, the efficiency of the unit ("back end") is still the problem spot. If a complete approximation is only +- $20 or so, that's still pretty good.

 

[DRWeig]

If you're interested, all of those equations have been approximated in the free program eQUEST from the US DOE. You could input a simple house with a DX unit that you define, and extract monthly (or hourly) data. You get to control the space load and pick the weather location.

Best to you,

Goober Dave

Haven't see the forum policies? Do so now: Forum Policies

 

[treddie]

Ah neat. I'm definitely interested in checking out the program! Thanks for the link.

 

[MintJulep]

The capacity and power use of a refrigeration compressor can be modeled with a third-order polynomial. COP = capacity/power

I wrote this Excel VBA function some years ago.

SST is Saturated Suction Temperature
SCT is Saturated Condensing Temperature.

Note that the constants are specific to a specific compressor.

Public Function Compressor(out As String, SST, SCT)
Dim constants As Variant

Select Case UCase(out)

Case "CAPACITY"
constants = Array("zero", 41873, 1517, -339.3, 21.9, -10.77, -0.45, 0.12, -0.15, -0.01, 0)
Case "POWER"
constants = Array("zero", 2359, 29.3, 79.3, 1.45, -1.75, 0.31, 0.01, -0.04, 0.02, 0.01)

End Select

Compressor = (constants(1) + constants(2) * SST + constants(3) * SCT + constants(4) * SST ^ 2 + constants(5) * SST * SCT + constants(6) * SCT ^ 2 + constants(7) * SST ^ 3 + constants(8) * SCT * SST ^ 2 + constants(9) * SST * SCT ^ 2 + constants(10) * SCT ^ 3) / 1000

End Function

 

[treddie]

I'm a VB guy so that's perfect. But will the manufacturers give up info on the compressor constants?

 

[MintJulep]

Not likely.

 

[treddie]

Bummer.

 

[DRWeig]

That's cool, Mint.

Along the same line, I screen-capped one of the basic inputs to eQUEST for a residential system for EIR (unitless Energy Input-output Ratio) as f(outdoor air temp and entering evap wet bulb temp). The coefficients you see are the built-in ones from DOE2, for the bi-quadratic function. There's another curve fit (not shown) that you need, the EIR as a function of part-load ratio for a cycling machine. It's in the program too. MintJulep's curve is more exacting, but these should be pretty close and repeatable.

You can VBA these pretty easily, treddie.

As for manufacturer's data, it can be had with leverage for these functions. Example would be if you were using eQUEST to fight for LEED accredation for a building. As a designer, I could pit Trane vs Carrier vs York for the project, and demand the coefficients from each. If the job is big enough, they'll give 'em.

Click here for screen captures. The first page is the basic choose-the-curve input. The second is the curve itself.

Best to you,

Goober Dave

Haven't see the forum policies? Do so now: Forum Policies

 

[treddie]

Thanks for all the help and links! I now need to start playing around with eQuest! Looks pretty darn fun.