An upcoming article on energy-modeling software
I am planning to write an article for Fine Homebuilding on energy-modeling software.
There are dozens of programs available. Some can be downloaded for free; some are expensive.
It seems to me that different professions need different types of software. The best software for a builder is probably not the best software for an HVAC contractor, an architect, or an energy consultant.
I’d love to hear from GBA readers about their favorite software and their most-hated software. What do you use it for? How long does data entry take? Do you have any idea if the software is accurate?
Here are some examples of software I might mention:
– Passive House Planning Package
– eQUEST
– EnergyGauge
– HOT2000
– REM/Design and REM/Rate
– RESFEN
– REScheck
– ACCA Manual J
Thanks for any feedback you can provide.
GBA Detail Library
A collection of one thousand construction details organized by climate and house part
Replies
I would have to say Retscreen, it's the most popular and intuitive program out there I would have to say but of course it depends on what your going to be doing as some of the programs you listed obviously report different types of data.
Two others I know of but have not used are HVAC-Calc (http://www.hvaccomputer.com/) and the
Writesoft packages (http://www.wrightsoft.com/).
I hope your article points out that the results of canned software are only as good as the input data and level of detail. Such packages are designed to make use of them easy, through such things as drop-down selection of R values of envelope components and air infiltration rates. Especially with highly energy efficient homes, the data for some things may have values outside the range of built-in values. Presumably the better packages let the user enter specific values where available and not rely on built-in values.
It takes a lot of time and attention to detail to come up with a good energy model, as you well know. I found it more satisfying (and more accurate) to build my own spreadsheet to do the calculations presumably being done behind the scenes of the canned software packages. The difference was that I knew what I was calculating as opposed to guessing at what sombody else's software does.
I have to admit that I have not used any besides the PHPP and several other German software (which would be irrelevant for your article). I like the PHPP for several reasons. The biggest one is similar to what Dick mentioned. It is Excel based and you can basically track everything the software does. It is not always obvious because the sitecalcs run in hidden fields. But you can unprotect the sheets, unhide the hidden fields and track every link if you really want to. Also you can insert sheets and produce the graphs you might want or make conversions (for example metric to Imperial). Obviously all these features are not intuitive (at least not for "non geeks"). But there is more that I know the PHPP can do and I am not so sure if the other above mentioned can!?
Just a few thinks the PHPP takes into account that I can think of of the top of my head:
- Solar heatgain to reduce heat demand and peak heatload
- Thermal bridging (Sitecalcs might necessary)
- Heatloss of plumbing pipes and hot water tank
- Ventilation losses according to the measured blower door test
Some of these points might be small enough to neglect in standart construction but make a real difference for high performance buildings. The PHPP however is probably not perfect either. It is still a stationary model and effects like thermal mass might fall short.
To your specific questions:
What do you use it for? Integrated design of high performance buildings
How long does data entry take? Depends on a lot of thinks :-). I personally think if you are skilled at it you get pretty accurate results quickly but refining it can take quite a while.
Do you have any idea if the software is accurate? It has been extensively tested in Europe and I don`t see why physics would change here. As you know the energy consumption depends a lot on the people who live in these buildings and unfortunately there is probably not enough data yet to really validate it over here. At least I have not seen any extensive studies.
From a design standpoint by far the best energy-modeling software package is undoubtedly the one which is fully integrated into the CAD software and directly uses the extensive information already in the three-dimensional CAD model: wall area, thickness and orientation: window dimensions, orientation and shading, roof aspect, slope and dimension, etc. etc. etc. Integrated dynamic modeling allows all design options to be explored on the fly to fully optimize every aspect of the performance of the home.
The only problem with this package is that, as far as I know, it doesn't exist.
Nicholas, Dick, Philipp, and James,
Thanks!
Readers: I look forward to more comments.
James,
That's what you call good software design. Personally as a person with a background in compsci/software eng, my preference would be for energy modeling tools that provide a list of required fields, which would enable me to build a abstraction layer above the different 'modeling theories', by which I could tie a single data input (via hand entry or automated CAD extraction), and feed it into the API's of various toolkits.
This really won't happen though until people stop protecting their IP. I keep hearing how Passive House is open and not-proprietary, and yet, last time I checked, I still have to pay for an excel spreadsheet.
The building world needs to try taking a page out of the software open-source community.
Just my 3 cents.
* http://uwspace.uwaterloo.ca/bitstream/10012/5267/1/Brittany_Hanam_Thesis.pdf *
Good abstract there Martin, about the state of affairs of the modeling world.
@james,
there are folks attempting to integrate PHPP with revit - which would accomplish what you're looking for.
@philipp,
regarding the thermal mass issue - i think that it tends to be utilized more in europe where typical construction still tends to be heavy mass (e.g. kalksandstein - similar to CMU). when playing around with PHPP & thermal mass (summer sheet) - it seems that locations with high insolation in winter (such as denver, salt lake, calgary) seem to get the bigger 'swing' in reducing heat loads. thermal mass also seems to be taken into account more on commercial buildings in EU, as way to absorb heat generated throughout the day and eliminated via night cooling.
Martin, We use REM/Design and PHPP.
We own Eco-Designer, a 3D BIM plug-in for ArchiCad that aims to do what James is dreaming of, but haven't gotten our templates and libraries aligned well enough to get valuable output yet. It's on the list for next year.
http://www.graphisoft.com/products/ecodesigner/
John,
PHPP is open-source, but not free (free as in speech, not free as in beer...). The source code of the calculations is visible and reviewable, it's not "black-box" like most energy software.
There is nothing stopping you from developing the add-ons you suggest, in fact, the community would welcome it, I'm sure.
I have experience using PHPP, REScheck, some of the Visual DOE variants including reviewing documentation for commercial projects trying to achieve LEEDs energy credits. I consider myself an amateur enthusiast not an expert.
Not too long ago I entered a residential design into a REScheck variant per a state code requirement. The architect of the project paid little consideration to the concept of a coherent and continuous thermal envelope. The project had a ton of windows and therefore more framing than space to insulate. The best I could describe the wall for the data entry in REScheck was that it was an R-19 2x6 wall. Moral of the story - energy modeling means little if the design doesn't follow the basics of a well detailed thermal envelope. At least some of the complexity of the PHPP forces recognition of this in the data entry.
A topic more interesting to me is the mathematics behind modeling. Calculating UA values is pretty straightforward. The math behind accounting for passive solar gains too is easy for me to imagine. Heat losses/gains through air infiltration/ ventilation is still a mystery to me especially considering Joe Lsti's recent paper on the topic. Passivhaus folk often point out the significant energy improvements attained by increased air tightness but I'd like to understand the math behind those determinations.
I like PHPP but I have lots of questions about it but no authority whom to ask. As Philipp states "its all physics" but also recognizes to create a reasonable model the 'real' physics needs to be simplified. I suspect the part of the Passivhaus paradigm of designing a thermal envelope is in fact to simplify the physics of the conditions being modeled - high level of air tightness, minimal thermal bridges, balanced heat recovery ventilation. So I wonder how accurate the PHPP is for designs that don't follow the standards strictly particularly around the issue of air tightness and ventilation.
J,
To echo your point, I've been specifically warned not to use PHPP to calculate heating loads for equipment sizing for buildings that aren't going to be built close in quality to Passivhaus, that PHPP assumes that it will be a high quality building that can "float" through short term temperature fluctuations which allows one to reduce the peak loads accordingly.
I don't have enough knowledge of the equations to know what exactly it is doing, but that could be a good example of the kinds of simplifications it must be making as a spreadsheet based program.
Martin,
Perhaps out of the arena that you were planning on, but you may want to get familiar with WUFI 3D also. I only caught half of the presentation, but the latest versions emerging from Fruanhoffer supposedly supply surface temps at variable locations including the face and within the assemblies themselves. If I had heard it right, then one could potentially track loads as well as moisture levels based on actual climate locations.
I don`t think this question has gotten enough attention so I am trying to give it a little boost in order to make it to the first page again. To include some technical stuff and may spark some vital discussions:
@Mike: To my knowledge the thermal mass in the PHPP summer sheet does nothing to the heat load or heat demand in any climate. What it does do is lower overheating. That is where my critique is coming from. In Germany it has been analyzed that the solar gains captured during the winter day do usually not cause overheating but in the summer the thermal mass plays a bigger role. Here in Minnesota we usually have a lot of sun in the winter but it gets really cold. Meaning in a passive house it might be really cozy during the day but the heat still have to kick in at night. That it when thermal mass gets handy and the PHPP does not really help. On that note I should also say that no software will ever replace good working human brains :-).
Philipp,
Agreed on the working brains part. My experience w/ uber cold climates is fairly limited, mostly familiar w/ 3000-6000 HDD climates. All our PH projects tend to have a somewhat passive-solar aspect to them. When we kick up the specific capacity in summer sheet our heating load drops, dramatically in some projects (depending on climate and quantity of glass to the south. We do a lot of studies to prevent overheating in summer... A lot of the reason we do this stems from us both having experience doing more passive-solar projects in EU and here in the states. And in studying up on the central European PHs, a lot of them utilize similar properties. For commercial projects especially, thermal mass appears to be highly beneficial when combined w/ night cooling. Even in the colder regions of CH/AT - I know of several PH projects that are almost completely glazed to the south.
For a project we've got underway here in Seattle, we're shooting for 3.5kBTU/ft2a specific space heating demand w/ all wood construction. When I mess around with specific heat capacities (BTU/°F per ft² TFA), here are resulting specific space heating demands...
Seattle, WA
11: 3.57kBTU/ft2a
23: 3.34kBTU/ft2a
36: 3.28kBTU/ft2a
We don't get much of a benefit up here thanks to the lack of insolation (which lead to awesome levels of seasonal affective disorder - fodder for another blog post, perhaps). Here are results when I 'migrate' the project to other locations:
Omaha, NE
11: 4.99kBTU/ft2a
23: 4.70kBTU/ft2a
36: 4.62kBTU/ft2a
Denver, CO
11: 2.98kBTU/ft2a
23: 2.52kBTU/ft2a
36: 2.40kBTU/ft2a
Boston, MA
11: 4.03kBTU/ft2a
23: 3.71kBTU/ft2a
36: 3.62kBTU/ft2a
Calgary, AB
11: 7.55kBTU/ft2a
23: 7.13kBTU/ft2a
36: 7.03kBTU/ft2a
The shift isn't enormous, but it's enough that if we're close, we will ponder use of a concrete topping to push us over the edge. There are other tricks as well - but if we're already utilizing concrete, we figure might as well make the most of it.
Martin,
For perspective I think you should consider the following mathematical formulations for your article:
Modeled Design x ‘the j factor’ = Real World Results
Whereas:
By definition ‘the j factor’ ≠ 0.
On early attempts ‘the j factor’ is at its greatest and always > 1.
‘The j factor’ can be ≤ 1, but only for projects executed in Japan or European countries (excluding Greece, Italy and Portugal.)
Furthermore:
Total Environmental Impact = Real World Results + ∑ƒ(a) + ∑ƒ(b) +∑ƒ´ (c) + ∑ƒ(d) + ƒ(e) + π
Whereas:
∑ƒ(a) is the embodied energy of the project.
This convoluted algorithm will be subject to subjective distortions by powerful lobbying efforts if it is even accepted as a worthwhile exercise.
Whereas:
∑ƒ(b) is the ‘Chesnut Algorithm’.
The ‘Chesnut algorithm’ accounts for ongoing operational impacts extraneous of electric or gas utility usage. Examples include transportation to and from site, toilet flushing, lifecycle maintenance, junk mail delivered to site, garbage collection, snow plowing, etc.
Whereas:
∑ƒ´ (c) is the ‘Riversong Algorithm’.
The subtle ‘Riversong Algorithm’ accounts for but is not limited to –
• The ∆ between the ‘Modeled Design’ and an environmentally regenerative ‘What Could Have Been’ such as planting climate adaptive plant communities or allowing natural succession to take its course unimpeded. (The inclusion of additional impacts for unrealized potential benefits is justified if you consider the practice of ‘shorting’ stocks.)
• The depreciation of the environment due to the entrenchment of the Enlightenment worldview via physical manifestation as built space.
• The lost opportunity for compounding capital investments to work towards the environment’s benefit due to excessive investment in air tightness measures.
Whereas:
∑ƒ(d) is the ‘ajbuilder Algorithm’.
This algorithm statistically forecasts the excessive stress over the environment’s ability to regenerate itself caused by all future human beings and their prodigy whose lineage can be traced back to all human conceptions occurring at the project over its lifecycle.
Whereas:
ƒ(e) is the ‘Cathedral Ceiling Penalty Formula’.
For projects employing a cathedral ceiling this formulation accounts for any additional energy losses - as compared to flat ceilings with a healthy heap of loose fill cellulose and properly sized energy heals - that less sophisticated energy modeling programs can’t capture and calculates the lost opportunity for compounding capital investments to work towards the environment’s benefit by multiplying the square footage of the roof’s area in plan by ‘Brook’s constant’.
. . . and π is thrown in for good measure.
In my estimation, the ‘Modeled Design’ accounts for less than 7% of the ‘Total Environmental Impact’ ; )
J,
OK, we're getting closer to modeling the real environmental effects of new construction! I like it!
Total Environmental Impact = Real World Results + ∑ƒ(a) + ∑ƒ(b) +∑ƒ´ (c) + ∑ƒ(d) + ƒ(e) + π
Can you please develop an Excel spreadsheet that implements your formula?
The only problem is the time it takes to do the data entry... Two months per iteration...
Mike, thanks for the reply. I usually just go with the annual method because it is easier to track what is going on but thanks to you post I learned that the monthly method actually takes the thermal mass into account. Every day we learn something new :-). I will definitely investigate this further and I have a feeling this might helps. when you determine the specific heat capacity do you go with the formulas given in the PHPP handbook or do you have a more sophisticated way? Also you seem to have build some houses using the PHPP. Can you share any results of accuracy in you climate zone?
@Martin: I guess what I would like to see in the article is limitations of the software. What does it good and what does it neglect. Also I do not know if any of the software mentioned by you uses dynamic analysis. I know the PHPP has been developed as a simplified tool (as J mentioned earlier) to omit the rather complicated use of a dynamic simulation and I guess this could potentially be a problem for different climate zones.
Philipp,
We typically just use the formulaes in PHPP - though there are a few projects where calculating it out might be advantageous (e.g. if we had concrete walls but wood floor framing). Annual method is usually easier/has better space heating demand results, however it's tough to get away from lots of glass (for us, anyway) - so our Qf/Ql (annual sheet - ratio free heat to losses) usually ends up being just a little north of 0.7. This means we need to use the annual method regardless. Wrote about that and some other PH optimization tips here: http://bruteforcecollaborative.com/wordpress/2011/08/26/10-tricks-for-optimizing-a-passivhaus/
Nothing built as of yet, though we've run several projects in PHPP now. Our biggest project is on hold but hopefully returns, as they'd have the potential for first CO2 neutral housing in NW.
To reinforce Philipp's comment on limitations of the software and my earlier comment on limitations of the input data used, I did a quick check on the impact of assumed air leakage on my own house. The use of a leakage rate based on 5 ACH @50 Pa (what I believe Energy Star has required for Zone 6), rather than the final 0.8 blower door measured, would add nearly a ton to the peak heat load. Furthermore, a run done by someone else on this house, using canned software, reported a 10,000 BTU/hr heat loss for an assumed but not present fireplace. Together, those two assumptions alone would have doubled the calculated heat loss at design conditions. Along with other smaller differences resulting from use of "closest" values in drop-down menus, the bottom line heat loss from that other calculation was even further from what I had obtained by spreadsheet.
@ J Chesnut
Perhaps this isn't the place to ask, but I'm sure I'm not the only one who'd like to learn more about
" ... all human conceptions occurring at the project over its lifecycle."
Martin,
I'm running into another problem. I may have been premature to publish my work.
My problem is resolving the units, some of which haven't been conceived of yet.
(I like the sound of 'a Holladay' so if I need to coin a new unit I'll go with that. The more 'Holladay's the better.)
I don't mean to make light of the perceived complexity of the PHPP. I think in the long run when even the US will be forced to implement meaningful energy consumption regulations only energy models like the PHPP that can accurately estimate actual loads inclusive of plug loads will be productive to use. So I appreciate the detail in which Philipp and Mike are conversing.
That being said I wonder if rules of thumb based on climate zones (the Building Science approach with more attention to leveraging solar heat gains for free heat in cold climates) developed with the help of energy modeling could demonstrate insignificant 'standard of deviation' of energy efficiency as compared to going through the process of modeling every individual project per the Passivhaus model. I think of course the optimal solution is somewhere in between.
@ JoeW,
Let's just say thats not a difficult calculation at my house ; )
J, LFMFO... Truly the best and most informed whole world look at this idea of a greener future via mathematical analysis. You do however need to change the plus signs to multipliers.. Then if you take the Aj factor down to 0(new births)/X(existing humans), the end result is a perfect natural world devoid of any further need for the formula, its use and of course the end of GBA, Martin's income and all these interesting posts!
Maybe we should stick to loving problems like the end of fossil fuel, rising seas, and how to build a net zero home affordable for those that make minimum wage....
Great thread, love your formula J.
AJ,
Good point.
Here is the correction according to your peer review-
Total Environmental Impact = Real World Results + ∑ƒ(d)dx [ ∑ƒ(a) + ∑ƒ(b) +∑ƒ´ (c) + ƒ(e) ] + π
J Chesnut,
The formula reads deeply into it's origins. This is truly building shelter with personality.
However... I do not consider over investment in air sealing an issue. There is no such thing. Proper air sealing and a faulty ventilation system will enhance ∑ƒ(d)dx.
Albert,
I was poking fun at our old friend Robert Riversong. If you haven't read the old GBA threads might be fun for you search for the heated debates on the appropriate air tightness threshold that occurred a year or two ago.
Joe Lstiburek's recent paper is the most elaborate treatment of the subject I've come across but Mike Eliason's summary of the mutliple European standards was illuminating also.
Until I know more about the mathematics behind the energy savings proposed by PHPP models or see an authoritative study based on demonstrated field results I don't know if there is a significant difference between .6 and 3 ACH50.
I tend to think it more prudent to error towards tighter but I don't necessarily want to use tapes and caulks moreso than need be because these are products that could prove to deteriorate over time. Sourcing these products also directs capital to the big chemical industries and I don't have faith in the regulations or motives of big corporations.
J,
No worries, I was not being serious at all. I got all of your references to the various personalities and enjoyed recalling John Brooks's rant as much as my own rants (far too many of my own to retain any credibility!) I was simply having fun at my own expense... Being a tape and other air sealing stuff guy... Kind of making fun of my own greed at air tightness, which is kind of a ridiculous concept.
While working together building higher quality shelter is serious business, your well thought out and well written "formula" was the funniest thing I've read in a while. Seriously: thanks for posting it!
Martin,
Not to bug you, but I'm wondering if you have a publishing date for your article yet?
Also, will you be covering the alternative of hiring specialists who are already familiar with the software; specifically, the accuracy, thoroughness, and the costs that can be expected?
Thanks for all your contributions to progress in thought and method.
David,
The article will appear at the end of 2012, so it's many months away.
Here's a thumbnail version of what I know now: the programs that are fast and easy to learn aren't particularly accurate, especially for superinsulated homes. The programs that are accurate for superinsulated homes take a long time to master and involve a lot of data entry.
In short, if you know a consultant who is familiar with PHPP and is willing to run the software for you, it probably makes more sense to hire the consultant than it does to learn how to use the software.
Thanks Martin. This was my suspicion, and your affirmation is very helpful. I will still look forward to your article!
Hi Martin
I have used HOT2000 for the past 3 years as part of the Canadian EcoEnergy Home Retrofit Program.
Data entry time is of course variable, depending on the house being modeled. A simple 1500sq ft bungalow might take about 30 minutes, while cathedral ceilings, multiple foundation types, variable wall assembly R values etc will add to this time as each building component has it's own entry.
Find the software is fairly easy to use. Accuracy for building energy loads is not bad, however I find the heat loads typically come in about 10 to 15 percent higher than what the client actually experiences (based on fuel bills). One thing I don't care for (at least in the version of software I use) is that "base loads" such as DHW, laundry and lighting are fixed. I think they do this to even out the results among samples. There may be a version that allows you to change these default inputs.
Joe,
Thanks. I appreciate the information you shared.
Joe,
Re: HOT2000
You can vary the base loads in General Mode. This won't affect the EG rating, but the change will be realized in the energy values if you run the Full Report.
I agree with you that this should affect the EG rating, especially for New Homes where efficient appliances are installed.
Martin, has your article been published yet, or do you know when it will be published?
The two I see most frequently didn'teven make your list.
DOE2 & BEopt.nrel
Are these user friendly?
Richard,
Both DOE2 and BEopt are on the list, and will be mentioned in the article (unless my words are edited by higher-ups).
DOE2 is not particularly user-friendly, but it is the "engine" used behind the curtain for BEopt.
BEopt is fairly user-friendly, but has its own quirks. For more information on BEopt, see BEopt Software Has Been Released to the Public.
I think a very good free program for Home Energy Efficient Design is HEED produced by the UCLA Architecture department. It is very user friendly. Search on the internet for HEED software.
Rick,
Thanks. The article was published a while ago. Here is a link:
http://www.finehomebuilding.com/tool-guide/articles/energy-modeling-software-is-it-worth-your-time-and-money.aspx?ac=fp
The article mentions HEED.