Infiltration issues whith more than one shell

13 posts / 0 new
Last post

Hi all,

I'm having an issue with infiltration and wondering if anyone else has noticed this. When I'm in wizard mode, I build a very simple rectangular shell and I change the infiltration rate for perimeter zones to 0.05cfm/ft2. I leave the core zone infiltration rate to the eQuest default value (0.001cfm/ft2). From my understanding, the perimeter zone entered value in wizard mode is based on exterior gross wall area. When I move on to detail mode, eQuest changes the value to an infiltration value per floor area instead of exterior wall area. The values so far make sense and I can do the math and get the same numbers eQuest calculates in detail mode. No problems here.

The problem however happens when I create different shells. For example, let's say I have the same rectangular building as mentioned above, but I split this up into two different shells next to each other (two square shells), therefore one of the walls of each square shell is an interior wall, which means I should not see an infiltration of 0.05 cfm/ft2 of the gross wall area. However when I finish building these two shells in detail mode (which would be identical to one rectangular shell), the infiltration value is not accurate anymore. Instead of the same numbers as my first model (one rectangular shell), the second model calculates an exterior wall infiltration value for the interior walls that are budded up against each other.

Has anyone come across this? Is this a quirk with eQuest and just needs to be changed manually in detail mode, or am I doing something wrong here?

Thanks for the help.

Cheers,

Menush Akbari
BEng, PEng, BEMP, CMVP
Senior Energy Engineer
Mission Green Buildings
(Mission Green Buildings is the trading name of Mission Green Limited)

t. 250 777 3380
makbari at missiongreenbuildings.com
http://missiongreenbuildings.com

[MGB 5 Years 2]

Menush's picture
Offline
Joined: 2016-03-02
Reputation: 0

An HTML attachment was scrubbed...
URL:

Bruce Easterbrook's picture
Joined: 2011-09-30
Reputation: 0

I don't think Bruce's comment is germaine to the problem posed by Menush, since Menush is
asking about infiltration and not heat transfer through interior walls.? As far as I know
from DOE-2 (always have to add that caveat because I don't work much in eQUEST),
infiltration rates are normalized per floor area, rather than external wall area, mainly
because that's how the infiltration models have been developed, i.e., effective leakage
areas have always been defined as per floor area.? If you don't think that's right, you
can always adjust the input values by the ratio of the external wall area to the floor
area, as you've described, but then you're on your own in coming up with the correlation,

Joe

Joe Huang
White Box Technologies, Inc.
346 Rheem Blvd., Suite 205A
Moraga CA 94556
yjhuang at whiteboxtechnologies.com
http://weather.whiteboxtechnologies.com for simulation-ready weather data
(o) (925)388-0265
(c) (510)928-2683
"building energy simulations at your fingertips"

Joe Huang's picture
Offline
Joined: 2011-09-30
Reputation: 406

An HTML attachment was scrubbed...
URL:

Bruce Easterbrook's picture
Joined: 2011-09-30
Reputation: 0

Bruce,

I have to say that DOE-2 normalizing the Leakage-Fraction by the floor area is not a
quirk, but a result of the infiltration studies done to date that are predominantly,
almost exclusively, done on single-family residences.? I'm thinking here of the work by
Max Sherman and David Grimsrud at LBNL that produced the Sherman-Grimsrud Model, the only
infiltration model I know of that's physically based, i.e., the model related the
infiltration to the temperature difference and wind pressure, coupled with the leakage
area of the building.? For single-family residences, that leakage area is best scaled by
the floor area, since a large part of the leakage area is in the ceiling. Therefore, I
don't agree that leakage scales better with the wall area than the floor area.? On an
intuitive basis, it should probably scale by the exterior surface area, i.e., walls +
windows + ceiling, but I have yet to see any study that quantified the amount of leakage
through each of these components.? As we move to multi-family or commercial buildings, the
surface/volume ratio will clearly dominate and for interior floors with only the wall and
windows exposed to the outside, yes, the infiltration probably does scale with the exposed
wall area. However, there's a dearth of information on the normalized leakage of walls.

Joe

Joe Huang
White Box Technologies, Inc.
346 Rheem Blvd., Suite 205A
Moraga CA 94556
yjhuang at whiteboxtechnologies.com
http://weather.whiteboxtechnologies.com for simulation-ready weather data
(o) (925)388-0265
(c) (510)928-2683
"building energy simulations at your fingertips"

Joe Huang's picture
Offline
Joined: 2011-09-30
Reputation: 406

All;
Confused as to how Bruce?s answer isn?t directly applicable. What he detailed in the first instance is directly in line with my experiences. eQuest wizard modes doesn?t realize that two abutting/touching shells should have interior walls. If a space wall touches a shell wall, an exterior wall is created.

- If a zone boarders a shell, eQuest creates an exterior wall.

- If a zone does not touch a shell edge, eQuest makes an adiabatic interior wall.

- If two shells are touching and each have zones touching the common wall, two exterior walls are created; one in each zone. Wizard eQuest doesn?t distinguish if two shells are 0 feet apart/touching or if they are 25 feet apart. (example described below)
If eQuest uses exterior wall area to calculate a floor area normalized infiltration rate, and there are ?extra? exterior walls which are actually interior walls where shells touch, then the infiltration rate will be calculated as though the two shells were never touching. This will increase the calculated infiltration rate since these are not actually interior walls.

Bruce?s solution only neglected to mention that the infiltration rate will either need to be corrected for this condition, or ignored/neglected. The second option is not to say infiltration is not important ? it is important ? but requires modeler judgement and some math. The length/area of the touching walls is also a factor to consider. If project is a large industrial setting as described in thread, yes, probably should check/revise. If a small connecting corridor, probably minor and can be ignored.

Did a quick (very simplified ) example. All examples have: 10ft high. No plenum. Set Perimeter infiltration to 0.05 CFM/sf. 0.0 core infiltration.
- Made two shells, 50ft x 50ft each. 2 spaces per shell, spliting shell equally. Placed the shells touching. The calculated infiltration is 0.0400 cfm/ft
- Made two shells, 50ft x 50ft each. 2 spaces per shell, spliting shell equally. Placed the shells 50 feet apart. The calculated infiltration is 0.0400 cfm/ft
- Made one shell, 100ft x 50ft each. 4 spaces per shell, spliting vertically,such that the zones are the same area/dimensions as the previous examples. The calculated infiltration is 0.0400 cfm/ft in the end cap spaces, and 0.0200 CFM/sf in the middle spaces. This is due to less exterior wall area (as expected).
- Made two shells, 50ft x 50ft each. 2 spaces per shell, same spaces as above. But ? offset the edge of the shells when drawing in wizard mode such that the edge of the spaces were not the edge of the shell(s) (refer to image). Placed the shells such that the two zones were touching. The calculated infiltration is 0.0400 cfm/ft in the end cap spaces, and 0.0200 CFM/sf in the middle spaces; matching as if done in a single shell.

The last example allows the user to change the wall from adiabatic to a heat transfer surface easily. Can also modify the infiltration rate as needed & detailed ablve. PNNL has published several reports on infilatration rates which can be references as well for good information.

Let me (and all of us) know if this seems at odd with previous experiences or overlooked a detail. Thanks in advance.

Thanks,
DARIC R ADAIR PE, CEM, BEMP
Mechanical Engineer, Energy Analyst

Henderson Engineers
Tel (913) 742-5530
daric.adair at hendersonengineers.com

Licensed in KS

Daric Adair's picture
Offline
Joined: 2017-10-21
Reputation: 0

An HTML attachment was scrubbed...
URL:

Bruce Easterbrook's picture
Joined: 2011-09-30
Reputation: 0

An HTML attachment was scrubbed...
URL:

Bruce Easterbrook's picture
Joined: 2011-09-30
Reputation: 0

Need to make a correction (& add clarification) to my answer below.
Clarified the square-footages: sf-ExtWall vs sf-floor.
sf-ExtWall = how infiltration is entered in eQuest wizard
sf-floor = the value in eQuest/DOE2 simulations for infiltration.

Sorry for error.

Thanks,
DARIC R ADAIR PE, CEM, BEMP
Mechanical Engineer, Energy Analyst

Henderson Engineers
Tel (913) 742-5530
daric.adair at hendersonengineers.com

Licensed in KS

Daric Adair's picture
Offline
Joined: 2017-10-21
Reputation: 0

The point I'm trying to make is that DOE-2 and presumably eQUEST derives the infiltration
rate from user input of the EFFECTIVE-LEAKAGE-FRACTION for the SPACE that is expressed in
CFM/ft2 of floor area. This infiltration rate is not affected by the amount of exterior
walls in the spaces or their proximity to each other.? In your examples, aren't the
infiltration rates for the building just the sum of the infiltration rates for the zones
(presumably perimeter and core) adjusted by their floor area ?

Joe

Joe Huang
White Box Technologies, Inc.
346 Rheem Blvd., Suite 205A
Moraga CA 94556
yjhuang at whiteboxtechnologies.com
http://weather.whiteboxtechnologies.com for simulation-ready weather data
(o) (925)388-0265
(c) (510)928-2683
"building energy simulations at your fingertips"

Joe Huang's picture
Offline
Joined: 2011-09-30
Reputation: 406

-eQuest Digest just came thru ? receiving emails out of order. Sorry in advance.

The infiltration rates from the example are from the eQuest Wizard input. The wizard input is expressed in a CFM/SF of exterior wall area. When the model is generated out of wizard and into the detailed/spreadsheet interface these infiltration rates are calculated/converted based on:

SpaceInfiltration [CFM/sf-floor] = (infiltration rate [CFM/sf-exterior wall area] * exterior wall area [sf]) / (Space Area [sf-space])
Note: INF-METHOD= AIR-CHANGE for this conversion/calculation.

I checked these for the example I sent out, and they aligned with the screenshots.

Joe - you are entirely correct that once in eQuest detailed edit mode, the Infiltration is expressed normalized to Space Area. Once past Wizard it is not possible to directly enter an infiltration in (CFM/sf-ExteriorWall area) without either doing some math outside of eQuest to normailze to the Space Area or using a UserExpression to do the same. (and very time consuming too. Learned the hard way -once).
The ?EFFECTIVE-LEAKAGE-FRACTION? input you reference is part of the Sherman-Grimsrud method; yes? The screen shots of the infiltration rates are of the ?INF-FLOW/AREA? input, which is default out of eQuest I believe. This may be part of the confusion. Again, getting emails out of order.

Is there a report which contains a ?whole building infiltration CFM?? I can find ACH by space, but nothing for the total building.

PNNL, NIST, and ASHRAE Journal have put out several reports in the last few years on the topic of energy models and infiltration. They use SF of exterior envelope (not just wall or roof) as a metric in several reports. ASHRAE 90.1-2013 Appendix G now has requirements on how Infiltration must be included in an energy model. One of the reports ??recommends a rate of 1.8 CFM/sf (0.3 inwc) of exterior above grade envelope surface area.? (whoa, that?s higher than I?m used to seeing!)

To go back to the original question from Menush. The issue was two shells touching and that exterior walls are created for each shell, not an interior wall between the two shells (& respective zones. The question (as I understand it) has nothing to do with the realities of infiltration or methods there-of. The issue was that eQuest Wizard does not know that when two horizontal shells are touching that this is most likely an interior wall between spaces. And thus it creates ?extra? exterior walls which inflate the infiltration rate. This is accurate and the only point I was initially addressing (and I think Bruce was addressing similarly). The methods on how to determine a proper infiltration rate for the impacted Spaces has sprung from this original question.

Menush ? does this answer your question(s) on the shell/space/exterior walls coming out of wizard?

Thanks,
DARIC R ADAIR PE, CEM, BEMP
Mechanical Engineer, Energy Analyst

Henderson Engineers
Tel (913) 742-5530
daric.adair at hendersonengineers.com

Licensed in KS

Daric Adair's picture
Offline
Joined: 2017-10-21
Reputation: 0

To round back and answer Menush?s query directly ? you aren?t doing anything wrong!

eQuest wizards do not take into account adjoining shells when determining whether a zone is ?core? vs. ?perimeter? for the purposes of assigning the static-but-tiny ?core? infiltration rate of .001 cfm/ft2 vs. those derived from other infiltration rate inputs for each shell.

For many models, this isn?t of much consequence, but I do recommend for any model where infiltration rates are a focus to give careful review post-wizards for where infiltration is applied. The wizards? default behavior is sensible enough for most cases, but sometimes requires correction beyond the adjoining shell ?quirk,? such as when core zones include clerestories or similar features subjecting them to more infiltration than a typical core zone. You may also wish to ensure core/perimeter HVAC zones who are in reality only separated by ?air walls? (i.e. open office with exposed perimeter) distribute the infiltration loads instead of isolating to just the perimeter.

If you?re thinking ahead, you can ?trick? the wizards into treating shell-perimeter-but-adjoining-another-shell zones as ?core? by pull the associated vertices a small bit off of the associated shell footprint? but that?s some next level shenanigans and you won?t be avoiding giving the whole model a pass later? I think just correcting core vs. other infiltration rates is probably most time-efficiently done immediately post-wizards.

~Nick

[cid:image001.png at 01D3B638.1A8B0AA0]
Nick Caton, P.E., BEMP
Senior Energy Engineer
Regional Energy Engineering Manager
Energy and Sustainability Services
Schneider Electric

D 913.564.6361
M 785.410.3317
F 913.564.6380
E nicholas.caton at schneider-electric.com

15200 Santa Fe Trail Drive
Suite 204
Lenexa, KS 66219
United States

[cid:image002.png at 01D3B638.1A8B0AA0]

Nicholas.Caton at schneider-electric.com's picture
Joined: 2016-02-26
Reputation: 0

Thanks everyone, that?s very helpful! I?ll start checking post-wizard, especially when I have a complex building with many shells and/or core/perimeter that are only separated by ?air walls?.

Cheers,
Menush

Menush's picture
Offline
Joined: 2016-03-02
Reputation: 0