Basic stuff - Infiltration

17 posts / 0 new
Last post

In building modeling programs one always has to provide precise values for
infiltration. In the real world, I will know one of two things: almost
nothing (The building appears to be kinda leaky with old windows), or I will
have a blower door test done at a specific pressure. How do I convert
subjective ("kinda leaky") or objective (Blower door test) leakage into
numbers that make sense in the program? Is there a guide one can use?

--Lawrence

lawrence Lile's picture
Offline
Joined: 2011-09-30
Reputation: 0

Hi Lawrence,

My copy of ASHRAE Principles of HVAC includes a table (5-1) excerpted
from ASHRAE Fundamentals, 2001 (Table 7, Ch 28). This table provides
air change rates as a function of subjective envelope airtightness
("tight" / "medium" / "loose") and as a function of the outdoor design
temperature. Upon reviewing the referenced Fundamentals chapter, I
learned this table is built from research surveying residential homes of
various vintages, so it helps to know that these are "tight" to "loose"
residential constructions.

In any case, I've used and cited this resource before when modeling
infiltration and calculating sizing loads for non-residential projects
as well. I've searched, but have yet to come up with an equivalent
table based on surveying and measuring commercial constructions from a
subjective/objective standpoint... That might be handier, but in the
meantime this is a good tool for "converting" your subjective
observations into the right ballpark.

~Nick

NICK CATON, P.E.

Nick-Caton's picture
Offline
Joined: 2011-09-30
Reputation: 805

Good ol ASHRAE Fundamentals! Why didn't I think of looking there? Thanks!

lawrence Lile's picture
Offline
Joined: 2011-09-30
Reputation: 0

An HTML attachment was scrubbed...
URL:

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

My own personal opinion is that ASHRAE Fundamentals has not rescued us here.

As Nick mentioned, the data presented there is for residential houses and relies on incredibly vague and subjective judgment calls by the modeler.

Furthermore, the values presented are in the unit's of air changes per hour. To me, this is a terrible way to use the data. Your entries for infiltration then rely on the volume of the room to determine the amount of infiltration. The volume of the room (zone is probably a better term) has NOTHING to do with the infiltration. What if I have a gym that's 150'x100'x50' tall, but it only has 10' wide by 50' tall of exterior wall??? If I use the ASHRAE method and rely on air changes per hour, the zone will be modeled with a HUGE GIGANTIC REALLY REALLY LARGE amount of infiltration.

Yet this seems to be the only recourse we have that is grounded in any sort of defendable data.

I have looked and looked for a reliable report or other source for real world commercial/institutional construction infiltration values to no avail. It would be so incredibly useful.

I have, over the course of my energy modeling career, developed a set of seemingly practical infiltration values to use, using the units of "CFM per sq ft of gross external wall area" available in Trane Trace (I don't believe these units are an option in eQuest). These values were developed by taking a number of buildings with no infiltration and arbitrarily adding infiltration in, until I get a reasonable utility consumption value. Hardly scientific and no way could I defend these values if they came under scrutiny, other than to say "well, you got a better idea????"

Some may say "Eh, who cares about infiltration anyway?". Well, it makes a bigger difference then you'd think.

A novice user who relies on the ASHRAE air changes per hour is likely significantly oversizing their cooling equipment in large rooms (ie conference, assembly, gyms, etc) which is exactly where you don't want to be oversizing cooling equipment.

Think about the components of a heating load - envelope conduction losses, taking in cold OA...and infiltration. Envelope losses are generally small, the design community likes to temper their OA (rightfully so), so where is a major component of the heating (consumption) load coming from? Infiltration really adds up.

How do you justify replacing leaky, wood sash windows? How bout making a switch to spray foam insulation? How about modeling door seals?

I've rambled enough. Point being, we all are forced to use arbitrary numbers for something that is a significant component of both equipment sizing and energy modeling and it just makes me mad and embarrassed when I have to explain to a client or colleague "well those are really important, but completely imaginary, numbers..."

Garasaki's picture
Offline
Joined: 2011-02-16
Reputation: 1

Hi John - thanks for your thoughts!

I was relating some of the same concerns with others off-list ... I'm am
no infiltration-auditing expert (though some lurk among us here on the
lists ^_~), but one thing I can say based on my past attempts to build a
better mousetrap regarding infiltration is that where ASHRAE may be
generally vague on the topic - they are actually doing a lot to say (if
not always directly) that whole building infiltration is a very
difficult thing to quantify at best (sorry for excessive use of
parentheticals (I mean it!)...).

The best guidance imaginable that would still be practical in day-to-day
use would have to rely in some part on subjective observations (guesses)
regarding envelope constructions. Two brick walls of certain
grout/masonry ratios weathered for the same period in the same climate
may still have different leakiness because the two masons used slightly
different grout mixes... What I'm getting at is you couldn't
realistically construct a table that covered every variable, and many
variables are not "knowable."

That's not to say the residential ACH table isn't useful for subjective
estimations, nor that research couldn't be undertaken to raise the bar a
notch. As John is alluding, a table providing representative commercial
envelope constructions (with accompanying illustrations!) and/or layer
combinations could be undertaken that would provide infiltration
performance as a function of time. Values could be given for new
construction, and after weathering for 1/5/10 years. While new
constructions/layers could be assessed in a controlled environment,
initial research on aged constructions would need to be done sampling
within a single climate zone. Separate/concurrent research could
explore determining multipliers on the weathering effects based on
varying climate and geography... All things being equal, a beachside
wall built in Miami, FL with lots of sun/salt/torrential rain seasons
and the occasional hurricane will weather differently over a decade than
the same wall in a milder climate. The net result of such research
could ultimately produce some really helpful tools in better assessing
existing and new constructions for a variety of industries and purposes
(energy modeling included).

Considering the growing presence and pressing need for better tools in
the world of energy modeling, I would put forward prime candidates for
whole construction assemblies would be ASHRAE 90.1 baseline
constructions as defined in Appendix A.

For all I know, such research may be underway or completed years ago -
my ear is not quite so close to the ground with the academic world...
can anyone comment?

To another point you brought up - eQuest is quite capable of using your
personally developed CFM/ft2 values - in the wizards even! In detailed
mode you'll find there are inputs for more involved estimations as well
if you wish to pursue other methods:

NICK CATON, P.E.

Nick-Caton's picture
Offline
Joined: 2011-09-30
Reputation: 805

Here is a file that might help. The infiltration values are given at a pressure of about 0.3 in w.c. so they may have to be reduced, but it is another guide.

Kelsey Van Tassel

Kelsey VanTassel's picture
Joined: 2011-09-30
Reputation: 0

Thanks for the response Nick.

As I recall from a recent foray into this subject in eQuest (in detailed mode), the cfm/sq ft entry is based on floor area.

It would be logical that if cfm/sq ft of exterior wall is an option in wizard mode, it would also be an option in detailed mode - I just haven't dug that far yet.

Garasaki's picture
Offline
Joined: 2011-02-16
Reputation: 1

John,

The infiltration as a function of exterior gross wall area is only available
in the wizard. I don't believe DOE2.2 is capable of having inputs that
relate to the sum of a parameter of it's child components. The DOE2.2 BDL
Functions do not have any references to child components (i.e. a Wall can
reference a parameter of its parent Space, but a Space cannot reference a
parameter of its child Wall).

Regarding Lawrence's initial question about converting a known tested value
to a value usable within eQUEST, the PNNL Report 18898, Infiltration
Modeling Guidelines for Commercial Building Energy Analysis (
http://www.pnl.gov/main/publications/external/technical_reports/PNNL-18898.pdf)
contains
formula for converting infiltration from a test case to an actual design
case. Based on the example presented in the document, a conversion factor of
0.112 can be derived.

So a tested air leakage of 0.40 CFM/SF at 0.30 in. w.g. would be modeled at
0.045 CFM/SF. This value is modeled at 100% when building fan system is off
and 25% when the building fan system is on.

Granted, this may be an oversimplification for eQUEST, as the document was
written for EnergyPlus which contains a wind-driven infiltration model, but
it seems to be a good starting point at least if you have test information
available.

*Robby Oylear, LEED**?** AP BD+C*

Robby Oylear's picture
Offline
Joined: 2011-09-30
Reputation: 202

Robby - awesome post, thanks for linking that paper!

As an extra heads up/thought: depending on which infiltration method is
selected, I do believe eQuest will also calculate hourly infiltration
rates working from windspeed in the weather file. Details would be in
the help files... I can't recall if this happens with the default
method, or if there is a single "default" method (it might vary based on
how you define infiltration at the wizard level?)...

~Nick

NICK CATON, P.E.

Nick-Caton's picture
Offline
Joined: 2011-09-30
Reputation: 805

Here is a recent response that I posted on Bldg-Sim.

Gaurav Mehta, LEED? AP BD+C

Mehta, Gaurav's picture
Offline
Joined: 2011-09-30
Reputation: 0
Robby Oylear's picture
Offline
Joined: 2011-09-30
Reputation: 202

Classification: UNCLASSIFIED
Caveats: NONE

Another infiltration issue to consider is the infiltration schedule. Most of
my designs include building pressurization during occupied hours. I make an
infiltration schedule and set the infiltration to 10% during occupied hours.

The biggest advantage is that all the outside air goes through an energy
recovery unit and therefore is tempered.

"Is Freedom a small price to pay to stop Global Warming?"

John Eurek PE, LEED AP

Jeurek's picture
Offline
Joined: 2010-10-07
Reputation: 0

Infiltration in non-residential buildings is for sure a big concern.

For example, a common assumption is that building pressurization keeps infiltration out (fortunately, John you keep some minimum rate). And it is very tempting to make this assumption of zero infiltration. However, this is not always the case, if we believe Kaplan and Caner ("Guidelines for Energy Simulation of Commercial Buildings", Prepared for Bonneville Power Administration, March 1992, p41):

"The assumption of zero infiltration during hours of HVAC system operation rests on the assumption that system operation will result in building pressurization. However, this presumes a well-designed, well-balanced, and properly operated air distribution system. It also presumes the absence of other infiltration-related effects such as tall building stack-effect, a high frequency of occupant or customer entry and egress, normally-open loading docks, and so forth."

A major factor is the difficulty to obtain and maintain uniform pressurization.

There is certainly a need for a lot more research on the topic of infiltration in non-residential buildings. A closely related topic would relate to the science of air distribution: development of guidelines for uniform pressurization.

At this point, let's make ourselves the favor not to forget this sensitivity analysis on infiltration rates to render the model more robust.

_______________
Demba NDIAYE

Demba Ndiaye's picture
Offline
Joined: 2011-09-30
Reputation: 200
Bobby Sy's picture
Offline
Joined: 2011-09-30
Reputation: 0

Hi Robert,

I can only speculate towards the circumstances of the DOE 'tests' you are referring to, but a contributing factor is likely that statements such as: "XXX CFM/SF reduction of infiltration will result in YYY% energy consumption savings" are inherently case-specific. Any set of numbers may be true for a specific model of a specific size and type of building in a specific climate (the variables go on and on), but as soon as you change any of number of contributing variables you open the door for the impact of infiltration to change.

This is analogous to the application of the Passivhaus and similar standards to buildings in completely different building types or climates than those the standards are built around. Someone trying to design/build a Passivhaus home in the Arizona is to some extent missing the point, for example, and will hopefully realize the annual 15kWh/m^2 heating/cooling mandate is itself intended for a specific part of the world.

Again, this is as speculative as it must be, but if you're implying DOE models calculate infiltration loads differently than something else, you'll need to first provide more specifics regarding what the 3% number is based on, and ensure the contributing variables to infiltration impact are being held constant against the measurements/model producing the 3% figure... I think you'll find eQuest/DOE2 models are capable of determining and modeling infiltration in a variety of calculation methods (among which you can mandate a specific ACH or CFM/SF of wall - both convertible from 10m? /h.m? of wall area), and can also be scheduled to impose any desired hourly level of infiltration upon a model's zones if that's desired.

Best of luck,

~Nick

NICK CATON, P.E.

Nick-Caton's picture
Offline
Joined: 2011-09-30
Reputation: 805

I am modeling an ice storage system that we plan to charge overnight
with a couple chillers. In eQUEST, this seems to work out alright
except the thermal storage is only discharged if the chillers cannot
meet the cooling load. We plan to use the thermal storage as the first
stage of cooling. Previous posts refer to the load management and
equipment controls for this, but it looks to me like you can only
specify chillers here. I don't see anything for thermal storage. Am I
missing something here?
Thanks,
Tim

Tim JohnsonCTA : PE, LEED AP

Tim Johnson's picture
Offline
Joined: 2011-09-30
Reputation: 1