[Bldg-sim] Internal mass modeling guidance in E+?

[Graham Carter & Megan Lyall]

I like some of Doug's points around bounding an assumption that is uncertain rather than arguing the exact assumption which if you are really lucky may be close to reality.

Books are probably a reasonably good insulator as they are cellulose and would have fine air gaps around them and possibly between the pages that act as an insulant so it gets back to the comment on 'fast' mass.  I think the CIBSE methods define something called thermal admittance as the ratio of conductivity to specific heat capacity.  What that means is a material can be heavy, but if it has a low conductivity or maybe finished with an insulative layer (think rubber backed carpet on concrete) then the ability for a material to absorb and release heat is diminished - the mass is no longer in close thermal contact.

I had the pleasure (or displeasure in the early stages) of diagnosing and rectifying a thermal comfort problem years ago where a new direct stick carpet was applied to a ground level slab that sat above a naturally ventilated carpark in a location that had an ambient winter design condition of 3C.  The slab was not fully insulated below so became quite cool in places where insulation was missing.  By our calcs (Therm and IES calibrated with an IR gun) the absence of a backing on the carpet resulted in the floor surface temperature being 1-2C cooler.  This was material from a radiant temperature asymmetry point of view but also in cooling low level air that resulted in an unacceptable dry bulb gradient.  We made the insulation continuous and the problem was solved but it was a real lesson in close coupling (fast) of thermal mass versus something that has less of an impact on the loads and thus temperatures.

So it does depend and you have to use engineering judgement, it isn't just the weight, it is how closely coupled (fast or slow …) the mass is to the space. 

Graham

On 15/01/2013, at 2:30 PM, Justin Spencer wrote:

> There's also an issue here about whether or not there's any modeling of internal partitions or fixtures included in your model. My personal experience in modeling residential models is that they always lag more than you expect, so I've regularly responded by jacking up the internal mass. I certainly can fit metered HVAC data a lot better when I add mass beyond what the BABM suggests. I've guessed that's because my partitions don't quite match the amount of mass we're interacting with. 
> 
> I get what Doug is saying that there isn't much temperature change occurring in his bookshelf, but when you have literally several tons of exposed mass, even a degree or two of temperature change is quite a bit of heat. Those granite counters currently in fashion do something for the building mass. 
> 
> What's funny about this is that the conventional rule of thumb is that residential construction is "light" and has little impact from mass, while commercial buildings are "heavy" and have lots of mass impacts. My personal experience in studying this is that mass is extremely important in residential construction because they "float" a lot of the time and also have higher solar gains than commercial spaces. 
> 
> One possible area of fruitful research I've thought about is to try and get some utility direct load control data and look at the "response time" of residences in terms of how long it took to have the temperature rise a certain amount when the HVAC system was constrained. You'd at least get some good data about how responsive the houses are supposed to be and alter what you can in your models to get them to better match real life (whether or not you actually figured out the root cause of the problem in the model). 
> 
> On Mon, Jan 14, 2013 at 7:58 PM, Doug Hittle <hittle.doug at gmail.com> wrote:
> I believe that E+ and predecessors take/took what ever was described
> as "internal mass" and parsed it into an equivalent "partition" of
> some "equivalent"  area. This can lead to some counter intuitive
> results. Most might expect that "internal" or "thermal" mass is going
> to moderate energy transfer ("store"). But suppose I describe a
> banquet hall full of folding mdf or plywood tables (light weight).
> Compared to an empty room, the area available to absorb radiation and
> convect it to the room is increased by easily 1/3. All of a sudden the
> room is "fast" and energy gains are translated into cooling loads much
> more quickly.
> 
> So let's agree with Joe that your question can be answered by "walking
> around." In looking around my office I find that one partition is
> covered with book cases and that the other side of the partition is
> also covered with books. If I thought this was an important feature
> that needed to be modeled (which I don't and which isn't) I would not
> describe any internal equivalent mass, I would simply describe a
> partition made up of books-air space-gyp board wall-air space-books.
> 
> But why do I think this is irrelevant additional work? I can weigh
> some books to determine their density. We think books are heavy but
> imagine putting an equivalent volume of concrete in your backpack.
> Then they are not so "thermally massey". Also the books never receive
> direct sunlight and the temperature in my office varies little because
> it is in my heavily insulated, air tight home. Since the guzinta
> equals the guzouta over time and since the guzinta and guzouta, even
> for this massive double book lined partition, are near zero because
> the materials are at constant temperature, more or less, the mass of
> the books is not very important to the room energy balance.
> 
> I don't know what thermal mass is anyway. The term must have evolved
> partly from passive solar designs, where energy storing materials
> exposed to the sun are important,  and from the brick trade
> association who left a confusing legacy of claims that a masonry house
> had a higher R-value because of the "thermal mass." Mind you I think
> bricks are great, I have some myself. However, there is nothing in the
> definition of R-value that is labeled "brick."
> 
> My apologies Dr. Wang. I expect that what I have just said is more or
> less useless as an answer to you questions. However, a productive
> guide to a first approach might be found in:
> 
> Consider a Spherical Cow: A Course in Environmental Problem Solving, John Harte
> 
> It is easier to bound a problem than to defend anybody's guess at the
> mean. The correct answer always seems to be "It Depends."
> 
> Doug Hittle
> 
> 
> 
> 
> On Mon, Jan 14, 2013 at 1:16 PM, Qinpeng Wang <qpwang at gatech.edu> wrote:
> >
> > Dear colleagues,
> >
> > I'm looking into furniture/internal mass modeling in energy models particularly within the environment of EnergyPlus but also not limited to that.
> >
> > I have two questions I guess:
> >
> > 1, Are there any modeling guide or reference building report that may give recommendations/instructions on how to represent furniture/internal mass inside a building if there is no much detailed information? I guess all I'm asking is are there any published data-set, or default values in terms of material property and area to represent furniture like what could be easily found about schedules and occupancy density?
> >
> > COMNET modeling guide has only one paragraph about it and it mentioned "The interior thermal mass and modeling assumptions in the baseline building shall be the same as the proposed design."
> > Currently what I have is from Building America House Simulation Protocols: "The internal mass of furniture and contents shall be equal to 8 lb/ft2 of conditioned floor space. For solar distribution purposes, lightweight furniture covering 40% of the floor area shall be assumed."
> >
> > 2, DOE reference commercial buildings have internal mass objects in the idf. file, where does the information come from?
> >
> > Appreciate your input!
> >
> > Best,
> >
> > Qinpeng
> >
> > --
> >
> > Qinpeng Wang, PhD Student
> >
> > College of Architecture
> >
> > Georgia Institute of Technology
> >
> > Atlanta, GA 30332-0155
> >
> >
> > _______________________________________________
> > Bldg-sim mailing list
> > http://lists.onebuilding.org/listinfo.cgi/bldg-sim-onebuilding.org
> > To unsubscribe from this mailing list send  a blank message to BLDG-SIM-UNSUBSCRIBE at ONEBUILDING.ORG
> >
> _______________________________________________
> Bldg-sim mailing list
> http://lists.onebuilding.org/listinfo.cgi/bldg-sim-onebuilding.org
> To unsubscribe from this mailing list send  a blank message to BLDG-SIM-UNSUBSCRIBE at ONEBUILDING.ORG
> 
> _______________________________________________
> Bldg-sim mailing list
> http://lists.onebuilding.org/listinfo.cgi/bldg-sim-onebuilding.org
> To unsubscribe from this mailing list send  a blank message to BLDG-SIM-UNSUBSCRIBE at ONEBUILDING.ORG

-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://lists.onebuilding.org/pipermail/bldg-sim-onebuilding.org/attachments/20130131/007d5366/attachment-0002.htm>