Re: [EnergyPlus_Support] Re: Zone Temperature Peaks too Early

OK, there's a lot of questions to answer so I'll try to be brief.

The slab is modeled with three layers: 6.75" of perlite in contact with the ground, 4" EPS insulation, and 4" concrete slab. Total R-value (IP) should be about 35. Ground temps are currently set to the default of 18C year round. Surface emissivity of the concrete is set to 0.9.

The exterior wall has 4" of polyiso insulation under the siding, so I don't think thermal bridging is significant. Also, like Joe, I would expect thermal bridging to cause the peaks to be earlier if it was a significant source of gain.

Infiltration is modeled using the ZoneInfiltration:DesignFlowRate object and is set to 0.0105 ACH (with coefficients 1,0,0,0). This is based on a blower door test of the actual house, which showed 0.21 ACH at 50 Pa. The 0.21 was then divided by 20 (based on a paper by Max Sherman) to estimate the natural infiltration rate. I'm also using the simplified models for zone mixing and zone ventilation. I realize AirflowNetwork would be a better way to model this stuff, but I don't have any experience with it and I'm not sure I have the time to figure it out.

The actual house also has a heat recovery ventilator, which has exhaust ducts in some rooms and supply ducts in others. I've modeled this using an air loop with a heat exchanger, outdoor air system, a dummy cooling coil that is always off, and a fan that is always on.

The solar radiation question brings up a good point. Due to the limitation of one shading surface per window, I have the operable windows modeled with only an exterior screen and fixed windows modeled with only an interior blind. In reality, the operable windows have both an interior blind and an exterior screen. I wouldn't think that the interior blind would make THAT much of a difference, but is it possible?

I'll try to get some charts posted of the measured vs. modeled temperatures. And, again, thank you all for your help.

Jeff

--- In EnergyPlus_Support@xxxxxxxxxxxxxxx, Joe Huang <YJHuang@...> wrote:
>
> There might be a slight mix-up about the original model mentioned by Jeff. Since that
> model was done in EnergyPlus, I don't think any supposed problems of DOE-2
> in handling solar radiation at low angles would apply. Having said that, this is the
> first I've heard of DOE-2 having such a problem, so could you (Jason) inform me about it
> (preferably offline since it doesn't have anything to do with EnergyPlus...)?
>
> The difficulty I have with exploring thermal bridging is that wouldn't it tend to quicken
> even more the time of the peak?
>
> Having exhausted the other proposed remedies, I would like to get back to the modeling of
> the slab foundation.
>
> Joe
>
> Joe Huang
> White Box Technologies, Inc.
> 346 Rheem Blvd., Suite 108D
> Moraga CA 94556
> yjhuang@...
> www.whiteboxtechnologies.com
> (o) (925)388-0265
> (c) (510)928-2683
> "building energy simulations at your fingertips"
>
>
> On 5/8/2013 3:00 AM, Jason Quinn wrote:
> >
> > How are you dealing with the linear thermal bridges? In a normal building these are
> > neglected but in a passive house they are usually considered. I would think that most
> > bridges would conduct rapidly since by definition they are weaknesses in the thermal
> > envelope.
> >
> > Also in doe2 I remember some cautions in how radiation was modeled for very low solar
> > angles. Could this be a factor?
> >
> > Feel like sharing your model so we can dig?
> >
> > On 8/05/2013 9:00 PM, "Joe Huang" <YJHuang@...
> > <mailto:YJHuang@...>> wrote:
> >
> > Well, at least you got a little bit of rise with one of the following (the 0.3
> > W/m2-K convection coefficient) :-) I was thinking primarily of reducing the
> > convection coefficient in steps, maybe by halves, but 0.3 W/m2-K equates to R-19
> > (IP units), which seems like a lot to me. Does this building have a slab-on-grade,
> > and if so, how are you modeling that? Can't think of anything else because from
> > your description the building is largely empty.
> >
> > Joe
> >
> > Joe Huang
> > White Box Technologies, Inc.
> > 346 Rheem Blvd., Suite 108D
> > Moraga CA 94556
> > yjhuang@... <mailto:yjhuang@...>
> > www.whiteboxtechnologies.com <http://www.whiteboxtechnologies.com>
> > (o) (925)388-0265
> > (c) (510)928-2683
> > "building energy simulations at your fingertips"
> >
> >
> > On 5/8/2013 12:13 AM, jeffreylauck wrote:
> >>
> >> OK, here's a summary of what I've tried and the results:
> >>
> >> Changed ZoneCapacitanceMultiplier from 1 to 2 ==> No change in peak time
> >>
> >> Set convection coefficient for AllInteriorWalls to 7.7 W/m2-K ==> No change in peak
> >> time
> >>
> >> Set convection coefficient for AllInteriorSurfaces to 7.7 W/m2-K ==> No change in
> >> peak time
> >>
> >> Set convection coefficient for AllInteriorWalls to 1 W/m2-K ==> No change in peak time
> >>
> >> Set convection coefficient for AllInteriorSurfaces to 0.3 W/m2-K ==> Peaks shifted
> >> to a later time for several of the days but it's not consistent
> >>
> >> I'm beginning to think it's something other than convection. Any thoughts?
> >>
> >> Thanks,
> >>
> >> Jeff
> >>
> >> --- In EnergyPlus_Support@xxxxxxxxxxxxxxx
> >> <mailto:EnergyPlus_Support%40yahoogroups.com>, Joe Huang <YJHuang@>
> >> <mailto:YJHuang@> wrote:
> >> >
> >> > From the sounds of it, i.e., hourly profiles are similar but just shifted
> >> forward, my
> >> > guess is that the amount of thermal capacitance is okay, but that the coupling to
> >> the air,
> >> > i.e., convection coefficients, are the source of the problem.
> >> >
> >> > Joe
> >> >
> >> > Joe Huang
> >> > White Box Technologies, Inc.
> >> > 346 Rheem Blvd., Suite 108D
> >> > Moraga CA 94556
> >> > yjhuang@
> >> > www.whiteboxtechnologies.com <http://www.whiteboxtechnologies.com>
> >> > (o) (925)388-0265
> >> > (c) (510)928-2683
> >> > "building energy simulations at your fingertips"
> >> >
> >> >
> >> > On 5/7/2013 3:39 PM, Griffith, Brent wrote:
> >> > >
> >> > > An easy thing to try is to increase the sensible thermal capacitance of the
> >> zone's air
> >> > > using the ZoneCapacitanceMultiplier:ResearchSpecial object.
> >> > >
> >> > > *From:*EnergyPlus_Support@xxxxxxxxxxxxxxx
> >> <mailto:%2AEnergyPlus_Support%40yahoogroups.com>
> >> [mailto:EnergyPlus_Support@xxxxxxxxxxxxxxx
> >> <mailto:EnergyPlus_Support%40yahoogroups.com>]
> >> > > *On Behalf Of *jeffreylauck
> >> > > *Sent:* Tuesday, May 07, 2013 4:02 PM
> >> > > *To:* EnergyPlus_Support@xxxxxxxxxxxxxxx
> >> <mailto:EnergyPlus_Support%40yahoogroups.com>
> >> > > *Subject:* [EnergyPlus_Support] Re: Zone Temperature Peaks too Early
> >> > >
> >> > > Jean,
> >> > >
> >> > > Thanks for your response. I suppose it's possible that internal surfaces are
> >> slowing
> >> > > convection heat transfer in the actual house, but there's not a lot of
> >> furniture in the
> >> > > space. A small couch, a chair, and a table are about it.
> >> > >
> >> > > The only way I can think of to test this theory is to increase the roughness of
> >> the
> >> > > internal surfaces in the zone. Do you have any other suggestions?
> >> > >
> >> > > Thanks,
> >> > >
> >> > > Jeff
> >> > >
> >> > > --- In EnergyPlus_Support@xxxxxxxxxxxxxxx
> >> <mailto:EnergyPlus_Support%40yahoogroups.com>
> >> <mailto:EnergyPlus_Support%40yahoogroups.com>,
> >> > > "Jean marais" <jeannieboef@ <mailto:jeannieboef@>> wrote:
> >> > > >
> >> > > > Could it be that the actual convection from surfaces is different or more or
> >> less than
> >> > > those modelled thereby transfering heat from surfaces to air faster or slower.
> >> Are there
> >> > > more internal surfaces like furniture which could contribute the these effects?
> >> > > >
> >> > > > Jean
> >> > > >
> >> > > > --- In EnergyPlus_Support@xxxxxxxxxxxxxxx
> >> <mailto:EnergyPlus_Support%40yahoogroups.com>
> >> > > <mailto:EnergyPlus_Support%40yahoogroups.com>, "jeffreylauck" <JeffLau ck@> wrote:
> >> > > > >
> >> > > > > Hello All,
> >> > > > >
> >> > > > > First off, thanks for a great community. I've learned a lot from searching
> >> through
> >> > > the message archives over the past few months.
> >> > > > >
> >> > > > > I have a model of a Passive House that I am trying to validate using
> >> measured data
> >> > > from the actual house. I've created an hourly schedule file for the lights,
> >> electric
> >> > > equipment, window usage, blind usage, and hot water consumption based on
> >> sub-hourly data
> >> > > collected at the site. I'm using a custom weather file from the roof-top
> >> weather station
> >> > > that monitors dry bulb temp, relative humidity, wind speed and direction, and
> >> global
> >> > > horizontal solar radiation. The diffuse radiation component was estimated using
> >> the Erbs
> >> > > model prior to importing the data into the weather utility. Currently I'm only
> >> looking
> >> > > at the summer of 2012.
> >> > > > >
> >> > > > > The problem I'm having is that the peak temperatures in the model occur 2-4
> >> hours
> >> > > before the measured data (2-3 hours in July and September, 3-4 hours in
> >> August). Here's
> >> > > what I've determined so far:
> >> > > > >
> >> > > > > 1. It's not due to a time shift in the data. I've verified that all data is
> >> GMT-8
> >> > > and that no Daylight Savings flags are included in the IDF.
> >> > > > > 2. It's not due to a lack of thermal mass. I added a bunch of internal mass
> >> as a
> >> > > test and the peaks were reduced but occurred at the same time as previous
> >> models. Also,
> >> > > I'm using CondFD with 9 nodes and a 1-minute timestep.
> >> > > > > 3. I don't think it's due to my custom weather file, as I get a similar
> >> results
> >> > > using TMY data.
> >> > > > >
> >> > > > > Does anyone have any suggestions on where to look next? Any ideas would be
> >> greatly
> >> > > appreciated.
> >> > > > >
> >> > > > > Thanks,
> >> > > > >
> >> > > > > Jeff
> >> > > > >
> >> > > >
> >> > >
> >> > >
> >> >
> >>
> >
>