T-Stat is acting Screwy

17 posts / 0 new
Last post

I have a lab space that is kept between 73 and 70 degrees all year long.
One of my parametric runs has the cooling and heating T-stat schedules
changing to have night time setbacks to 80 and 60, respectively. When i use
these, i get more energy spent overall. specifically, i have a chilled
water meter, steam meter, electric meter and hot water meter to monitor
everything. The chilled water increases while the others slightly decrease
when i use the set back schedules. Any ideas?

--
Rob Hudson

rdh4176's picture
Offline
Joined: 2011-02-14
Reputation: 0

Are you changing them in DD? If you are and it's not working I'd do it in
the .inp deck. Not too hard to fix in BDL. Just find the specific schedules
you want and don't change and usernames.
Carol

cmg750's picture
Offline
Joined: 2010-10-05
Reputation: 0

I am working in the DD and i have created my own schedules.

rdh4176's picture
Offline
Joined: 2011-02-14
Reputation: 0

Try opening up your inp file and retyping it there. You might be able to see
why it snaps back. Have you used BDL before? Its not too hard to see your
schedules.
Good luck'
Carol

cmg750's picture
Offline
Joined: 2010-10-05
Reputation: 0

It is constant volume VAV system.

rdh4176's picture
Offline
Joined: 2011-02-14
Reputation: 0

Hello partners,

I ask for your help because I am having problems exporting data from eQUEST.

Two weeks ago I were able to make a .DAT file with hour by hour in a year the values of Electric n Meter.

This was the name of the filed that I could create.

100510 Barcelona - Baseline Design_HourlyData2_01.dat

Today I have tried to do the same with different files that are from differente cities and I can?t. I can?t remember what I did to generate this file.

Please can someone help me?

Thanks in advance.

Mateo de Guadalfajara

Mateo Guadalfajara's picture
Joined: 2011-09-30
Reputation: 0

Rob,

I have a number of questions:

1.
Do you have many lab hoods? And are exhaust and make-up air
revised for the lab hoods at the night set back time?
2.
What is the percent of outside air to the lab? Does this
percentage change during the night set back time?
3.
Do you have a pre-heat or exhaust heat recovery and pre-cooling
on your make-up air? What is the temperature setting of this pre-heat
or heat recovery and pre-cooling? Does the temperature settings on
these change at the night set back time?
4.
Do you utilize an air or water side economizer for 1st stage
cooling? Does this use change any at night set back time?
5.
Does the fan run continuously day and night? Have you ran
models with the fan running continuously, and with fan cycling based on
demand at night and off completely at night, to see how the results
compare?
6.
Do you have a internal load watts per square foot and latent
load on the space from interior lighting and equipment? Do these loads
change at the night set back time?
7.
Are you adding humidification or dehumidifying the space based
on some humidity settings? Does the humidity settings change during the
night set back time?

One of these things is probably causing the increased chilled water
cooling load. Often if you have high internal loads, even at night,
then changes to the fan cycling and the amount of outside air brought in
at night, will increase the chilled water cooling load. I would review
all of these areas and run various test models to see how they each
affect your energy usage when modified.

Our firm just completed modeling some very large lab facilities with
more than 55 exhaust hoods in the building, with high internal loads
24/7 and 100% outside air. As long as the models are set up correctly
they are normally right. It takes some real design and thermal dynamic
thought and often may models to really get your mind right with what is
truly going on with the facility. Its important to keep an open mind to
what is going on. Having been doing computer hourly modeling for over
17 years, it is often easy to think you have a handle on what is going
on with the building, but it is important to keep an open mind and
investigate all the different avenues that you can think of until figure
out what is going on. I have 99% of the time that I have blamed the
screwy program having problems that I have found that I just didn't look
at all the different angles enough.

David A. Bastow

David Bastow's picture
Offline
Joined: 2011-09-30
Reputation: 0

Thanks for your thoughts and here is your answers:

1) yes, but there is no night set back time.
2) everything is 100% oa and constant volume 24/7/365
3)no exhaust heat recovery, and steam preheat. the thermostat is set for 70
- 73 heating and cooling and the set back is to 60 and 80 at night for my
parametric run.
4) the cooling comes from a central chilled water plant for the entire
campus, and i just created a chilled water loop, added a meter and a pump
and it seems to be running happily.
5) the fans run all the time, day and night
6) I do have internal loads with latent heat, and it cycles down at night to
almost nothing. These loads are currently the same for the model and the
parametric run.
7) i have set humidity levels to 20% and 53%, which also do not change with
my parametric run.

hope this give you enough information.

thanks again

rdh4176's picture
Offline
Joined: 2011-02-14
Reputation: 0

Rob,
I think the problem is in the thermostat schedule and/or the space temp
and/or the throttling ramge. A quick look at your .inp file should suffice
to figure out which one. Look at it in Word. And, don't worry R & H I am
working on your project. :)
Carol
Second thought, I think it's your TR. As Steve Gates has told us the if your
SA temp is 70 and your TR is 2 the result is 70 +1+1 or 73 on the high side
70-1-1, or 68 on the low side. Steve says it much better. You can find all
of the discussion in the archives. Nick has sent you links and everything to
get there. Go there!

cmg750's picture
Offline
Joined: 2010-10-05
Reputation: 0

Rob,
OK, I can't resist a bit of lunchtime "Stump the Chump." First I'd say verify the setpoint change is actually working by examining zone SS-Os for the baseline and parametric run. But I'll assume it is, and offer my explanation.

One piece that is left out in all this is the supply air temperature (SAT) control, which is very important for a CV-reheat application. And of course location.

My bet is that eQuest is resetting your SAT down at night based on hitting the upper RH limits during the zone setback scenario. This is why the CHW load increases. This is also why HW savings might not be as great as expected.

Why? Holding all else the same, the hourly temperature in a lab zone will with minimal internal gains as described will follow the heating setpoint down; the cooling setpoint will not be in control. Generally this will lead to high internal relative humidity (which is why this setback is only done carefully and to a limited extent in labs) due to the moisture in the OA, even if internal latent gains are nil. If the model has any SAT reset, the MAX-HUMIDITY resets the hourly SAT back down to the MIN-SUPPLY-T when the model zone RH exceeds the RH setpoint, in this case a relatively low 53%. This mode occurs more often in the "parametric" case as the zone temperature decreases.

Fred Porter

P.S. The only way to truly follow the SAT is to set up some hourly reports. I recommend the use of variable 225 for your AHU.

Thanks for your thoughts and here is your answers:

1) yes, but there is no night set back time.
2) everything is 100% oa and constant volume 24/7/365
3)no exhaust heat recovery, and steam preheat. the thermostat is set for 70 - 73 heating and cooling and the set back is to 60 and 80 at night for my parametric run.
4) the cooling comes from a central chilled water plant for the entire campus, and i just created a chilled water loop, added a meter and a pump and it seems to be running happily.
5) the fans run all the time, day and night
6) I do have internal loads with latent heat, and it cycles down at night to almost nothing. These loads are currently the same for the model and the parametric run.
7) i have set humidity levels to 20% and 53%, which also do not change with my parametric run.

hope this give you enough information.

thanks again

Fred Porter's picture
Offline
Joined: 2011-09-30
Reputation: 0

I'm trying to wrap my head around this the best i can.

I looked at my SAT and it is controlled by a reset schedule. the
temperature of the space never reaches anything above about 72 deg, which
makes me think that what you are saying is very true and the heating t-stat
schedule is in control all the time.

So can you tell me what it is that i can do to fix this?

rdh4176's picture
Offline
Joined: 2011-02-14
Reputation: 0

Hi Rob,
You can send your .inp and .pd2 files to whoever you trust to look at this.
You are aware, of course, that we are all working: some for others some for
themselves. Someone will either give you the gift of a free peer review or
you can request one of us to do a peer review for a sum of money. Always
remember that you get what you pay for.
Regards,
Carol

cmg750's picture
Offline
Joined: 2010-10-05
Reputation: 0

I believe what you did to push your hourly report data out to a .dat file
was to go to the unlikely location of "non-hourly reports" tab under the
Project & Site tab and then set the Hourly Data Save Format from "No Save"
(which imbeds the hourly report info within the .sim file) to "Formatted".
Make sure you do it for all report types (loads, system, plant & econ --
although I imagine it is plant reports you are generating).

That said, there is now the ability in either D2SimViewer or eQUEST
directly to parse the hourly report data directly into an Excel
Spreadsheet. In D2SimViewer, just hit the Hourly Reports button - top
right. In eQUEST, it is under the file menu -> File - Export Hourly
Reports.

Hope this helps.

Brian

bfountain's picture
Offline
Joined: 2011-09-30
Reputation: 201

Rob,
I'm a little unclear about statements like "I looked at my SAT and it is controlled by a reset schedule." Does this mean that the reset schedule a) came from "The Wizard," b) actual AHU sequences, c) a departed modeler (you know who you are!) or some other source. This is important and should be checked.

What's happening to increase the cooling coil load is that at certain hours the baseline reset SAT (e.g. 60F high RH) is not providing 53% RH indoor conditions at reduced 65F zone temp even though it did at 70F. So DOE-2.2 resets down to the MIN-SUPPLY-T (probably 55F) in the "parametric" with the setback zone Ts. Thus, coil load shoots up at those hours, and there are no hours where the cooling load is decreased by changing, even increasing, the zone Ts.

The lab users need to decide whether the appropriate upper RH limit is really 53%. It if can be increased, then you can save money by increasing it, and instituting setbacks. And perhaps a setpoint decrease to only 67-68F might give better results in reality and model land. But it sounds like you are unsure of some of the model and the lab requirements. If this is the case I would be very careful about making recommendations to lab owners and users along the lines of "....reduce zone heating setpoint to 60F," because if they could just do that, they might not spend millions of dollars on Phoenix valves to control VAV. You need to model the correct SATs, and only reasonable zone conditions.

Fred Porter

I'm trying to wrap my head around this the best i can.

I looked at my SAT and it is controlled by a reset schedule. the temperature of the space never reaches anything above about 72 deg, which makes me think that what you are saying is very true and the heating t-stat schedule is in control all the time.

So can you tell me what it is that i can do to fix this?

Fred Porter's picture
Offline
Joined: 2011-09-30
Reputation: 0

A couple ideas.

1. Try removing certain schedules (let eQuest auto-calculate); then
you might be better able to highlight problem areas.
2. Check the throttling range again for zones in question.

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

Very interesting. It hadn't occurred to me that you might bump into the
dehumidification set-point during night setback (then again, a lot of
things haven't occurred to me). But it seems to need to be under a set
of just right conditions. When could the space cool down fast enough at
night, while still having a high enough humidity level to trigger
dehumidification?

I don't think this would occur during the winter. Let's assume this
building is up north where shell losses can add up and cool a space down
relatively fast during winter. Even if you're humidifying to 40% at 70F
during the day (which he's not) and dropping to 60F at night, you're
only hitting 57% RH, assuming constant absolute humidity. For a lab, the
make up air (which has a low absolute humidity because of the low
ambient temperatures) would probably dry out the space before the indoor
temp hits a point low enough to set of dehumidification.

So the other possibility I see (which is what Fred described, I believe)
is that during some nights the make up air is bringing in moist, cool
air, and the system isn't tempering it because the zone is satisfied. So
this extra air cools down the space and drags up the RH, triggering a
dehumidification mode that would be unnecessary at 70F indoor set-point.
But if this was the case, wouldn't your heating energy go down more than
slightly due to the reduced ventilating loads?

What am I missing here? How large are those internal latent loads (seems
to be a wild card to me)? I'm curious as to whether you've solved this
one Rob.

Cheers,

Eric

Eric O'Neill's picture
Joined: 2011-09-30
Reputation: 0

I have not yet solved it. My boss came over to me this morning and said we
are not to change the thermostat set points, so everything is as it was
prior. This was a major ECM for the project, but i guess its gone now...
hehe
I think if i have more time i will attempt to take a crack at it again. The
internal loads were minimal in the otherall design ( i think 1.4 w/sqft for
lighting and ~3 w/sqft for equip loads.)

rdh4176's picture
Offline
Joined: 2011-02-14
Reputation: 0