Ashrae 90.1 - Unmet hours

29 posts / 0 new
Last post

Ashrae 90.1 (2004) Appendix G3.1.2.2 requires a baseline building to
have less than 300 unmet hours. What exactly does this mean?

To illustrate my question: assume you have a building with 301 zones,
and each zone has 1 unmet hour per year. This gives you a total of 301
unmet hours, and requires you to increase your baseline equipment
capacity. But you could argue that, on average, the building has only 1
unmet hour per year.

Have any of you run into this? Is it addressed in an addendum
somewhere, etc?

Any help is appreciated. Thanks,

Jim Crockett, P.E.

Crockett, Jim's picture
Offline
Joined: 2011-09-30
Reputation: 0

Jim,

That's actually a really good question that I was afraid to ask when I
first encountered it - kudos to you! I've currently resolved to follow
what others seemed to be doing within and outside of my office: Sum up
all unmet hours for cooling and heating between the zones just as you
describe. In your example, I'd agree that the unmet hours of your 301
zone building total 301.

I do agree that this doesn't seem intuitively to be the intent of the
standard, however between what is suggested within 90.1, the LEED
handbook, and the LEED credit templates - I honestly can't see any clear
indication either way on which is the appropriate interpretation.

I think the appropriate metric for ensuring appropriately sized systems
should be something like: "hours of the modeled year in which at least
one zone has an unmet cooling/heating load," but I think that was
avoided by all concerned parties because it's too wordy!

My acting interpretation, again referencing your example, is that all
systems of your 301 zone example affecting the zones with unmet
cooling/heating hours should have their heating/cooling/overall sizing
capacity ratios increased incrementally until the design hours fall
below 300 (and/or within 50 of the sum from the other model, depending
on your situation).

Afraid I'm only really adding to the discussion here without providing a
solid answer. Would like to echo the desire to see anyone's experiences
that would help us know the "right" way to interpret this (in my case,
specifically in the context of a LEED submittal).

NICK CATON, E.I.T.

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

Nick,

Your interpretation is the correct one. According to the definitions in
Standard 90.1.

unmet load hour: an hour in which one or more zones is outside of the
thermostat setpoint range.

Michael Rosenberg

Rosenberg, Michael I's picture
Joined: 2011-09-30
Reputation: 0

I agree that Nick's interpretation is correct, but this is my biggest
frustration with 90.1. For multi-zone buildings, the limit should be on the
sum of "non-concident" unmet load hours for each zone. For standard energy
modeling output reports, you can determine the total annual unmet hours for
each zone, but you don't know which hours of the year they occur.

In Jim's example, what if each of the 300 zones had 1 unmet load hour that
all occurred during the same hour of the year? Or better yet, what if you
had 300 zones with 30 hours of unmet loads (all at the same hours)? By the
90.1 reasoning, you would have 9000 unmet load hours (greater than the
number of hours in a year!).

I have thought about requesting a change to 90.1, but if the standard were
to require people to determine the sum of non-coincident unmet load hours
for each zone, it would put a large burden on the energy modeler to develop
hourly output reports for each zone and hand calculate this. It seems like
we're stuck with what we have until the energy modeling programs can produce
an output report that describes the sum of non-coincident unmet load hours.

Kendra Tupper, PE, LEED AP

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

Mike,

I appreciate you bringing this up. This definition drives straight to
the heart of defining the issue-at-hand... Since I kinda presented both
sides of the issue at once - I'd like to clarify where I stand regarding
what the correct interpretation should be.

The logic is as follows: If a modeled year has 8,760 hours, can there
be 10,000 unmet load hours? By strict reading of the standard's
definition below, I would put my foot down stating there can only be
8,760, at most.

By common practice however, it appears a majority (myself included) sum
unmet cooling/heating hours between the zones, even if they should fall
on the same modeled hour, against the intent of the standard.

My pure speculation (for what it's worth, as a young EIT) is this
practice developed because eQuest BDL reports don't present the crunched
numbers in a way that makes the sum of unmet load hours, as intended by
90.1, easy to determine. I wouldn't be shocked to learn other energy
modeling software packages generate LEED compliance summaries featuring
unmet load hour totals in sync with the real intent of ASHRAE 90.1.

If there's anything I've learned from going out on a limb, it's that I'm
sure to learn something whether I fall or not!

NICK CATON, E.I.T.

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

I think eQuest and any DOE2 based software does report the hours of
loads not being met as required by Appendix G. In the BEPS report it
gives "PERCENT OF HOURS ANY SYSTEM ZONE OUTSIDE OF THROTTLING RANGE". My
understanding of this number is that it is a percentage of scheduled fan
run-time hours, so some calculation may be necessary.

Michael Rosenberg

Rosenberg, Michael I's picture
Joined: 2011-09-30
Reputation: 0

Dear All,

Explanation of Unmet Hours is as follows:

Unmet Hours:

Unmet hours of a building are the summation of the number of hours when the
heating or the cooling set point temperature of a zone is not met either by
the HVAC system or by the plant.

Understanding/Interpreting/Calculating the number of unmet hours:

* Unmet hour is for a particular zone when the zone indoor temperature
is higher than the heating or cooling set point specified in that hour.
* The number or the percentage of unmet hours in a building is usually
given as one of the outputs of the simulation.
* Zone wise unmet can also be read from the various output files
provided by the software used for simulation.

(Example:

Visual DOE: "SS-J System Peak Heating and Cooling Days" report &

Energy Plus: Output Variable, "Time Cooling Set point Not met")

* When two zones are unmet at the same hour, this will count to one
unmet hour for the building.
* When two zones have unmet hours during different non overlapping
times of a day, the total number of unmet hours in that day is the summation
of these unmet hours of each zone. This total for the year should be
considered as the total unmet hours of the building.

Example:

When each zone is unmet in the specified hours as beside,

Zone 1 unmet during: 6 8 14
16

Zone 2 unmet during: 6 8 12
16

Zone 3 unmet during: 7 8 12
13

Total number of unmet hours of the building: 7 hrs and not 12hrs.

6 7 8 12 13 14 16

* When percentage of unmet is specified, than this is the percentage
of total number of hours (1 year- 8760 hours) for which the simulation is
performed (not just the occupied hours)
* As per ASHRAE 90.1-2004, the unmet hours of the total building
should be less than or equal to 300 hours and the difference in the base
case and proposed case should be less than or equal to 50 hours.
* If unmet load hours in the proposed building exceed the unmet load
hours in the baseline building by more than 50, then the size of equipment
in the baseline building shall be reduced incrementally, until the condition
is satisfied.

Thanks

Ashu Gupta

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

dear All...

the entire issue with unmet hours has been analysed on my website, on this
link

http://greenbuildinggenie.com/tips/leed-issues/item/15-understanding-total-number-of-unmet-hours-as-per-ashrae-901-

regards

Deepak Agrawal's picture
Offline
Joined: 2011-10-02
Reputation: 0

Thanks Deepak for pointing this out. If anyone has any other
interpretations/suggestions, please let us know so that the website can be
updated.
Vishal

Vishal Garg's picture
Offline
Joined: 2011-10-02
Reputation: 0

Am I missing something or does this table and the corresponding answer
(7 unmet hours) not make sense.

Zone 1 unmet during: 6 8 14
16

Zone 2 unmet during: 6 8 12
16

Zone 3 unmet during: 7 8 12
13

If the columns (hours outside of throttling range) all represent
simultaneous hours in the model then wouldn't the total number of hours
be the sum of the maximum number of hours for each time period?
(7+8+14+16=45 hours outside of range) I have looked at the chart
multiple ways and do not see how seven (7) could ever be the correct
value.

Yes, projects with hundreds of zones may be penalized however ASHRAE
allows similar spaces(orientation, load , HVAC system) to be combined
into a single zone which can reduce the impact of this issue.

I personally sum all of the hours ANY zone is outside of range without
regard to when the time period is. I feel that this is the most accurate
method of correctly sizing the HVAC systems.

Seth Spangler, LEED(r) AP

Tiger06's picture
Offline
Joined: 2011-02-14
Reputation: 0

To get 7 unmet hours you need to interpret the numbers listed as hours of the day when the load was unmet. There are 7 unique hours when the load was not met. There are differing opinions on how unmet load hours should be calculated but this seems like the most logical but possibly not the most practical depending on the simulation software used.

David Wasserman

David Wasserman's picture
Offline
Joined: 2011-10-02
Reputation: 0

The unmet hour table meant to show the hours of the day in which that
particular zone is unmet and not the number of hours.

The same information is shown in the below table for one particular day-
suppose Jan 1st

In the table below:

? Yes ? represents that the hour is unmet by the system

? No ? represents that the temperatures are met in the zone

*Hour of the day*

*Zone 1*

*Zone 2*

*Zone 3*

*Number of unmet hours of the building*

*1st*

No

No

No

*2nd*

No

No

No

*3rd*

No

No

No

*4th*

No

No

No

*5th*

No

No

No

*6th*

Yes

Yes

No

1

*7th*

No

No

Yes

1

*8th*

Yes

Yes

Yes

1

*9th*

No

No

No

*10th*

No

No

No

*11th*

No

No

No

*12th*

No

Yes

Yes

1

*13th*

No

No

Yes

1

*14th*

Yes

No

No

1

*15th*

No

No

No

*16th*

Yes

Yes

No

1

*17th*

No

No

No

*18th*

No

No

No

*19th*

No

No

No

*20th*

No

No

No

*21st*

No

No

No

*22nd*

No

No

No

*23rd*

No

No

No

*24th*

No

No

No

*Total number of unmet hours in each zone*

4

4

4

*-*

*Total number of unmet hours of the building*

*7*

As seen in the table:

Total unmet hours in the building is not the sum of number of unmet hours in
each zone (which is 12hrs) but is the sum of number of unmet hours of the
building (which is 7hrs).

I am also attaching the table as an image.In case the formatting goes wrong
in the mail, the attached image can be referred.

Regards,

Surekha Tetali

surekha tetali's picture
Offline
Joined: 2011-10-02
Reputation: 0

Thanks to all for the insight and information! This makes much more
sense. Is there a report in eQUEST that calculates unmet hours in this
way, or do I need to do some custom analysis?

Jim

Crockett, Jim's picture
Offline
Joined: 2011-09-30
Reputation: 0

Some of DOE recommendations is using R-50 in the attic ( above the ceiling). If the HVAC ducts are also in the attic, is it true that this might result in condensation on the ducts espicially in moderate to hot climates? is there any literature on this or any practical experience.
Any help is really appreciated.
H. Abaza

Abaza Hussein's picture
Offline
Joined: 2011-10-02
Reputation: 200

Ashu,

I am in agreement with everything you stated except possibly this;

"When percentage of unmet is specified, than this is the percentage of
total number of hours (1 year- 8760 hours) for which the simulation is
performed (not just the occupied hours)"

My understanding is that some simulation programs (DOE2 and eQuest in
particular) report this as a percentage of hours that the fan is
running. I could be wrong, and I am sure some of you DOE2 experts can
comment on what is reported in the DOE2 BEPS report.

Michael Rosenberg

Rosenberg, Michael I's picture
Joined: 2011-09-30
Reputation: 0

I also interpret the 90.1 PRM definition as Sureka's table illustrates.
I imagine the intent of the original writers was that you would get
about one hour/day for either morning startup or at/near design
conditions, when MANY zones might be "outside the throttling range"
during that hour. Any other interpretation penalizes buildings with more
zones. And it should not matter whether Zones 1,2 and 3 are served by
the same system or separate systems.

For DOE2/eQuest, as far as I know, this is essentially how the BEPS %
unmet hours are determined, though apparently the percent references
system "on" hours. I THINK this is calculated as a fraction of the
system with the greatest "on" hours, but I'm not sure how cycling "on"
is treated. I think that if a model includes any system, even a tiny
virtual independent system, with continuous operation, then the overall
% from BEPS can be multiplied by 8760 (hopefully NOT 8760 + design day
hours)!

Unless it has been added to the newest versions of eQuest, there are NO
DOE2/eQuest summary outputs which decompose "unmet hours" into heating
and cooling for the BUILDING, as requested by the PRM and EAc1 forms.
However this decomposition is not specifically required by the language
of the standard. AND these #s might not necessarily be additive, a good
reason NOT to try to decompose and then sum these. During startup on a
winter day, many interior zones might be "undercooled" while many
exterior zones are "underheated," yielding ONE unmet load hour, but also
ONE hour cooling loads not met and ONE hour heating loads not met...
which would appear to become TWO hours if entered separately then
summed.

Fred W. Porter
Architectural Energy Corp.

P.S. Before increasing system or zone capacities, it's always wise to
check myriad other inputs to assure that some schedule or setting hasn't
turned off the heating or cooling or reset SA temps in some
unanticipated fashion.

The unmet hour table meant to show the hours of the day in which that
particular zone is unmet and not the number of hours.
The same information is shown in the below table for one particular
day- suppose Jan 1st
In the table below:

?Yes ? represents that the hour is unmet by the system
?No ? represents that the temperatures are met in the zone

Hour of the day
Zone 1
Zone 2
Zone 3
Number of unmet hours of the building

1st
No
No
No

2nd
No
No
No

3rd
No
No
No

4th
No
No
No

5th
No
No
No

6th
Yes
Yes
No
1

7th
No
No
Yes
1

8th
Yes
Yes
Yes
1

9th
No
No
No

10th
No
No
No

11th
No
No
No

12th
No
Yes
Yes
1

13th
No
No
Yes
1

14th
Yes
No
No
1

15th
No
No
No

16th
Yes
Yes
No
1

17th
No
No
No

18th
No
No
No

19th
No
No
No

20th
No
No
No

21st
No
No
No

22nd
No
No
No

23rd
No
No
No

24th
No
No
No

Total number of unmet hours in each zone
4
4
4
-

Total number of unmet hours of the building
7

As seen in the table:
Total unmet hours in the building is not the sum of number of unmet
hours in each zone (which is 12hrs) but is the sum of number of unmet
hours of the building (which is 7hrs).

I am also attaching the table as an image.In case the formatting goes
wrong in the mail, the attached image can be referred.

Regards,

Surekha Tetali

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

I believe that eQuest system and zone reports (SS-R and SS-F) only
report unmet hours when the fans are running. However, the percentage
of total hours as reported by the BEPU and BEPS reports use the entire
year as the denominator (ie 8760). I've seen a previous LEED project
accepted by stating that there are fewer than 300 unmet hours based on
the percentage reported on the BEPS report (in that case it was 3% or
263 hours).

Luka Matutinovic, B.A.Sc., LEED(r) AP

Matutinovic, Luka's picture
Joined: 2011-09-30
Reputation: 0

An HTML attachment was scrubbed...
URL:

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

EnergyPlus produces a simple summary table for this called "Time Setpoint Not Met." It has hours not met by zone and for entire facility, taking into account coincidence. Results are separated by heating and cooling as well as total time vs occupied times.

EnergyPlus uses a deviation from setpoint of 0.2C to trigger unmet time. I hear DOE2 uses a deviation of 1.0F as a trigger. I think this is one area where 90.1 could be clearer.

Griffith, Brent2's picture
Joined: 2011-10-02
Reputation: 0

It's splitting hairs, but one item in the email from Ashu is not quite
correct. The following sentence is from his email, and the incorrect
line is in bold:

* As per ASHRAE 90.1-2004, the unmet hours of the total building
should be less than or equal to 300 hours and the difference in the base
case and proposed case should be less than or equal to 50 hours.

What Ashrae actually says is, ". . .unmet load hours for the proposed
design shall not exceed the number of load hours for the baseline
building design by more than 50."

If you have a building where the baseline has 300 unmet hours, and the
proposed has 0 unmet hours, no correction is needed. A correction is
only needed if the unmet hours in the proposed exceeds those in the
baseline by more than 50. (or if either exceeds 300). Worded
differently, if the baseline has 300 unmet hours, and the proposed
system has 0 unmet hours, you do not need to make the baseline more
stringent.

Crockett, Jim's picture
Offline
Joined: 2011-09-30
Reputation: 0

You are right. The Unmet hours of the proposed design can be less than base
case unmet hours and the difference can be more than 50 hours (or upto 300
hours as shown by you).

However, the ASHRAE 90.1 2004 user manual on page G37states:

"If the number of hours that loads are unmet by either the systems or plant
of either the baseline or the proposed model exceeds 300 hours of the 8760
hours simulated, or *differs by more than 50 hours between the two models*,
these simulation results will not be accepted as valid." I think this is
incorrect.

According to ASHRAE 90.1 2004 - "unmet load hours for the proposed
design *shall
not exceed* the number of unmet load hours for the baseline building design
by more than 50." Reverse is not required.

So, the unmet load hours for the proposed design can be less than the number
of unmet load hours for the baseline building design by more than 50.

Vishal

Vishal Garg's picture
Offline
Joined: 2011-10-02
Reputation: 0

I use to agree with the statement and methodology you have provided.
However, there have been discussions that definition of "exceed" is the
"difference of". Hence making this statement false. In addition the
ASHRAE User's Manual also indicates that it is the "difference of" and
does not use the word exceed.

ARTURO HERNANDEZ

Hernandez, Arturo's picture
Joined: 2011-09-30
Reputation: 0

I'd be very interested to read up on the previous discussions, but a
search of the archive didn't turn up much.

90.1-2004, 90.1-2004 User's Manual, and 90.1-2007 all use "exceed" in
this sentence. I don't have a copy of the '07 User's Manual, but even
if it does use the phrasing "difference of", I don't see how USGBC can
hold anyone to it; it is not written that way in the standard. End of
discussion. (Just my opinion.)

James D. Waechter, PE, LEED AP

James D. Waechter's picture
Joined: 2011-09-30
Reputation: 200

The USGBC received confirmation from ASHRAE that it is OK if the baseline unmet load hours exceed the proposed unmet load hours by more than 50, given that they are both below 300.

Any review comment stating anything to the contrary was written before the guidance was received and it is completely acceptable to reply to the review comment with a comment stating it is OK that the baseline exceeds the proposed by more than 50.

Best Regards,

Gregg Liddick, EIT, LEED? AP

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

This is good to know because I know in the past USGBC did not allow it. I think it makes more sense now that they have gotten the OK from ASHRAE.

ARTURO HERNANDEZ

Hernandez, Arturo's picture
Joined: 2011-09-30
Reputation: 0

I have done a small experiment; I created a block model and simulated it on
auto sizing.
There were no unmet hours in this case. Hence,I manually decreased the
supply cfm (flow/area) in the zone which started increasing the unmet hours.
The % of unmet was varying as a % of number of hours HVAC fan is ON.The cfm
was decreased till the unmet hours were 100% (cooling unmet). In this case
the total number of unmet hours of the building are similar to the total
number of hours the system fan is ON. I changed the cooling setpoint
temperature when the system fan is OFF and this had shown no effect on unmet
hours.

These percentage unmet hours (in Visual DOE) hence seem to be the % of the
total number of hours the HVAC fan is ON.

Surekha Tetali

surekha tetali's picture
Offline
Joined: 2011-10-02
Reputation: 0

We also interpreted the unmet loads hours as Sureka's table illustrates below. We believe this to be the correct interpretation and have included tools in the Virtual Environment for assessing unmet load hours accordingly. The tools counts only occupied hours for each zone (per the schedule for that particular zone), and unmet load hours are determined in keeping with the controls throttling ranges established by the user for heating and cooling of individual zones. The occupied hours and set points can thus vary from one zone to another, without double-counting a particular hour during which two or more zones are out of range. That said, in support of other needs and interpretations, we also provide means for users to count unmet loads or other hourly bin data for any set of hours in the day and to total all hours or just the non-coincident hours via a logical ?OR? test

Regards,

Timothy Moore

Timothy Moore2's picture
Offline
Joined: 2011-10-02
Reputation: 200

Scott,

Wanted to express thanks I'm certain many of us in the community share
for you, Steve Gates and the rest of the DOE-2 team. I know I'm not the
only one looking forward very much to this new feature!

NICK CATON, E.I.T.

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

I have over R-58 in the attic with the first layer R-19 so it is short
cut - only 22 1/2 inches wide
to allow for the wood nominal 2 inch trusses. In the US fiberglass batts
R-25 and up are full cut -
a full 24 inches or 16 inches wide for application above the framing.

I put Arches of R-25 foil faced over the return air duct with the foil
facing out and the running direction
of the batt is across the duct and the edges are taped with heavy duck
tape. It works very well.

I am developing detailed records of the utility bills and heating /
cooling degree days. These records
go back to January 1987. I am very near the Sterling Virginia USA
weather station so I can get
daily and hourly data and also RADAR data going back to 1991.......

I do not recommend putting the return ducts in the attic anymore. I
have developed a much better
approach, I just can't seem to get funding support to build a demo.

John R Ross III PE

JRR's picture
JRR
Offline
Joined: 2011-10-01
Reputation: 0