search

kbenne / cbecc

Automatically exported from code.google.com/p/cbecc
0 stars 0 forks source link

Process Systems #630

Open GoogleCodeExporter opened 9 years ago

GoogleCodeExporter commented 9 years ago 
What NACM Section(s) are relevant to this issue?
5.1.2 Table 6

Explanation of issue:
The requirement for using System 10 is "Total computer room design cooling load 
is over 3,000,000 Btu/h)".  What if the user has more than one thermal zone 
with the classification of "Computer Room".  Is it the total design cooling 
load for the baseline building that determines this system type, or is it the 
total for one thermal zone?

Also, just want to confirm the intention is the that the CRAC/CRAH units are 
single-zone, i.e. one of these systems per baseline thermal zone.

Proposed resolution:

Please provide any additional information below.

Original issue reported on code.google.com by da...@360-analytics.com on 3 Jul 2014 at 11:56

GoogleCodeExporter commented 9 years ago 
Revising this issue to be general to all 'Computer Room' issues and related 
rules. 

Thinking more on the rules for implementing the 3,000,000 Btu/h CRAC->CRAH 
switch, this will be more difficult to implement than if the threshold was 
based on area or some other metric, non-load based.  This is because to 
properly size the baseline ChW system, it is best to include establish and size 
the baseline equipment in the sizing run.  Since the TOTAL Computer Room load 
is not accurately known until the sizing run is performed, we can't determine 
the applicability of the CRAH until after sizing. To add it in the annual run 
is possible, but it is more complex.

I'd propose the CRAH unit to be applicable to buildings with at least some 
threshold of Computer Room area. At 20 W/ft2, 3000000 Btu/hr equipment load is 
43949 ft2.  I'd suggest pegging the threshold at 40,000 ft2 or less to give 
some margin on the typical load.

Original comment by da...@360-analytics.com on 4 Jul 2014 at 6:13

GoogleCodeExporter commented 9 years ago 
Also, I noted in Appendix 5.4A that Computer Rooms have a design occupancy, 
service hot water load, and minimum ventilation requirement. My guess is that 
the hot water use would be associated with other, surrounding spaces, and that 
the occupancy to be intermittent rather than regular.

Original comment by da...@360-analytics.com on 4 Jul 2014 at 6:23

GoogleCodeExporter commented 9 years ago 
The rules for LAB and KITCH process systems are inconsistent. The descriptions 
in Section 5.7 Cooling System indicate the system is PVAV, but then in Cooling 
Capacity, it indicates 'One fan system per room'. 

If there are more than one of these process ThrmlZns in the building, is the 
intent that they are single or multizone systems?  

If multizone, one system/per floor, as defined for other multizone systems?

Original comment by da...@360-analytics.com on 4 Jul 2014 at 11:05

GoogleCodeExporter commented 9 years ago 
I agree, we don't want to base determination of baseline systems on cooling 
load or any other "variable" characteristic.  One option here is to base it on 
the process plug load in these spaces.  3,000,000 Btu/h ~ 880,000 Watts.  Maybe 
make the switch a plug load of 800,000 Watts.  This is a user input that is 
used for both the proposed and baseline in these spaces.

Original comment by rhedr...@archenergy.com on 7 Jul 2014 at 7:07

GoogleCodeExporter commented 9 years ago 
I agree with Roger's #4 comment, and was thinking something similar. Rather 
than base it on ft2, base it on the load. The user can enter a load in W/ft2, 
unlike the receptacle loads for non-process spaces, and it may be much greater 
than 20 W/ft2 for the data center.

CRAC/CRAH units are single zone systems.

Will look at other comments soon.

Original comment by JohnJArent on 8 Jul 2014 at 6:27

GoogleCodeExporter commented 9 years ago 
Comment 2 response: I don't see why the computer room would have a service hot 
water load. Occupancy should be very low but >0. I don't have a strong 
preference for the values we use. If you lumped the data center and support 
areas together in one zone, you could have a SWH load, but I don't think that's 
the best thing to do. Recommend setting SWH load to 0. 

Comment 3: I don't see the place in the ACM where it states, "One fan system 
per room". But for LABs, the baseline a PVAV or VAV multizone air handler (we 
could do either one per floor, or one per building), with the design CFM in 
each zone as the max of: cooling cfm requirement, ventilation cfm requirement, 
exhaust cfm, and the AHU cfm is the sum of the cfm for the zones. I don't have 
a strong preference for LAB system mapping. I recommend one per floor to be 
consistent with other system map rules. We should have a rule in there that 
selects between PVAV and VAV with central plant, as well.

For KITCHEN, it should be one make-up unit per kitchen (single-zone system).

Note: on VAC from 7/11 through 7/15. For urgent ACM-related questions, you can 
call my cell. Will return on Wed 7/16.

Original comment by JohnJArent on 10 Jul 2014 at 5:19

GoogleCodeExporter commented 9 years ago 
Re SHW loads for computer rooms - I disagree that this should be zero.  Workers 
in computer rooms will use the washroom and wash their hands just as often as 
any other workers, so the SWH/person should be non-zero.  I recommend using the 
office rate.

Original comment by rhedr...@archenergy.com on 10 Jul 2014 at 5:30

GoogleCodeExporter commented 9 years ago 
Re #7, the  'Computer Room' space function cannot be combined with other 
spaces.  I agree that support areas should have loads, and they do.  One 
fundamental question: should a standalone data center that has no have support 
areas or restroom have a shw load?  If there's no restroom or shower in the 
building, does it make sense to have shw energy use in the compliance analysis?

Original comment by da...@360-analytics.com on 10 Jul 2014 at 8:55

GoogleCodeExporter commented 9 years ago 
[deleted comment]
GoogleCodeExporter commented 9 years ago 
A data center with no restroom?  That means we are talking about a completely 
automated data center with zero occupants?

Original comment by rhedr...@archenergy.com on 10 Jul 2014 at 9:11

GoogleCodeExporter commented 9 years ago 
Yes, seems like it is not unreasonable for the data center building to be 
detatched from the admin/support building.

For example:
http://en.wikipedia.org/wiki/Utah_Data_Center

In this case, it is conceivable that the different buildings would be 
permitted, and therefore, modeled separately, right?

Original comment by da...@360-analytics.com on 10 Jul 2014 at 10:34

GoogleCodeExporter commented 9 years ago 
I think these types of massive data centers will be the exception, and we 
should base CBECC assumptions on a more typical computer room that will be part 
of a building with other space types.  Even in the Utah Data Center, I'll bet 
there are bathrooms in the data halls.

Original comment by rhedr...@archenergy.com on 10 Jul 2014 at 10:53

GoogleCodeExporter commented 9 years ago 
Re: #6 'One fan system per room", see Table 25 or attached screen shot.  I 
think what you've outlined makes more sense, so I will go with that.

So, to summarize what I think are the appropriate resolutions:
- CRAC, CRAH, and KITCH are all single-zone systems, LAB is multi-zone (if 
applicable).
- One LAB system per floor
- 0 SHW/people/ventilation assumed by rules for Computer Room spaces.  If the 
user specifies them for these spaces, they will be the same in the baseline and 
the proposed.
- The threshold for the CRAH baseline is 800kW of TOTAL building Computer Room 
receptacle load.  

Let me know if any of these assumptions are incorrect by COB Friday.

Original comment by da...@360-analytics.com on 10 Jul 2014 at 10:53

GoogleCodeExporter commented 9 years ago 
Attached is Table 25 screen shot

Original comment by da...@360-analytics.com on 10 Jul 2014 at 10:54

Attachments:

GoogleCodeExporter commented 9 years ago 
I think we should retain SHW and Ventilation rates at non-zero values per 
person.  Then if the occupancy is zero, then those loads are zero as well.  If 
the user specifies occupancy, then those loads are also non-zero at standard 
rates.

Original comment by rhedr...@archenergy.com on 10 Jul 2014 at 10:56

GoogleCodeExporter commented 9 years ago 
I think we are actually saying the same thing.  Here is an expanded version of 
what I am proposing:

For the 'Computer Room' SpcFunc, the default occupancy is 0.  If occupancy is 
user-defined > 0, the default ACM SHW and ventilation assumptions are used. 
Furthermore, I am recommending that the Table 5.4B assumption for 'Computer 
Room' occupancy should be set to 0, and the rules that override occupancy and 
the related calculation of ventilation and SHW need to be revised to make the 
baseline the same as the proposed for this SpcFunc. 

I think the more we think about how models are zoned and developed, there may 
be additional SpcFuncs that we want to classify in this way...  'Electrical, 
Mechanical, Telephone Rooms' are one other that come to mind...

Original comment by da...@360-analytics.com on 10 Jul 2014 at 11:18

GoogleCodeExporter commented 9 years ago 
I agree with your approach.  

Also, I have always been a fan of more space functions.

Original comment by rhedr...@archenergy.com on 11 Jul 2014 at 3:17

GoogleCodeExporter commented 9 years ago 
I don't quite agree with #13, "0 SHW/people/ventilation assumed by rules for 
Computer Room spaces.  If the user specifies them for these spaces, they will 
be the same in the baseline and the proposed."  

SHW in a true data center space will very likely be 0. It will. But if the 
space is defined to include any support areas, then it will likely be low and 
non-zero.  Also, I think occupancy will be very low/ft2, but should be 
prescribed. The dominant load (the servers) should be user-specified, but I 
don't see a reason to allow the user to specify arbitrary occupancy levels 
and/or SHW for this space in particular

This is comment #18!  Should we have a very brief call to resolve this, if 
there are still outstanding issues??

Original comment by JohnJArent on 16 Jul 2014 at 5:58

GoogleCodeExporter commented 9 years ago 
My proposed resolution:

1. Computer room occupancy (3/1000 ft2), SHW load (0.180 gal/h-person) are 
prescribed and cannot be modified by the user.

2. I'm ok with mechanical/electrical rooms not having SHW load, but some large 
buildings with large mechanical penthouses with central plant equipment will 
have SHW. We should set occupancy to zero in these cases. Maybe we can have 
another space type for "mechanical penthouse"? These can be 1000s of ft2.

3. It seems like we may need to handle the "No SHW in Proposed Design" case, as 
discussed.

CHANGING TO RESOLVED -- PLEASE MODIFY STATUS TO "DISCUSSION" IF YOU DON'T AGREE.

In summary:
1. All spaces types that have a non-zero occupancy in App 5.4A have a SHW load.
2. There should be an option for no SHW in proposed.

Original comment by JohnJArent on 31 Jul 2014 at 5:24

GoogleCodeExporter commented 9 years ago 
Resolution:

1. All occupied spaces, with only a couple of exceptions, will have a SHW load. 
Computer rooms will have a SHW load. Exceptions are handled directly in App 
5.4A and include mechanical and electrical closets.

2. Noah is adding language to allow for "No SHW" in the proposed design. In 
this case the standard design will have no SHW as well.  See attached language.

Original comment by JohnJArent on 13 Aug 2014 at 5:36

Attachments:

GoogleCodeExporter commented 9 years ago 
Sent email on 10/15/14. 

For this issue there were several items that came up.  Let me summarize with 
what I believe to be the best solution.

1.  Criteria for CRAC vs. CRAH and central plant system type:   I agree that 
this should be load-based (800,000 W or similar), for ALL spaces of type 
computer room (combined load).  The criteria would be load based (entered by 
user), rather than by floor area for those spaces.
2.  Occupancy, SHW load:  I believe occupancy should be a FIXED low value, 
rather than user defined, and SHW load should be a fixed value, rather than 
user-defined.  It looks like CBECC-Com now allows for user entry of occupancy 
for computer rooms and maybe other space types.  I don’t see any significant 
benefit to allowing user entry of the occupancy level.  Even in David’s case 
of a space that is unoccupied, that small occupancy is going to add a 
negligible amount to the system load. For hot water, I think it’s also best 
to fix hot water loads to be consistent with whatever fixed occupancy is used 
(3 persons/1000 ft2, for example). Are SHW loads user-editable as well?
3.  I think a related, but separate, issue is the ability to allow for a 
proposed design case with no hot water system, for any building.  I believe we 
are handling this, and if this is the case, that exception would override any 
SHW load values at the space level, per my understanding.
4.  I agree with Roger that I would tend to add more space functions, rather 
than extra rules to handle scenarios that are unlikely.  For example, a 
mechanical closet that has some small motors or equipment is much different 
than a mechanical penthouse that has chillers, pumps, fans and facilities 
people, and most likely a bathroom.

User entry of occupancy level wouldn’t be my preferred solution, but I 
don’t feel strongly enough against it to change it at this point, either.  
Please confirm that this matches the current CBECC-Com implementation, and I 
can change the ACM to be consistent.

Original comment by JohnJArent on 15 Oct 2014 at 9:10

GoogleCodeExporter commented 9 years ago 
Changing to FIXED. Please verify the following changes have been made in CBECC, 
per comment #21.

#1: CRAH/CRAC selection:  CRAH is the system type if the computer room process 
load for the building is > 800 kW. See section 5.1.2.

#2: No ACM change. I don't have clear guidance on this one.  

#3: No SHW is allowed.  See section 5.9.1.

Original comment by JohnJArent on 21 Nov 2014 at 3:21

Attachments: