You make a good point, and it also can apply to centrifugal chillers,
which are sometimes most efficient at 60-80% loaded. But you still have
to remember that the client has to pay for the larger equipment, which
costs more, plus larger system components (which cost more) and more
space (which costs more).
You also maake a good point about the system loads 90% of the time,
which raises an interesting question regarding sizing for optimum
operating efficiency. What if we sized certain types of equipment to
meet the most predominant loads within their most efficient capacity
ranges? That is not difficult when using hourly simulation software.
Of course, you have to keep an eye on peak loads and capacities. I
haven't thought this through yet, but I'll bet someone else has.
Thanks for your insightful response. To a large degree I concur. You
encounter various situations out there as a design consultant, and you
need to understand the plans and needs of your client. But surely you
recognize that there are some cases where over sizing is not required,
and the client may actually prefer to pay less up front and get a better
operating system than always be able to maintain set point under the
most extreme combination of conditions. When maintaining perfect
environmental control is critical, then you must oversize accordingly,
but when it isn't, you may assign more priority to proper sizing, or, in
rare cases, under sizing.
I personally would prefer a slightly undersized AC system in my home to
an oversized one if I had to choose between the two. I understand the
advantages, and prefer to risk having to tolerate a little discomfort
for a few hours a year than to pay for and operate an excessively
oversized system for the next 15 to 20 years.
But to offer this to a client may be risky business unless you get him
to sign a waiver, and then it is still risky. On the other hand, if you
can determine the maximum probable system load and add a reasonable
10-25% to that, you will have satisfied the client's needs with a
reasonable margin of safety. The only angry clients I have ever
encountered were angry because their system was failing to deliver a
significant amount of the time when circumstances were only normal, in
which cases the engineer had made a serious error and specified a system
that was grossly undersized. I don't recommend over sizing to make
certain this doesn't happen. Instead I recommend checking and double
checking your sizing calculations so you don't make such a mistake.
This is a hot issue because nearly every HVAC engineer is going to
sooner or later make that proverbial mistake, lose self-confidence, and
from then on join those that oversize their systems more than necessary
just to cover potential mistakes. Most clients are none the wiser
anyhow, and, after it has all been paid for, some actually enjoy the
feeling of knowing they have a "honker of an air conditioner" that is
capable of cooling the place down in 60 seconds or less. So what do we
have to lose? I think it might be some of our own professional and
personal integrity! Not much else comes to mind.
Now I have said too much, so I will bow out of this, knowing that last
paragraph may ruffle some feathers.
I changed the topic name to ?Equipment Sizing?. I thought I would add my 2 cents worth also. So now Bldg-Sim has 8 cents worth.
The first building project that I was associated was at the tail end of the M-E design at JB&B of the Federal Reserve Bank Building in Minneapolis.
http://www.lera.com/projects/ofc/federalreservemn.htm
The building opened in January (I think) in the early 70s. ASHRAE winter design at 1% is -16 F. When the building opened it was about -30 F (it can go down to -40 F there) with strong winds. Occupants in the US are not dressed (with full arm and leg heavy woolen underwear) to tolerate low indoor temps even for one day.
Summer design in Minneapolis at 1% is 92 DB 75 WB but the DB can exceed 100 F. Occupants can tolerate a little discomfort on the few days that extreme summer conditions occur. It?s still better than no AC. Few, if any, buildings were air-conditioned, even in the US, before 1940. The moral of this story is design for extreme winter conditions in very cold climates. Judgment, experience, and common sense have to be applied. It depends on the location. In a place like Singapore the temp varies from a low of 75 F to 95 F all day and all year. There are no extremes. All buildings in tropical countries do not need heating systems.
Inefficient energy use occurs when there is only one unit of the equipment and it is oversized. When there are two or more units, one unit starts until it reaches maximum, then second unit comes on and the two shares load. The units are rarely operated at minimum load. This is the default in DOE2 but LOAD-MANAGEMENT allows you to sequence the use of primary equipment in any way you want that is appropriate for the project.
The lighting and equipment design criteria was 5 watts /sf and 3 watts /sf for buildings designed before the energy crises in 1974. No one cared about energy before then. Actual lighting density was nowhere close, and there was very little equipment in offices. This means for a million sqft building you end up with three chillers. One is standby which comes into operation when a chiller fails or one chiller has to be shut down for maintenance. Specifying standby primary equipment affects first costs and does not affect equipment efficiency.
Energy programs are for comparing alternative energy conservation measures. There is no need to size equipment for occasional extreme weather conditions. But I think auto-sizing is based on weather data (not design data or median weather data) so equipment is sized for the worst condition of that year. This means at every other hour of the year the equipment is operating inefficiently at part load conditions. Equipment sizes estimated by energy programs are used in the design process.
Equipment schedules in design documents are based on actual equipment selected from manufacturers catalogs. The name of the manufacturer and the model number are specified and then ?or equal? added. Equipment schedules are not based on design calculations or generic data because it is possible that real equipment cannot meet the performance data. You need the dimensions of the equipment (with clearances for maintenance) to design them into the space. TRACE energy program is based on actual real equipment performance data specified my model numbers. HAP is based on real equipment, but they don?t mention model numbers.
I think the Code of Federal Regulations 10CFR434 (ENERGY CODE FOR NEW FEDERAL COMMERCIAL AND MULTI-FAMILY HIGH RISE RESIDENTIAL BUILDINGS) used to allow you to size two identical units of primary equipment each at 66% of the maximum design. This would be commercial buildings with large heating and cooling loads where you would have two units. Perhaps it was a special case where failure to perform at all times was not an option. I couldn?t find it the latest register.
http://www.wbdg.org/pdfs/10cfr434.pdf
10CFR434 - 403.2.2 Equipment and System Sizing.
Heating and cooling equipment and systems shall be sized to provide no more than the loads calculated in accordance with subsection 403.2.1. A single piece of equipment providing both heating and cooling must satisfy this provision for one function with the other function sized as small as possible to meet the load, within available equipment options. Exceptions are as follows:
(a) When the equipment selected is the smallest size needed to meet the load within available options of the desired equipment line.
(b) Standby equipment provided with controls and devices that allow such equipment to operate automatically only when the primary equipment is not operating.
(c) Multiple units of the same equipment type with combined capacities exceeding the design load and provided with controls that sequence or otherwise optimally control the operation of each unit based on load.
I fully agree with you, and most of the following diatribe is not meant for you, but for the larger design community. None the less, I invite you to correct me publicly where you believe I may have missed the boat. I'm still learning!
When you can specify more than one cooling unit to meet the same load, you can afford to oversize a lot before energy efficiency is affected. It's mainly the residential and small commercial sectors that concern me regarding excessive over sizing vs. efficiency and humidity control.
But in every case, more capacity costs the client (owner) more money, so proper sizing is important, especially during difficult economic times like these. If you are like me, you tend to take ownership of the HVAC design, and costs may not be as important to you as they are to the client. An excellent design may be rejected by the design team based on cost, and the easiest way to increase cost, along with the probability of rejection, is to oversize the systems.
With regard to sizing heating systems, I believe there is more leeway in terms of equipment (but not floor space or subsystems) costs, and the practical need for greater capacity is more common (morning warm-up after setback, for example). A good design will consider the real capacity needs of the building without exceeding them more than necessary. It's more difficult to define over sizing in the large C&I sector because proper sizing depends on more variables, so it comes back to the design engineer's integrity more often. A highly skilled designer will tend to have more confidence in his ability (unless he has been "burned" a few times due to design errors) to calculate the real loads more accurately, considering all the variables.
One less confident (possibly including a highly skilled professional with burn scars) may tend to assume "conservatively" on all variables simultaneously, thus over-calculating the required loads, and then beef those up more than necessary. Who, except another highly skilled (and brave, and independently wealthy) professional, is capable and willing to challenge this approach? And even then it is one's opinion against another's.
If redundancy is needed (hospitals, for example), then there is still a practical limit to the need. It just becomes more indeterminate, or more difficult to define. This is where interaction with the design team is more critical; this time the HVAC designer needs to ask appropriate questions, listen carefully and offer verbal guidance to the team before he can properly size the systems. We should not fail to consider that installation costs are always greater with larger systems, regardless of building sector. All else being the same, this usually applies to O&M costs as well. The integrity question is really this; "Do I mitigate my personal risk more through willful overdesign, or do I assign more value to the owner's financial objectives and/or limitations?"
To achieve a proper balance, the designer must first determine the true capacity requirements with accuracy and confidence, and then add only a modest oversize factor to that. It isn't easy, but our choices here eventually establish our levels of self-respect and our professional reputations within the design community.
Chris,
You make a good point, and it also can apply to centrifugal chillers,
which are sometimes most efficient at 60-80% loaded. But you still have
to remember that the client has to pay for the larger equipment, which
costs more, plus larger system components (which cost more) and more
space (which costs more).
You also maake a good point about the system loads 90% of the time,
which raises an interesting question regarding sizing for optimum
operating efficiency. What if we sized certain types of equipment to
meet the most predominant loads within their most efficient capacity
ranges? That is not difficult when using hourly simulation software.
Of course, you have to keep an eye on peak loads and capacities. I
haven't thought this through yet, but I'll bet someone else has.
Thanks,
Glenn
I did all that 30 years ago using the Meriwether ESAS program for Hilton Hotels. Before DOE-2. Before DOE-1 !!
John,
Looks like I might have awakened a sleeping giant...been around awhile,
huh? Are you going to tell us how it worked out?
Glenn
Jason,
Thanks for your insightful response. To a large degree I concur. You
encounter various situations out there as a design consultant, and you
need to understand the plans and needs of your client. But surely you
recognize that there are some cases where over sizing is not required,
and the client may actually prefer to pay less up front and get a better
operating system than always be able to maintain set point under the
most extreme combination of conditions. When maintaining perfect
environmental control is critical, then you must oversize accordingly,
but when it isn't, you may assign more priority to proper sizing, or, in
rare cases, under sizing.
I personally would prefer a slightly undersized AC system in my home to
an oversized one if I had to choose between the two. I understand the
advantages, and prefer to risk having to tolerate a little discomfort
for a few hours a year than to pay for and operate an excessively
oversized system for the next 15 to 20 years.
But to offer this to a client may be risky business unless you get him
to sign a waiver, and then it is still risky. On the other hand, if you
can determine the maximum probable system load and add a reasonable
10-25% to that, you will have satisfied the client's needs with a
reasonable margin of safety. The only angry clients I have ever
encountered were angry because their system was failing to deliver a
significant amount of the time when circumstances were only normal, in
which cases the engineer had made a serious error and specified a system
that was grossly undersized. I don't recommend over sizing to make
certain this doesn't happen. Instead I recommend checking and double
checking your sizing calculations so you don't make such a mistake.
This is a hot issue because nearly every HVAC engineer is going to
sooner or later make that proverbial mistake, lose self-confidence, and
from then on join those that oversize their systems more than necessary
just to cover potential mistakes. Most clients are none the wiser
anyhow, and, after it has all been paid for, some actually enjoy the
feeling of knowing they have a "honker of an air conditioner" that is
capable of cooling the place down in 60 seconds or less. So what do we
have to lose? I think it might be some of our own professional and
personal integrity! Not much else comes to mind.
Now I have said too much, so I will bow out of this, knowing that last
paragraph may ruffle some feathers.
Glenn
I changed the topic name to ?Equipment Sizing?. I thought I would add my 2 cents worth also. So now Bldg-Sim has 8 cents worth.
The first building project that I was associated was at the tail end of the M-E design at JB&B of the Federal Reserve Bank Building in Minneapolis.
http://www.lera.com/projects/ofc/federalreservemn.htm
The building opened in January (I think) in the early 70s. ASHRAE winter design at 1% is -16 F. When the building opened it was about -30 F (it can go down to -40 F there) with strong winds. Occupants in the US are not dressed (with full arm and leg heavy woolen underwear) to tolerate low indoor temps even for one day.
Summer design in Minneapolis at 1% is 92 DB 75 WB but the DB can exceed 100 F. Occupants can tolerate a little discomfort on the few days that extreme summer conditions occur. It?s still better than no AC. Few, if any, buildings were air-conditioned, even in the US, before 1940. The moral of this story is design for extreme winter conditions in very cold climates. Judgment, experience, and common sense have to be applied. It depends on the location. In a place like Singapore the temp varies from a low of 75 F to 95 F all day and all year. There are no extremes. All buildings in tropical countries do not need heating systems.
Inefficient energy use occurs when there is only one unit of the equipment and it is oversized. When there are two or more units, one unit starts until it reaches maximum, then second unit comes on and the two shares load. The units are rarely operated at minimum load. This is the default in DOE2 but LOAD-MANAGEMENT allows you to sequence the use of primary equipment in any way you want that is appropriate for the project.
The lighting and equipment design criteria was 5 watts /sf and 3 watts /sf for buildings designed before the energy crises in 1974. No one cared about energy before then. Actual lighting density was nowhere close, and there was very little equipment in offices. This means for a million sqft building you end up with three chillers. One is standby which comes into operation when a chiller fails or one chiller has to be shut down for maintenance. Specifying standby primary equipment affects first costs and does not affect equipment efficiency.
Energy programs are for comparing alternative energy conservation measures. There is no need to size equipment for occasional extreme weather conditions. But I think auto-sizing is based on weather data (not design data or median weather data) so equipment is sized for the worst condition of that year. This means at every other hour of the year the equipment is operating inefficiently at part load conditions. Equipment sizes estimated by energy programs are used in the design process.
Equipment schedules in design documents are based on actual equipment selected from manufacturers catalogs. The name of the manufacturer and the model number are specified and then ?or equal? added. Equipment schedules are not based on design calculations or generic data because it is possible that real equipment cannot meet the performance data. You need the dimensions of the equipment (with clearances for maintenance) to design them into the space. TRACE energy program is based on actual real equipment performance data specified my model numbers. HAP is based on real equipment, but they don?t mention model numbers.
I think the Code of Federal Regulations 10CFR434 (ENERGY CODE FOR NEW FEDERAL COMMERCIAL AND MULTI-FAMILY HIGH RISE RESIDENTIAL BUILDINGS) used to allow you to size two identical units of primary equipment each at 66% of the maximum design. This would be commercial buildings with large heating and cooling loads where you would have two units. Perhaps it was a special case where failure to perform at all times was not an option. I couldn?t find it the latest register.
http://www.wbdg.org/pdfs/10cfr434.pdf
10CFR434 - 403.2.2 Equipment and System Sizing.
Heating and cooling equipment and systems shall be sized to provide no more than the loads calculated in accordance with subsection 403.2.1. A single piece of equipment providing both heating and cooling must satisfy this provision for one function with the other function sized as small as possible to meet the load, within available equipment options. Exceptions are as follows:
(a) When the equipment selected is the smallest size needed to meet the load within available options of the desired equipment line.
(b) Standby equipment provided with controls and devices that allow such equipment to operate automatically only when the primary equipment is not operating.
(c) Multiple units of the same equipment type with combined capacities exceeding the design load and provided with controls that sequence or otherwise optimally control the operation of each unit based on load.
Varkie,
I fully agree with you, and most of the following diatribe is not meant for you, but for the larger design community. None the less, I invite you to correct me publicly where you believe I may have missed the boat. I'm still learning!
When you can specify more than one cooling unit to meet the same load, you can afford to oversize a lot before energy efficiency is affected. It's mainly the residential and small commercial sectors that concern me regarding excessive over sizing vs. efficiency and humidity control.
But in every case, more capacity costs the client (owner) more money, so proper sizing is important, especially during difficult economic times like these. If you are like me, you tend to take ownership of the HVAC design, and costs may not be as important to you as they are to the client. An excellent design may be rejected by the design team based on cost, and the easiest way to increase cost, along with the probability of rejection, is to oversize the systems.
With regard to sizing heating systems, I believe there is more leeway in terms of equipment (but not floor space or subsystems) costs, and the practical need for greater capacity is more common (morning warm-up after setback, for example). A good design will consider the real capacity needs of the building without exceeding them more than necessary. It's more difficult to define over sizing in the large C&I sector because proper sizing depends on more variables, so it comes back to the design engineer's integrity more often. A highly skilled designer will tend to have more confidence in his ability (unless he has been "burned" a few times due to design errors) to calculate the real loads more accurately, considering all the variables.
One less confident (possibly including a highly skilled professional with burn scars) may tend to assume "conservatively" on all variables simultaneously, thus over-calculating the required loads, and then beef those up more than necessary. Who, except another highly skilled (and brave, and independently wealthy) professional, is capable and willing to challenge this approach? And even then it is one's opinion against another's.
If redundancy is needed (hospitals, for example), then there is still a practical limit to the need. It just becomes more indeterminate, or more difficult to define. This is where interaction with the design team is more critical; this time the HVAC designer needs to ask appropriate questions, listen carefully and offer verbal guidance to the team before he can properly size the systems. We should not fail to consider that installation costs are always greater with larger systems, regardless of building sector. All else being the same, this usually applies to O&M costs as well. The integrity question is really this; "Do I mitigate my personal risk more through willful overdesign, or do I assign more value to the owner's financial objectives and/or limitations?"
To achieve a proper balance, the designer must first determine the true capacity requirements with accuracy and confidence, and then add only a modest oversize factor to that. It isn't easy, but our choices here eventually establish our levels of self-respect and our professional reputations within the design community.
Glenn