First of all - this will be more of a theoretical question than LB/HB/E+/OS related, I hope you do not mind.
Secondly - I am an architect, not an engineer (please bear with me, I have limited knowledge on the topic but would like to change that). I use pretty basic models in the early stage of design. I do that (through Ideal Air Loads + set points) to see how different facade configurations affect overall predicted heating/cooling load + peak loads.
We got these latent loads values from our co-working team of engineers for core&shell office building:
lightning: 15W/m2 equipment: 35W/m2 people - 5m2/person (@72W per person)
I would like to note here that they usually size equipment using steady-state model - outdoor temp+humidity+indoor loads and solar gains. Last ones are calculated by glazing g-value and size (so, for example, room orientation does not matter). These are all used for sizing of equipment.
I am not sure about that but from what I found/read (also in HB components ) these seem relatively high (although I am aware these values can vary A LOT). The thing is - we usually design for multi-teanants with no assumption for layout they would use.
I would like to ask you a couple of questions here:
I ran a couple of test models with provided (mentioned above) loads and what I realised is that Cooling Sizing Value are often very different (mostly higher) than what is calculated by engineers through steady-state model. I assume it is due to different method of calculating solar gains - am I right?
I am not trying to be smarter than engineers - I am sure their sizing calculation is right and provided according to local law. I just would like to be able to provide models of variants that are at least somewhat close to correctness. That would save a lot of time on both sides. But - as my results are very different - should I use different loads in that case?
In HB I can set number of people per area, but cannot see any settings on what is the load per person. I know that typical values are 75-125 W/person. Can it be adjusted anywhere in HB? If not - do you know what is the value used by E+ for calculations?
Could you provide me with any reliable source of knowledge where I could find “typical” values for indoor loads (lightning, people, equipment)? What values would you use to calculate likely cooling/heating loads + sizing predictions?
I’d be very grateful if you could share your knowledge and insights on these topics.
Hi,@Wujo I am a HVAC engineer from China. As a architect, you have done awesome job for the calculation of heating/cooling load + peak loads.Here is my answer for your 4 questions.
Q1: You should post more information for it, such as cooling load from you and engineer.
Q2: You shoud tell me which variable maked the difference.
Q3: You can use schedule to set the person load. This input is for setting of the activity level in Unit of W/m2.
Few examples or comparison - we used same rooms/zones (also areas). Few examples of peak cooling loads calculated:
S orientation:
engineer: 4,8kW
mine: 4,8kW
SW orientation:
engineer: 6,2kW
mine: 7,3 kW
W orientation
engineer: 4,8kW
mine: 5,4 kW
I am not sure if I understood you correctly. Do you mean what I think may cause the differences? My guess is that solar loads may affect the values most (maybe temperature from epw. as well as it may be different for peak loads than assumed in steady-state 30 celsius) , as interior loads are the same.
Thanks, exactly what i was looking for!
I thought about ASHRAE standards, I must admit (although I never dug into them). I am located in Europe, unfortunately i could not find similar standards in my country so far.
The thing is - I cannot really relate to ASHRAE Zones because of that. I guess I could compare weather files to find similarities and try to relate to that?
For further discussion, you can compare the detailed load such as exterior wall load/ windows conduction load/windows solar load/. You will find the difference between you and engineer.
You should tell us which country you are . Or you can ask the engineer for the local standards.
That is all. Hope my answer can help you.
Thanks for giving such good answers @minggangyin . I wanted to make a few observations this might be helpful:
You say that you are using “latent loads” at the top of the post but I am nearly certain that you mean “sensible loads”. Just make sure you are modeling the right type load fr everything your are modeling.
Those loads that you are using for lighting, equipment, and people are definitely on the higher side compared to most programs. They are within the realm of possibility but those equipment loads are more characteristic of a hospital, computer room, or a restaurant/kitchen than they are of an office or residence. Those lights would also appear to be incandescent or they are illuminating a palace where high lighting levels are needed (like a surgical room in a hospital).
I would ask your engineers how they are accounting for thermal lag if their calculations are steady state. In E+, you are doing a transient calculation that accounts for the thermal mass of the floor and other surfaces and how they dampen the peak over the day as you see here. Your engineers likely have some way that they are accounting for this and, with a steady state calculation, they often just multiply it by a factor (or they are using some software to relate the heat capacity of the floor to a thermal lag factor). I would venture a guess that this it probably where a lot of the divergence in calculation methods happen since you can often get the other loads to match (eg. lighting, equipment, and even solar). But matching transient thermal lag to steady stat thermal lag is tough.
The ASHRAE 90.1 standard that @minggangyin cites is the origin of a lot of international building energy codes. If you are in Europe, I would venture a guess that you are using some form of the IECC codes and these codes tend to draw a lot of their values from the ASHRAE 90.1 standard. Generally, the two codes tend to move in lockstep and the following versions of ASHRAE 90.1 can be related to a corresponding version of IECC like so:
ASHRAE 90.1 2016 | IECC 2018
ASHRAE 90.1 2013 | IECC 2015
ASHRAE 90.1 2010 | IECC 2012
ASHRAE 90.1 2007 | IECC 2009
ASHRAE 90.1 2004 | IECC 2006
To provide some more background details on why I started this thread:
Engineers plan using cooling beams, which, as I was instructed, are less efficient than fan coils when it comes to cooling, but it is a tradeoff for acoustic comfort and better heating (?) efficiency. That is why we were provided with some max. cooling loads from their calculations. Than we tried to meet them by manipulating the layout of the facade (WWR, shading, etc.)
I tried calculating that through HB/E+. To my surprise - I got values that were much higher (in some zones) compared to engineers results.
Unfortunately, now any sort of communication with engineers (as with everyone, due to virus) is very limited but I will for sure dig into what standards/methods they use and spend some time with them to understand each other better.
My next step will be checking how mentioned components (wall and window conduction and window solar load) influence the results. I already know how to do that (i think) as I am friends with hydrashare for quite a while .If I encounter any problems I would reach out to you, If you do not mind.
You are right! I used sensible loads of course (double checked, phew!). In my native language we use other order of words - “sensible and latent” instead of “latent and sensible” that, as far as i know, is usually used in english. That is the source of my mistake, but simulation was fine! .
I thought so. My guess is that engineers are being super-careful, as we design it as core&shell - this would be a multi-tenant building - we do not know any layouts (for example - where tenant would locate conference room where people loads would be higher for short periods of time) or what kind of lightning would be used. But, for sure, this wouldn’t be anything else than office space.
It would be strange if incadescent lights would be assumed though, I wouldn’t think anyone used these in an office space nowadays. I am also pretty sure it is not a hospital building :).
As for equipment load - I found some information that one pc+monitor is about 150W. I think, that - assuming 5m2/person gives 1 pc each 5m2. So PC load (to be more specific) would be 150W/5m2 = 30W/m2. I think this might be the source of such high value provided by engineers.
Unfortunately (as mentioned earlier) I cannot confirm it now.
In the end I decided to run another set of analysis, changing especially people loads (as architects we design toilets and elevators, for example. to be sufficient for 8m2/person in such buildings).
I will compare results soon. It is a 90-zone model with 5 different variants of facade + I calculate sDA and ASE (btw. thank you so much for your example on hydrashare!) for each run so it takes quite a while!
Just to be sure - do you mean thermal lag (caused by thermal mass of materials) affects cooling load (needed to cool a space to a set point) to be lower than peak cooling loads calculated by E+? (sorry for that sentence, didn’t know how to rephrase it to be more understandable)
I am familiar with this example you posted but now I am not sure I understand it right. I’d be grateful for explatnation: Do the two graphs in the file illustrate phenomena you mentioned? I can see that Peak cooling load value, is in fact higher than Cooling demand (as I understand it - the whole zone was pre-cooled in previous “time-steps” and themal lag causes cooling demand to be actually lower than peak load would suggest?) So - in theory I would need smaller-sized cooling equipment than peak cooling loads would suggest? Also - I cannot see that in the file itself, but in the pictures on hydrashare there is a third graph for “Cooling Load” and its values are lower than “Cooling demand” graph - I am confused
I am ashamed to say that, but I’ve never heared of these. I also did not find single Polish page that would mention IECC, but I will defienately take a look at these. I will consult which I should use with engineers (when we would be able to talk for some time)!
Sorry for the late response. I’ll try to answer the questions that you highlighted in bold:
do you mean thermal lag (caused by thermal mass of materials) affects cooling load (needed to cool a space to a set point) to be lower than peak cooling loads calculated by E+?
Thermal lag affects the cooling demand that needs to be delivered by the HVAC to keep the space at the setpoint. EnergyPlus accounts for this thermal lag in it’s HVAC sizing calculation because the thermal mass of constructions is specified in the IDF. However, you can also just sum all of the loads in the space over the design day to get the instantaneous peak load, which (because it is instantaneous) does not account for thermal lag.
Do the two graphs in the file illustrate phenomena you mentioned?
Yes.
So - in theory I would need smaller-sized cooling equipment than peak cooling loads would suggest?
I’m just saying that, if you are using the sum of the instantaneous peak load to size the system, this would be very conservative and you likely don’t need a system that big. It’s also possible that you and the engineer are both accounting for thermal lag but you two are assuming different amounts of thermal mass in the space.
) these seem relatively high (although I am aware these values can vary A LOT). The thing is - we usually design for multi-teanants with no assumption for layout they would use.
.If I encounter any problems I would reach out to you, If you do not mind.