Energy Analysis - HVAC loads

[paul.soler]

Hi,

I created a small house with:

- 1 floor

- 4 walls

- 1 roof (2 sides)

Every elements has thermal properties.

When I create the Energy Analytical Model, I can see all the surface with different type in surfaces analytics and the room/space. Strange thing is that in the report I only have wall and roof thermal lost but nothing for the floor.. Why ?

I updated OpenStudio with the latest version.

Thanks,

[iainsavage]

I'm not up to date with how the analysis now works but, for reasons I could never understand, it used to ignore floor losses for floors in contact with the exterior/ground.

The OpenStudio EnergyPlus literature from their website explains in more depth than the Revit help topics.

[paul.soler]

Strange behavior knowing that 10% of heat losses come from the floor of a house..

Anyway I don't know if it's a good way of doing thing but I create a crawling space under the house, set the floor with the parameter "outside", set space with a type "crawling space" and energy analysis using rooms (with space I got lower heat loss).

Now in the report for a room of 55m² with a floor U = 0.22 W/m²K, I got -146watts.

If I do the math manually I got 55 x 0.22 x (21 - 5) = 193.6 watts.

I make the assumption that the temperature inside the crawling space will not drop below 5°c when it's -3 outside.

Not the same result as revit but maybe it's taking more parameters into account idk 😕

[iainsavage]

Firstly I think you are now getting a floor loss because the floor is adjacent to another space and is not in direct contact with the ground or external environment. So its still not properly calculating floor loss for a ground floor.

Secondly I don't think you can do a simplistic comparison using the manual steady-state equation against the heat balance methodology used in the analysis software. You're also relying on your guesstimate of a temperature of 5C in the void but what if its actually 4C or 6C?

Thirdly, and very importantly, heatloss from ground floors does not in any case follow the simple Q=UAdeltaT equation because most of the loss is horizontally through the perimeter of the floor (if in direct contact with the ground) or the perimeter of the void (if the floor is suspended). It is fundamentally different from loss through a wall, window or roof which is in direct contact with outside air over all of its exposed area. Manual methods require a modified U-value to be used taking account of the perimeter to area ratio and loss through a floor of a given area will always be less than for a wall of the same area for that reason. 

That said a lot has been written on this forum about the heat load and energy analysis calculations in Revit most of it not in a positive sense.

Do some searching and see what has been discussed before.

Revit uses the US Dept of Energy approved EnergyPlus software as its calculation engine, however that software is only as good as the data passed to it by Revit and having trawled through the various documents on the EnergyPlus website some years ago I came to the conclusion that EP is very comprehensive and should be as accurate as you're likely to get (in fact I think several expensive commercial applications use this freely downloadable software as their calculation engine). In my opinion therefore the problems lie with the way that Revit collates data to send to EP and the results which it then extracts from EP.

You could do a comparative test by exporting gbxml data from Revit and then importing it into the full version of EP rather than using the built-in command line interface version, however users have also queried the gbxml data which is output from Revit so this test is perhaps not 100% reliable either.

You could also as a comparative test (if I remember correctly) manually input data to EP so that it is processing raw data and not reliant on the output from Revit.

Hope this helps.

[iainsavage]

Oh and you might also want to look at some free plugins developed by @Kevin.Lawson.PE of Ripple Engineering

[KoosU]

I have went the same route as you and have tried Revit on Multiple buildings to see if I can get any confidence in the results.

As of yet, I have not. I tried both methods Revit offers for Heat Load Calculations and had to hack my Revit 2024 to get the older option back.

Conclusion: I could not get results from Revit that gave me anything near the confidence of Carrier HAP 5.

It seems OK if you want a fair bulk value very early in the project, but as a specialist HVAC designer, I would not put my neck on the block with this.

The problem is they went too complex and missed the essence of what heat loads is about. In this regards, Carrier HAP 5 stays king.

[RLY_15]

In a similar boat. My only dislike of HAP as a product (the 5.x version specifically) is the limited compatibility with GBXML files; limiting each space to 8 exposure definitions ultimately results in having to rebuild your wall/window definitions. But otherwise, every component of the load calculation procedure is very well-documented. Whereas in Revit it takes a fair amount of poking around to parse the parameters that were used in the deprecated (but not gone) Heating&Cooling Loads tool vs Analytical Surface Energy Analysis vs Analytical Systems analysis, etc. etc... I think we're about due for someone at AU to present the methodologies in a concise 'Day In the Life' format. 

[KoosU]

The documentation on HAP5 is the key reason I use it.

There calculations and assumption as well documented and the engineer can verify how they get to the answer. From there he can decide if and how he wants to compensate for it.

My one concern with Revit is exactly that: That we have little reference in how calculations are done.

The heat load portion apparently use EnergyPlus. We are familiar with that through Designbuilder. Designbuildier is a great piece of software, but it is very difficult to work with as it has a million options and is not so well organised.

In the end, as engineers we do not have a lot of time for design and we need a good answer with simple inputs and Carrier HAP does just that.

[paul.soler]

I agree with you it's ideal but I was hoping to get something from Revit to help me size my heating floor.

But with what you told me it looks like I will have to use another software...

Before posting I search a lot on Google and in the forum and since many posts was "old" I was hoping this was fix/new features in the new revit release !

Sad to see that Autodesk don't want to invest more time into this.

I will try exporting GBXML data and see if I can have something with either energyplus or other software.

Not as quick as I was hoping ^^

Thank you for your reply 🙂

[iainsavage]

"since many posts was "old" I was hoping this was fix/new features in the new revit release"

There have been "improvements" and the method recommended by ADSK now is their System Analysis method which is supposed to give more accurate results.

You can create a very simple system for each space using a simple electric heater as the system so it doesn't need to be complicated.

The older Heating & Cooling Loads tool was retired in version 2021 (I think) but then brought back in response to user requests but as far as I know it is no longer fully supported or developed.

I've not personally used the Systems Analysis method much because by the time they sorted out the processor compatibility problem I had retired.

Did you investigate the addins available from Ripple Engineering? Users have reported that they greatly improve Revit's built-in methods and @Kevin.Lawson.PE who developed them has been an active participant on this forum in relation to the energy analysis methods within Revit.

[KoosU]

I often struggle to get output from the latest heat load calculator. It will go on and work for many minutes just to return an empty report with no reason as to why it failed. Very frustrating.

The retired version gave arguably some better output. I yet have to find a solution where I can calculate a fairly reliable heat load for a space, so that I can pick up that value with Dynamo to to assign flow values to diffusers.

At the moment though, I am still manually calculating typical loads for area types and have modified my diffusers to calculate airflows based on spacing and my set deltaT. This already helps very much as that values feeds into the duct, and the fancoil, from where I make a calculation of water flow required.

Usually we opt for some conservative values on the floors, so a floor will typically have 50%-100% more water than required at that time based on the pipe sizing, and 50% more on air.

This have helped a huge load when all of a sudden an open plan office layout is changed to a huge training room - or the gazing that was selected as performance in the beginning, is reduced to something inferior. So, load picks up, I change W/m2 on the permiter diffusers and everything recalcualtes.