Revit Systems Analysis - Incorrect Outside Temperatures, Humidity and unrealistic psychrometrics at time of peak condition

[eng58HND]

Hello everyone,

I hope you are doing well.

I'm wondering if anyone can help regarding the Revit Systems Analysis toolkit. I appear to get inaccurate peak outside temperature conditions at Time of Peak. Namely, a 0.0 C Dry Bulb and a corresponding Wet Bulb of 32.9 C, the design psychrometry for this space, particularly the outdoor air relative humidity is also not realistic, I would expect somewhere in the region of 60-70% not 28.5%, it is Cape Town not the Sahara desert. I have seen multiple posts on this forum related to other issues around Systems Analysis where I notice their temperature values report more realistically and I would like to get assistance with my issue, which I believe is the last piece of the puzzle that stops me from progressing further with the Systems Analysis tool.

I am trying to perform a basic analysis on a simple room with a single space to test this issue.

I have done the following steps on Revit 2024.2:

1. Created an arbitrary room 5.7 x 4 x 3.1 m in size consisting of a 300 mm generic floor set to level 0, 200 mm generic wall with its base at level 0 and top at level 1 with no offsets and a simple flat roof the same size as the floor. Default construction sets were used. There are no overhangs present.
2. Assigned a space within this room with extents at level 0 and up to level 1, occupiable and condition type set to heated and cooled. No other settings were adjusted.
3. Two water loops, one chilled water and the other hot water were created and assigned to a Four Pipe Fan Coil zone equipment unit and set to one per space. A system zone line was drawn across this space and assigned to the zone equipment.
4. Location was set to Cape Town, South Africa (WMO# 688160), all other energy settings were kept as default, this includes default building types, space types and resolutions.
5. Analytical Surfaces schedule was consulted and no air surface types were found, indicating an airtight model.
6. HVAC Systems Loads and Sizing workflow was selected and run.

I have attached my eplusout.err file to this post in hopes that I am missing something obvious but I suspect the problem lies either with my copy of Revit (how else would it explain other users getting reasonable outside design temperatures?) or a connection issue to the weather data servers.

I will assist in any way to get to the bottom of this issue.

Kind Regards,

Tazden

[Onur_Bozkurt]

I have a similar problem

Although the dry bulb temperature is taken correctly from the station (it appears correct in the detailed openstudio results) it shows 0 in the loads report.

[iainsavage · Accepted solution]

ASHRAE cooling design data for Cape Town is 31.2C db, 19.3 wb (or at least it was in the 2013 of Fundamentals and I don't have access to more up to date versions) and this would put the RH in the region of 30%

The Sahara would be nearer 10 to 15% RH

It looks like the calculation is using the correct data based on the bottom table in your screenshot but maybe there is a presentational error in the summary which is pulling the design DB into the WB field and leaving the DB field blank??

This should be raised with Autodesk via the product feedback or by submitting a support request - its the only way that they will know about it and fix it.

[iainsavage]

Per response to @eng58HND it looks like the summary is showing Dry Bulb in the Wet Bulb field and leaving the Dry Bulb field blank, BUT it is using the correct data in the calculation??

Alert Autodesk to this via product feedback or by a support request so that they can take corrective action.

[eng58HND]

Hi @ianinsavage, thanks for the response. I will submit a support ticket to Autodesk regarding this.

Apologies if I am discussing it on the wrong forum but does the ASHRAE annual cooling temperatures (DB and WB) take into account the relative humidities from the Monthly Climatic Wet Bulb and MCDB design conditions? I'm aware that the RH is 30% but in practice it is often much higher than shown in the annual conditions. I'm mainly concerned about latent loads not being fully captured.

[iainsavage]

Firstly apologies for the lecture and please just skip over any bits that you already know.

I don't intend to insult your intelligence or education but you did ask!

Even if its not helpful to you it might help someone someday:

There will be days/months when you might have higher DB and there may be days/months when you have higher WB but the "design day" is based on statistical analysis of DB & WB conditions which occur simultaneously and therefore give the greatest enthalpy values. That is what the MCWB tag means, Mean Coincident Wet Bulb.

There is then a built in acceptable failure rate - in the case of cooling design usually taken as 0.4% - so 0.4% of the hours over the whole year might exceed that design condition in a "typical" year.

0.4% is 35 hours, probably not sequential, spread over the whole year. For human comfort that is unlikely to be significant (other failure rates are also quoted for less critical cooling applications and extreme values are quoted for more critical (process) cooling applications).

When you talk about high humidity are you meaning relative humidity or absolute humidity?

Your Absolute humidity or moisture content (kg of absorbed water vapour per kg of dry air) is likely to be high because of your warm maritime climate.

The Relative humidity (ratio of how much water is actually in the air versus the maximum value which could be absorbed at that dry bulb temperature) varies with DB so even though you may have a high Absolute humidity (moisture content) the Relative humidity could be quite low due to the extreme DB.

You need to plot the coincident DB and WB values on the psychrometric chart to see the RH values and lower down the page for Cape Town you will see design values for each month so you should be able to track the RH over the year.

The sensible load on your cooling coil will be the difference between DB outside and DB required in the supply air to offset sensible gains in the room/zone.

The latent load on your cooling/dehum coil will be the difference between the Absolute humidity outdoors and the Absolute humidity required in the supply air to offset latent gains in the room/zone. The Relative humidity is not used in this calculation, it is only an indication of how "dry" the air feels for human comfort and how much the evaporation rate from skin is likely to be.

I managed to find the 2017 ASHRAE weather station data so I've attached the page for Cape Town.

Hope this helps.

[iainsavage]

Just as a brain teaser for me I plotted a rough psychrometric chart based on the monthly values.

See attached.

It seems that your peak dehum rate is not actually very large - only 0.001 kg/kg, and that for much of the year you would actually require humidification rather than dehum.

It also seems that the absolute humidity in your zone is almost precisely the same as that for the supply air condition which would imply that you have near zero latent gain in the zone - does this seem reasonable to you? Have you accounted for all gains including people and infiltration air which might add moisture to the spaces?

[eng58HND]

Hi @iainsavage,

Thanks for the wonderful explanation, I'm starting to appreciate the world of HVAC more and more. 🙂

As per your recent snip on the chart for Cape Town, I submit an additional use case for a more extreme climate (closer to the equator) for discussion. As it would have it, both Cameroon and Cape Town have February as their hottest Annual Cooling design condition.

Plotting both the Annual design point and corresponding Monthly Design Wet Bulb + MCDB data for February results in the graphs attached. If I look at the enthalpies, I get the following, which is supported by your own analysis, considering the two design points are fairly close from an enthalpy point of view:

For Cape Town:

Monthly Design Wet Bulb Condition: 58.9 kJ/kg

Annual Design Cooling Point: 54.8 kJ/kg

Indoor Set Point (22 C @ 50 RH): 43.3 kJ/kg

For Cameroon:

Monthly Design Wet Bulb Condition: 81.1 kJ/kg (20.13 g/kg Abs. H)

Annual Design Cooling Point: 67.16 kJ/kg (13.57 g/kg Abs. H)

Indoor Set Point (22 C @ 50 RH): 43.3 kJ/kg (8.33 g/kg Abs. H)

Using Q=mc(Tf-Ti) as a reference, I would expect the overall cooling load to be higher in the monthly design condition versus the annual cooling design point? Below is the tabulated data from the monthly design wet bulb section of the ASHRAE weather report if it helps.

Perhaps I am approaching this from the wrong angle?

Kind Regards,

Tazden

[iainsavage]

You can't use Q=mcΔT to assess load on a cooling coil unless your load is completely sensible load.

The correct method is to use Q=mΔh where Δh is enthalpy difference between the air entering the coil and the air leaving the coil. That way you take account of both the sensible and the latent components.

The entering coil condition will be the outdoor condition if you are using 100% outdoor air but it will be less than that value if you are recirculating a percentage of room air. 

The leaving coil condition must be lower than the room condition in order to offset gains in the room.

Again please don't take offence if I'm telling you stuff that you already know.

I live in a temperate climate in UK (station 031400) where we tend to favour temperature control rather than humidity control (except for critical applications) so we would not normally use design conditions biased towards maximum WB, although in our case it makes little difference, but if that is the practice in Africa then the chart that I did earlier maybe understates the required dehum.

[eng58HND]

Hi @ianinsavage,

Thank you for the explanation, I had erroneously used Q=mcΔT when I meant Q=mΔh to refer to the overall difference in enthalpy as a way to gauge load on coil as a combined sensible + latent.

If we fit this discussion now into the context of the Revit System Analysis (EnergyPlus engine, not the older Heating and Cooling Loads), the analysis engine uses the ASHRAE design conditions as you've stated and is shown in the report, barring the mixed up on the DB/WB values used in the report.

Summarizing our discussion, the weather data values used in the Systems Analysis tool appear to be biased towards more temperate climates where the enthalpy difference between Annual conditions and monthly WB conditions are minimized. Conversely in more extreme situations closer to the equator where there is a significant latent load present due the high humidity, this difference is exaggerated, and the results become less reliable.

As it stands now, unless we get the ability to manually edit the weather input data such as is the case with the older load est. method, the only way I can see us adapting the System Analysis tool for humid climates is to manually design and select a dehumidifier that will move the design point on the psychrometric chart from the Monthly Wet Bulb and MCBD value to the Annual Cooling Design Point and then run the simulations assuming all outdoor fresh air entering the building is mechanically driven from central fresh air handling units with dehumidifiers.

This is a reasonable approach?

Kind Regards,
Tazden

[iainsavage]

To be honest I'm not sure how you would manipulate the data to get the results that you want.

They were supposed to be reintroducing the ability to edit weather data and I understood that it had been done:

https://www.autodesk.com/support/technical/article/caas/sfdcarticles/sfdcarticles/Weather-tab-is-mis...

[eng58HND]

Hi @iainsavage,

Apologies for the back-and-forth!

You are correct. As you are aware (I'm explaining for the sake of the others that might not be aware) there are two methods of performing HVAC load estimation within Revit, See Below:

By default, Option 1 (Based on old ASHRAE Fundamentals) was hidden from the menu along with the Weather Tab in the Location Window that allowed for manual adjustment of DB, WB and Daily Range values. This was reenabled but only shown through editing of the .ini files. (See Revit Heating and Cooling Loads Revit 2022/2023/2024 – Ripple (rippleengineeringsoftware.com))

This produced more reasonable results as you could override the weather data values to account for the higher WB temperatures, but this method only reported peak loads for a month, day and hour and not a diversified load across the year. Output data is fairly limited and only good for peak capacity sizing.

Option 2 is the new shiny method based on the EnergyPlus engine that prints out the nice .html report with full EnergyPlus data that shows basically everything you could need. The reports can also be customized if you are tech-savvy enough to code in your own workflow. (Of which, I do not count myself amongst)

The only issue is that despite weather data being reenabled, from my experiments, Option 2 does not use these manual values, it will always default to the weather data obtained from the Location tab (Internet Mapping Service) despite editing the weather tab. This presents a larger problem as these weather data values are often decades old and not indicative of current climatic conditions, even if you ignore the high humidity issue we were exploring in the previous posts.

One more thing I will add to the support ticket, I suppose. 🙂

Thanks for your help @iainsavage, much appreciated!

Kind Regards,

Tazden

[iainsavage]

I wasn't aware that systems analysis didn't use the same weather file as the older Heating & Cooling Loads tool but @Kevin.Lawson.PE knows more about this stuff than I do.

Regarding " these weather data values are often decades old and not indicative of current climatic conditions", as far as I can see the data that you are using corresponds to the ASHRAE Fundamentals weather station data which should as far as I know be updated every time the Fundamentals chapter is reissued (every four or five years?).

The data is however smoothed by averaging over something like a 20 year period so that it is not unduly affected by short term events such as sunspots, solar flares and the El Nino which is happening just now  etc.

I also assume that ASHRAE get their data from NOAA and other similar organisations so there won't, I don't think, be anything better out there unless you know differently.

Hopefully you are making some progress with your calculations and can figure out a way to get the results that you need and it might be worth reaching out to @Kevin.Lawson.PE for a bit of help - he has also created some free apps on the Ripple Engineering website which are intended to improve upon the built-in calculation tools (but I've never used them so you would need to explore them yourself).

[Kevin.Lawson.PE]

Thanks for the shout out @iainsavage 

For the classic 'Heating and Cooling Loads' tool, you can override weather in the Weather tab of the location dialog box. 

For the new Systems Analysis tool, you can write custom measures to use whatever weather files you want: https://www.autodesk.com/autodesk-university/class/Revit-Systems-Analysis-Features-and-Framework-Cre...