revit-ifc icon indicating copy to clipboard operation
revit-ifc copied to clipboard

PR: When exporting, scale parameter is missing in georeferencing

Open Moult opened this issue 1 year ago • 5 comments

Problem Description

Here is a blank project in metric millimeters as the project unit. The coordinates at the PBP is 10,000mm and 20,000mm.

image

The projected CRS is EPSG:7856.

image

Revit will export this data in IFC: Project2 - Copy.txt

#19=IFCSIUNIT(*,.LENGTHUNIT.,.MILLI.,.METRE.);
#97=IFCUNITASSIGNMENT((#19,#21,#22,#26,#27,#30,#31,#33,#34,#35,#37,#40,#42,#46,#47,#48,#49,#50,#51,#52,#53,#54,#55,#56,#61,#64,#66,#69,#70,#71,#72,#73,#74,#75,#76,#77,#79,#80,#81,#82,#83,#84,#85,#86,#88,#89,#91,#92,#94,#95,#96));
...
#105=IFCSIUNIT(*,.LENGTHUNIT.,$,.METRE.);
#106=IFCPROJECTEDCRS('EPSG:7856','Map Grid of Australia Zone 56, GDA2020 (7Parameter)','GDA2020-7P',$,$,$,#105);
#107=IFCMAPCONVERSION(#100,#106,20.000000000000004,10.000000000000002,0.,1.,6.1230317691118863E-17,$);
...

As you can see:

  1. Project units are in mm (good)
  2. Projected CRS defined (good)
  3. Projected CRS has a map unit of meters (sure)
  4. Map conversion has an easting and northings of 20 and 10 (in meters, good :) it matches the map unit which is great!)
  5. Map conversion has a null value for Scale (bug)

IFC states:

The scale factor can be used when the length unit for the 3 axes of the map coordinate system are not identical with the length unit established for this project (see IfcProject.UnitsInContext), if omitted, the scale factor 1.0 is assumed.

and repeats itself

Scale to be used, when the units of the CRS are not identical to the units of the engineering coordinate system. If omited, the value of 1.0 is assumed.

on the docs for IfcMapConversion

Because it is omitted, the value of 1.0 is assumed. But the units of the CRS (m) are not identical to the project units (mm). So this is not correct - there needs to be a value for scale.

According to this equation as you can see first we scale the project units XYZ then we add the eastings/northings. So the scale should be equal to 0.001 (to go from mm -> m, then we can add the eastings/northings in m).

(ignore the FactorXYZ that is for IfcMapConversionScaled but the general equation still applies)

image

Revit Version

2023.1.x

IFC for Revit Addon Version

24.x.x

Windows Version

11 22H2

Moult avatar Jul 01 '24 13:07 Moult

This looks like a likely oversight to add this scale - will add issue to address this shortly.

AngelVelezSosa avatar Jul 03 '24 13:07 AngelVelezSosa

This looks like a likely oversight to add this scale - will add issue to address this shortly.

Thank you, this is a real problem to address properly georreferencing.

enruilo1 avatar Jul 31 '24 15:07 enruilo1

Hello @Moult,

This is considered a confirmed bug that has been addressed and is no longer reproducible in recent versions of the IFC exporter.

We have confirmed that the issue is resolved in Revit 2025 and 2026. The Scale parameter is now correctly exported when project units (e.g., millimeters) differ from CRS units (e.g., meters).

We are closing this issue as resolved.

Thank you for your contribution to improving the IFC exporter!

Testing environment:

  • Autodesk Revit 2025 (IFC 25.4.3.0)
  • Autodesk Revit 2026 (IFC 26.3.0.0)
Image

ArtemHnatko avatar Aug 22 '25 12:08 ArtemHnatko

Hi @Moult,

Thank you for the clarification. We understand your concern that the Scale parameter's implementation might be more complex than a simple bug fix and requires specialized geodetic knowledge.

From our side, the bug report for the missing Scale parameter has been addressed, as the field is now present in the exported IFC file.

To move forward and ensure the exported data meets your expectations, could you provide us with a detailed description of the expected behavior? This would greatly help us understand the geodetic requirements and allow us to investigate the specific issue with the parameter's value.

We believe that with your expertise, we can define the problem more precisely and work towards a correct solution. Thanks.

ArtemHnatko avatar Aug 26 '25 13:08 ArtemHnatko

Gday, I think someone else posted under the name CEM-EnriqueRuiz, not me. Then he deleted his comment. I havent made any further comment on this or had a chance to review the fix, sorry.

Dion Moult

Sent from Proton Mail Android

-------- Original Message -------- On 26/8/25 11:20 pm, ArtemHnatko wrote:

ArtemHnatko left a comment (Autodesk/revit-ifc#784)

Hi @.***(https://github.com/Moult),

Thank you for the clarification. We understand your concern that the Scale parameter's implementation might be more complex than a simple bug fix and requires specialized geodetic knowledge.

From our side, the bug report for the missing Scale parameter has been addressed, as the field is now present in the exported IFC file.

To move forward and ensure the exported data meets your expectations, could you provide us with a detailed description of the expected behavior? This would greatly help us understand the geodetic requirements and allow us to investigate the specific issue with the parameter's value.

We believe that with your expertise, we can define the problem more precisely and work towards a correct solution. Thanks.

— Reply to this email directly, view it on GitHub, or unsubscribe. You are receiving this because you were mentioned.Message ID: @.***>

Moult avatar Aug 26 '25 13:08 Moult