Suddenly moving ACS

Hello!

We have a problem with ABD's Floor defined ACSs. So far it have occurred three times in three different models/projects.

The thing is that we have been working on the model ,everything going OK, but suddenly (and for me it seems that without reason) the origin of the ACSs jumps up 188.788 metres as shown on attached image. I can't even imagine where this number comes from, for the Ground Floor's Elevation is 188.140m. The GF ACS should sit exactly on the top plane of the slab below the house.

I've checked BB_FloorMaster.dgnlib and it seems OK. I've compared the "malfunctioning" model with other ones, but can't find anything.

Does anyone know the reason for this or any suggestions on what could be done?

Csaba Szabó

  • The first thing I would check is that the floormaster.dgnlib and the 3D models have the same working units and resolution.   We've come across similar situations before. but they're not always reproducible and therefore not always able to be resolved.    Any kind of pattern you can identify would be helpful.



  • Hi Steve!

    I've checked the design file settings of the two files (image attached). The resolution is the same, only the system differs - don't know why, both should be metric. We worked on two separate model files and both have the same problem. Is there something that can change resolution's system during working on a file? A seed or reference perhaps?

    I've set the model's resolution from imperial back to metric, but didn't change anything.

    Csaba

  • Hmmm....  it looks as though someone had taken an imperial file and changed the Working Units to metric, but not the resolution.  However, since the Working Units are the same and the Resolutions are equivalent, this may not be the cause.   I'm guessing all of the 3D models have the mixed metric/imperial settings, and not just the problem file(s)?

    In regards to changing the resolution, you may be able to simply change it to metric since you're not really changing the solids resolution per se, but rather its display value - mm vs. FT.   However, I'm not sure how the DataGroup values for existing components will react; e.g., solids that were 3 units changing to 914 units (FT to mm).

    But back to the original issue...  do you have side by side cases where one file wants to place elements 188.788 metres higher than others, while both pointing to the same BB_FLOORMASTER.DGNLIB file?



  • We have three model files, all of them behaving the same while having the same settings (same floormaster, metric working units, imperial resolution).

    Out of curiosity I opened masterproject.dgnlib since it had a modification date matching that of the model file. It has imperial working units and resolution. Actually I've no idea what this file does, so can it be the cause?

    I'm asking because after this I checked it for the project I'm working on (which had no problems so far) and it is imperial also. Then I opened the model file and all of the elevations (relative and "absolute") in the Floor Selector were ~3,28 times bigger than they should (eg.: 1st floor relative: 12,959m in stead of 3,95m) with the problematic ACS of course. Opening Floor Manager settled them back to normal.

  • AFAIK the masterproject.dgnlib is only (or at least mainly) used for link sets; e.g., Project Explorer.  However, I'm not really following the logic of what happens (in the background) when opening masterproject.dgnlib and then a DGN file and seeing the floor elevations temporarily 3.28X larger - the FT > Meter conversion factor - and then those same elevations returning to normal after opening Floor Manager.   Obviously it happened, but why I can't really explain off the top of my head.   I'd have to think on that one... for a while...   :P

    But what I would add is that the dataset files should all have matching units, and those units should be consistent.  Using a mixed set amongst various files is simply not something that would be tested in-house, so it's difficult at best to guess what might happen.  I would have almost bet on the result being nothing whatsoever.   But then there are the results you mentioned above, which shoots down that theory!

    So going back to the original issue...  after the above steps, are you still seeing the incremental difference in elevations?