Local Magnetic Variation

[trombe]

Hi folks, a query please :
Elevations and plan views in question particularly.
Advanced Model Graphics - Cast Shadows - Sun and Shadow Settings - By Date, Time and Place - Place - Place tab - select location of latitude and longitude.

I am trying to decide how to modify the settings to take account of the local magnetic variation.
If setting up these location settings in the rendering dialogue under Sun etc. , you are able to adjust for the local variation after settling on your location, however, this is not so in Advanced Model Graphics.

Is this module taking any note of the previously set relationship between Project North , and True North ? when it casts shadows for (a solar study) no matter if elevation or plan is used as the view to cast shadows onto.

If it is not, can I assume that there is no provision for local magnetic variation for a given latitude and longitude ( here, there is no set up for any rendering, just using for sun shading)
For my location ( for example) , Wellington, NZ is approximately 23 degrees East (magnetic) of True North.

Hope this is a clear enough explanation.

thanks for any help as local help is not forthcoming .
regards
trombe

[dbaldacchino]

I understand there is a difference between magnetic bearings and "real" ones, depending on where you are (for instance in my area, the compass south would be about 6 to 7 degrees off from solar south) but if you have the true longitude/latitude information for your project, doesn't that bypass the problem?

[trombe]

Hi David,

I don't think so / am not sure about that.
There is a difference of something like....1200 km ? ? ? between the geographical north pole and magnetic north pole. The local latitude and longitude settings define a place relative to the originally determined above 2 places, and then local alterations are made to take account of the fluctuations of the magnetic field at that place (Variation = declination),
Deviation is something else altogether.
I just cannot find anything in documentation to clarify and state specifically either way or anything about this important condition.

All that can be said really seems to be that your project north and true north setting, is reflected in the Advanced Model Graphics setting as ("internal" on the Tab) OR, as you might choose.
While you might set your true north (in your project / site plan) from a surveyor dataset and that might provide for this already, other methods can require adjustments to take account of this local variation.

Obviously I cannot afford to have a solar study off by 23 degrees, so I may be missing something else. Unfortunately, my local support has not replied to repeated queries.
I think I am going to have to seek clarification direct from AutoDesk on this one.

Thanks anyway David B
regards
trombe

[dbaldacchino]

With regards to the orientation, I believe you are correct. I have no clue whether Revit is using Solar North or magnetic North. And if you're trying to orient a sundial, it makes a LOT of difference.

With regards to location though, I think the latitude/longitude numbers are absolute, geometric coordinates....slicing the globe in perfectly sized regions, so magnetic versus non-magnetic doesn't come into the equation. Orientation is another problem though. So you could locate your project on the globe perfectly using latitude/longitude figures, but then you don't know whether Revit's north is the same as your survey and how far off you really are. Since the relationship between magnetic north and solar north vary depending on where you are on the earth's surface, I would suspect Revit is using the "geometric" north (where all longitude circles cross each other). So my guess would be that you need to make corrections due to this...in my case I'd need to rotate the building by about 6 degrees (cannot remember which way!).

I sure hope someone from the factory would give us a clarification on this, or if you hear something from Support, please post back.

[dbaldacchino]

I just thought of something this morning. I don't have time to try right now, but maybe someone can.

Place a column and run solar studies and take snapshots at noon. If the shadow is exactly vertical in a perfect N-S orientation at any time of the year, then Revit is using Solar North/South. If it varies to either side when you change dates (summer-winter dates), then you can rule out Solar North, but wouldn't be able to tell if it's adjusting for magnetic variations.

[trombe]

closure on this issue probably.
Also to David B.....
have not got to AD yet and probably won't as the support request would have to be (morally) rated at lowest priority and would probably never get answered...am still waiting for earlier information return from RB 9.0 so...

I think that since the lattitude and longitude lines are struck from the N & S poles and not from the magnetic poles, the location / position given by the lattitude and longitude values in the Advanced Model Graphics for a place, are relative to solar / true north , thus, local magnetic variation does not come into it as its already corrected sp to speak.
I should probably have known this !!
How this relates to options within the Accurender settings I have yet to get to.
regards
trombe

[dbaldacchino]

Thanks for replying. That's what I was thinking. But here's where it would be really important to know what the magnetic variation is...

A couple of years ago I worked on a school and we integrated a meridional, vertical sundial on the interior....it's actually a reversed sundial as the entry is from the north, so we came up with a way to cast the shadow on a translucent screen, reversed the graphics and still achieved a functioning sundial. Anyway, if something like this is simulated in Revit, you would need to know what the variation between solar North/South and magnetic North/South is, because otherwise you'd be off (about 6 degrees in my case) and thus, it wouldn't be accurate. So you'd need the surveyor to give you this info. Conversely, if Revit corrected for magnetic variation, then we wouldn't need to bother the surveyor for that info. :)

[Henry D]

Related to this topic: Here is a link to a nice simple Geomagnetic variation calculator...just enter the zip code and it will give the precise variation for the time and place. The Magnetic Pole keeps shifting a bit from year to year.
The type of Architecture I do requires that the buildings be precisely aligned with True North.. we don't use a compass to get True North, but rather get the surveyor to do a Sun Shot

http://www.ngdc.noaa.gov/seg/geomag/jsp/Declination.jsp

[tomnewsom]

If your surveyor is using a compass to set his North, then sack your surveyor. You should never need to know magnetic north when designing a building, unless it has a decorative compass in it.

[Alex Page]

Interesting link to revit city forum- looks like you have to work out the 'true north' compared to the 'magnectic north' if you want your shadows to be correct.. this scares the sh*t out of me, since I have done Resource consent shadow diagrams to get council permissions to build

http://www.revitcity.com/forums.php?action=viewthread&thread_id=4865.


have a read

[Mike Sealander]

It seems the gist of the RevitCity thread was that Revit uses true north, and that several people had surveys that were using magnetic north, and therefor they had to adjust for declination.

[tomnewsom]

It seems the gist of the RevitCity thread was that Revit uses true north, and that several people had surveys that were using magnetic north, and therefor they had to adjust for declination.
That is mighty strange. I can honestly say I've never come across a magnetic north survey.

[Mike Sealander]

It is strange, but I have a survey I'm working with right now done to magnetic, not true. It's old, and my feeling is that pre-gps, magnetic was fairly popular.

[tomnewsom]

If it's that old, magnetic north has probably shifted since then :) It's moving quite quickly at the moment, IIRC.

[Mike Sealander]

Yes, I went online and tracked the declination for my coordinates (about 68 N 44W) for the month in question.

[ron.sanpedro]

closure on this issue probably.
Also to David B.....
have not got to AD yet and probably won't as the support request would have to be (morally) rated at lowest priority and would probably never get answered...am still waiting for earlier information return from RB 9.0 so...

I think that since the lattitude and longitude lines are struck from the N & S poles and not from the magnetic poles, the location / position given by the lattitude and longitude values in the Advanced Model Graphics for a place, are relative to solar / true north , thus, local magnetic variation does not come into it as its already corrected sp to speak.
I should probably have known this !!
How this relates to options within the Accurender settings I have yet to get to.
regards
trombe

My worry is, even if this is true, there are still times you still need a magnetic north, with accurate declination, to go with project north (the paper) and true north (the sun). When a new building is being laid out on a rural site, with no existing roads and such, it is a compass and magnetic north that is used. And if your declination is wrong, then the layout is wrong, and the orientation to the sun is wrong, and your passive solar design is wrong, along with orientation of solar panels and such. I would love to see Revit add another flavor of North, with the ability to define the magnetic declination. Until then you really need to use a bit of a kludge to get it all to work right.

Gordon

[dbaldacchino]

As long as a survey is based on True North, then any solar calcs. will be just fine. And Revit's True North is Solar North with no account for declination. Once you know what you're working with, it should be easy to make adjustments.

[trombe]

As long as a survey is based on True North, then any solar calcs. will be just fine. And Revit's True North is Solar North with no account for declination. Once you know what you're working with, it should be easy to make adjustments.

Hi Dave B,

So are you saying that your True North settings which were done with latitude and longitude, for your (eg) Site Plan for instance while correct for site plan uses, need to have the (attached image) option adjusted for rendering and solar analysis ?
Is this the correct use for this ? (the help is somewhat vague I thought)
If so this brings me back to the original question and scary answer that all of my solar shadow work to date, is either about 23 degrees out for Wellington and 19 degrees out for Auckland work and no local authority processing officer picked it up , nor did anyone else ?
As Alex Page says #$!!&*
Say it isn't so ?

regards
trombe

[dbaldacchino]

I'm not sure about that dialog, I'll have to research it a bit. I believe that the True North that Revit uses is the Solar North (ie: point of highest altitude during any one day). Now I know that that point during the day actually varies also on a day to day basis due to Earth's wobble around the axis, but am not sure if Revit takes that into account at all. This topic will make a good SR question, I might file one to see what we can get out of it.

[ita]

Trombe, latitude and longitude have no effect on the North bearing. All you can do is stand on that point (which is simply a position on the surface of planet Earth) and then you can face any direction that you want – rotating through 360deg – latitude and longitude stays the same. All Latitude and Longitude give you is a position so you can determine you angle of solar inclination at any time of the day - so even with that even Longitude becomes irrelevant – apart from determining the time of day!



If your survey bearing is magnetic (and I request the surveyor nominate the method of determining the North point) you will also need the date and time of when that bearing was taken. Go to a web site like Geoscience Australia and you can find the declination tables for that location, date and time.



Take note. Make sure you instruct your surveyor that you require notification of what the North bearing is – True North, Magnetic North or North based on a national mapping grid – which is often what surveyors provide in Oz – some even give you North from the Road Directory!! Note that up to 10deg east or west of True North does not have a great deal of impact on energy calcs – but it will have an impact on the modelling of sun screening – so determine what the basis is that you are dealing with.



That will allow you to calculate the variation from true North of the magnetic bearing provided. Then that is the direction that you rotate your model to in the Element Properties>Orientation> . . and select True North view. You can now orient your model to true North and your modelling will be correct. The 23deg that show on the Sun and Sky setting says that that your model in the Revit North view is oriented 23 deg west of north from true north.



So unless the North Elevation is parallel with the top of your monitor in the True North view, your setting is probably correct if the north Elevation is in fact 23deg West of North. If that is not how it is meant to be, then recalculate the Normal of your north elevation and adjust its bearing to the True North Axis.



Now for all that there are other issues to resolve – not so much for you in New Zealand – but here in Western Australia for instance, Western Standard Time in Oz is based on the Longitude of Kalgoorlie (not Perth) some 500km to the east of Perth – so in Perth we are in fact (time wise) ahead of the sun’s real position, while in Melbourne they are somewhat closer to the mean position for Eastern Standard Time.



In terms of shadow studies, that all does not mean very much; but in terms of heat analysis on the envelope of your project and comfort levels within your project – it is significant. Particularly if you are endeavouring to determine max heat gain of loss or internal comfort levels etc. That is where applications like Ecotect are so invaluable because not only do they do the shadow analysis and demonstrate the design decisions far better than Revit, but the modelling of heat loads is a significant design tool before you need to finally model the impact of solar radiation on your building.



Hope that makes it a little clearer – probably its as clear as mud!

[truevis]

The best minds at MIT have trouble figuring the exact alignment of sun and existing building.

Article on MIThenge : http://web.mit.edu/planning/www/mithenge.html

[dbaldacchino]

Haha, nice one! I feel better about myself now ;) We are the not-so-advanced civilization, since there are so many examples of perfectly aligned, solar-conscious monuments out there built thousands of years ago with limited tools but probably much better brains.

[trombe]

Hi Ian thanks,
yip, clear as mud.

trombe

[dbaldacchino]

Trombe, this is weird...my Sun and Sky settings dialog does not look like yours. I don't have that option to enter a rotation there. What version are you running?

[Mike Sealander]

David:
I don't have it either, and I can't enter lat-long coordinates directly. When I do, the numbers bounce back to the city showing up in the dialogue box. I was hoping someone could tell me what I'm missing, because the online help says that if you enter coordinates directly, the location changes to User Defined, as opposed to one of the out-of-the-box cities.

[ejburrell67787]

I'm running the current build of Revit 9.1 and I can see the same dialgue boxes that Trombe posted.

- Setting user-defined lat and long values is under advanced model graphics -> Sun and Shadow settings -> still --> Manage Place and locations -> Place. Ignore the city and entre the co-ordinates and time zone

- Setting the North angle etc is under Render Scene Settings - but you have to un-check the box that says "Use Sun and Shadow settings from the view" to get access to the detailed Sun and Sky Settings.

:beer:

[Mike Sealander]

Elrond:
I'm running 9.1.
I think you can't set lat-long coordinates under advanced model graphics. I can set them under Render Scene... by creating a new scene. However, that new scene's location is not available under AMG. No matter how many times I try, AMG only allows changes in cities.
The Render Scene... dialog also gives me the ability to see when the sun is due south, with the azimuth indicator, which can act as a check on the accuracy of one's true north orientation.

[ejburrell67787]

In AMG, have you tried entering the lat and long co-ordinates without clicking in the city drop-down? Once you have entered the figures and clicked out of that section the city sets itself to User-defined. Maybe you have to save a new location first though? Can't remember if I did that first or not. I did try and overtype the city name a few times before I realised you can't do that though.

[dbaldacchino]

Thanks Elrond, found it thanks to you. I have not experimented much with rendering so I'm not very familiar with these settings.

[MikeJarosz]

On the original Freedom Tower (before the redesign) the television broadcasters needed the exact geographic location of the tower. After some massaging and manipulation of shared coordinates, we got a number that we submitted to them. They came back and wanted it in latitude and longitude. This is not something I do everyday. One of the engineers at the Port Authority of NYNJ said they had a contract with the Corps of Engineers for global information, and our site survey was in NAD83. Figured I was a phonecall away from an answer, so I emailed them the coordinates.

I got back a question: At what altitude?

Damn engineers want everything.

[dbaldacchino]

Well, the z coordinate tends to be quite important ;)

[k.armstrong]

Well here's something to confuse me

I thought that at midday local time to where you are (regardless of where you are on the earth) the sun always was DUE NORTH and a shadow would fall due south (just differing lengths depending on the season)

so here's my little solar study at midday for a year

!!

[dbaldacchino]

Nice! What you plotted is called an Analemma (http://images.google.com/imgres?imgurl=http://www.mysundial.ca/tsp/images/eot_curve.gif&imgrefurl=http://www.mysundial.ca/tsp/time.html&h=604&w=550&sz=40&tbnid=YTcOf4MlNCfbSM:&tbnh=135&tbnw=123&prev=/images%3Fq%3Dequation%2Bof%2Btime&start=3&sa=X&oi=images&ct=image&cd=3): a graphical plot with the Equation of Time on one axis and the sun's declination on the other.