🚀 go-pugleaf

On Wed, 02 Jul 2003 19:52:33 GMT, nobody@uunet.ca (Ralph) wrote in 
news:3f0337d3.245537603@news2.istop.com:

> On Wed, 02 Jul 2003 07:38:26 -0400, Tegger
> <teggeratistopdotcom@changetheobvious.invalid> wrote:
> 
>>Ralph wrote:
>>> I'd like to convert CO V&H coords to Lat/Long.  I found a V&H map
>>> online, but it's a low-res gif so it is not useful for precise
>>> conversion.  I've found references to lots of commercial software that
>>> will do it, but I've never seen a formula to convert it.
>>> 
>>> Even if I could find the lat/long of the 4 corners of the V&H map I
>>> could interpolate the rest.  Any pointers?
> [...]
>>
>>Is this the gif you encountered?
>>http://www.colorado.edu/geography/gcraft/notes/coordsys/gif/vandh.gif
> 
> Close, but the one I found was in B&W.
> 
>>Does this help?
>>http://www.ngs.noaa.gov/PC_PROD/pc_prod.shtml#UTMS
> 
> Hmmm... maybe.  I'll check it out if I can get it to download.

After looking at the things turned up in a search for Donald Elliptic
Projection, I've found out enough to understand what the V&H is...

Since the grid is not north/south oriented and covers quite a large
area (all of North America) there's little hope of getting any
accuracy from simply trying to register or calibrate the image
to anything resembling lat/long.

It seems the formula is a bit of hocus pocus designed in 1957 to
guesstimate long-distance rates based on geographical distance.
Aparently the innards of the calculations are not reverse-engineering
friendly. In other words, nobody seems to know how the constants
for the equations were derived. Which means you need the equations...

There's a java class described at:

http://openmap.bbn.com/doc/api/com/bbn/openmap/util/VHTransform.html

to do conversions just as you request.
It seems the equations appear at the bottom in the toLatLon methods.