The situation:

Enter the geographic location of a known point (CORS BLWY L1)in AS using Geodetic, North American 1983-CONUS, Ellipsoid Elevations, Meters, as Settings. Fix both horizontal and vertical.

N = 42 46 01.63661
E = 109 33 28.03606
Eh = 2216.456 (m)

Go back to Settings and change to Grid, State Plane Coordinate 1983, Wyoming (West) Zone, Orthometric Elevations, US Feet.

AS now reports the position as...

N = 1154586.728'
E = 2765782.970'

Corpscon 5.11.08 gives the following...

N = 1154586.727'
E = 2765782.970'

Essentially the same. But if we try the same thing in State Plane Coordinate 1927, the results are more troubling. AS reports the position as:

N = 765798.932'
E = 641307.683'

Whereas Corpscon gives us...

N = 765799.067'
E = 641307.708'

A difference of 0.137 feet. Why? And which one is "right?"

Skylar

Skylar...

Please let me know what "AS" is, as I have always used CORPSCON.

I have checked CORPSCON a hunred times with conversion software I have written, as well as with other software, and have never found an error.

I suspect that the AS software is using a different set of equations and constants. There have been more than one method published in the past. Prior to about 1984, coordinate positions were calculated in a different manner, although they should yield positions within 0.03 feet state plane.

First, let's agree that any conversion between NAD83 and NAD27 is an approximation of the truth but we do have a referee available in the form of the NGS. Use the Geodetic tool kit on the NGS web page to do the conversion. There is your official answer.

As for the NAD83 conversion, what is 0.001 foot among friends?

Both of us have used many versions of software and have seen (and probably been bitten) by incorrect parameters when doing translations...
Use the NGS programs, and then you can at least document the results...IE, NGS did it, not me!At least use them to check your work..Most of their programs now have online version where you can input via internet and get the results without downloading any software..


You'll never get good survey results going from NAD-27 to NAD-83...Not to say that a good NAD-27 survey will not have good relative positions, just not good overall fit to the NSRS...

If you have a good tight NAD-27 survey it's probably best to hold one point and using a NSRS tie, rotate and translate to this single point. This will maintain the integrity of your survey, but will have a small error in the absolute NAD-83 reference frame...


Which one is right? There both right, but one is based upon a straight transformation(probably the AS solution) while the other is based on modeled error in the system (how many ties, how well did they fit, what is the relationship to NAD-83 points). The NGS program is the best 'to get you close', but when the NGS says "mapping grade", then you can count on it...

Sky...Copy your post and send it to whoever makes AS...They should be aware of the difference and if it's a software glitch, then they should fix it, but most likely it's not and they can give you the best explanation of why you are seeing a difference..

I do give you good marks on 'checking' and questioning the outputs...After being burned many times by software output, it's the only sane thing to do..

Trimble Man wrote:

Sky...Copy your post and send it to whoever makes AS

He has, in effect, already done that. Unless I'm mistaken, AS is Ashtech Solutions, the processing and adjustment software provided with Ashtech GPS receivers.


Modified By Jim Frame on 10/20/2001 at 10:07 AM

As you can see by my login name..I am receiver challenged, as they say...Please use my shorthand TM when replying to me...Don't want to get the natives restless..
We have many common problems with GPS and I've picked up some good tips from this side of the fence...

Sincerely,
Deral Paulk, PLS OK

Yes, TM is correct. I was referring to Ashtech Solutions 2.4. I understand that NAD27 isn't the best projection to work with, but we have an awful lot of control in 27, and our county GIS is all NAD27. I just wondered why Ashtech Solutions didn't match the results of Corpscon for NAD27 positions. Thanks.

Skylar

Skylar,

For surveying purposes, I would never rely on conversion between NAD27 and NAD83. If you have to work on NAD27, perform the adjustment on that datum, constraining the NAD27 control. This can be a little tricky, because there are some significant distortions in '27.

Regards,
Brian

Without knowing how Ashtech is handling each of the conversions/transformations, it's not easy to give a good answer. CORPSCON uses various utilities written by NGS. That being said, neither computation is correct. In your process you failed to account for the fact that the CORS position is consitent with the HARN - NAD 83 (1993) in Wyoming - and in CORPSCON you must treat it as an NAD 83 HPGN position. The proper transformed values for NAD 27 should be N = 765796.978', E=641309.267' Assuming that ASHTECH is running NADCON properly, I would suspect that they have somehow truncated the algorithm that converts lat/long to State Plane on NAD 27. To echo what others have said -- DON'T USE NAD 27!!! Having your GIS linked to NAD 27 would be link having all their computer software in DOS! Increasing, issues of homeland security are being linked to quality spatial data and NAD 27 does not provide that framework. I realize I'm probably preaching to the choir.

Dave,

Thanks for your response. I didn't realize my CORS station was on HARN. (Good thing I haven't done any work where I had to provide state plane positions.) CORPSCON (single point conversion) asks for Latitude, Longitude, and Elevation. What elevation is it looking for? Ortho height? Ellipsoid height?

Finally, there are two positions given for the CORS station in question (BLWY); the ARP position, and the L1 Phase Center position. Since Locus units are L1 only, my impression was that we should use the L1 Phase Center position, but I have nothing to base this on, really. Do you know which one I should use?

Skylar Wilson

Either one will do, but using the ARP requires that your software knows the offset from the ARP to the phase center. Using the phase center position allows you to use an antenna height of zero.

P.S. As I understand it -- an important caveat -- the L1 phase center position is provided because an L1 fixed solution is as good as it gets. Dual-frequency receivers allow for faster integer fixing and calculation (as opposed to modeling) of ionospheric delay, but both single- and dual-frequency users are striving for an L1 fixed solution.

Doesn't matter what you use (arp or phase), just as long as you correctly input which one you measure to..
The software just needs the correct one to apply the appropriate height adjustment.

TM

Thanks, Jim and TM,

What you say makes sense.

Skylar