Actually, it's your fault.
You should always keep attention on where exactly your input goes. It is indicated by underscore, so if you see that you "selected" not your current block, but whole formula, you should use arrows to go to required block prior to entering next part.

But here, SMath could make it apparent showing bracckets...

SS-2331 filed.

List2.sm (12kb) downloaded 27 time(s).

Their purpose is to compensate the difference between equatorial and pole radiuses of Earth, and displace of your arc from Earth major corcumference. Generally, you measure the distance not over equator, nor ober meridian, but over some arc that is both displaced and rotated. You need to know its radius to be able to measure its length.

But anyway, their source is unclear, and they do give wrong results.
You may simply take average radius of Earth to het descent results.

Edited by user 27 April 2016 04:26:14(UTC)  | Reason: Not specified

A "corrupted region" is by the software, not by you. By you: just incorrect.
"Corrupted rgion" creates by hemselves, most annoying. Aslo, sometimes "corrupted region"
doctor by themselves. If you supect "corrupted region", recreate on blank sheet from A..Z.
There is no way to detect the corruption. It may just be within some of the primary
function as simple as missed some coefficients in sin(x), for instance.
Corrupted regions did happen in Mathcad 11 [rarely], mostly on long work sheet, even more
annoying. The wrost case: the backup is not up to date and nothing to recreate .

Moscow-Kiev2.sm (15kb) downloaded 28 time(s).

Thanks Mike:

Vicenty sanity check very closely with some on-line calculator [5060/5045]
Haversine NASA is quite in eccess 5915 km.

Cheers, Jean.