We have been using multiple coordinate systems for a number of years, and it seems to work perfectly.
We mix and match at will between lat-long, NZMG and NZTM (ie old and new datums) within the same cave survey and between cave surveys.
So, Footleg, unless you have other reasons to do so, there is no need to convert your old datum coords to new datum coords.
 
There seem to be a couple of rules to keep things working smoothly.  Have a look at Therion Book pg 14 and 46-47 and keep in mind the distinction between input and output. 
  1. Each centerline with fix statement(s) should have a date and cs specification (prior to the fix I think).
  2. So that you can be sure of the cs applied to the outputs, specify the output cs that you want in the th-config (not in a layout).
  3. I have avoided using cs in layout, as, even if Therion is consistent in the way it applies it, in the situation where there are multiple coordinate systems in use, it is not abundantly clear to a user which cs will be used, for example where some centrelines have no explicit cs or for the output.  (Although reading Therion Book pg 47, maybe that is for Atlas outputs - been a while, I forget).
Now that we have output variables in 5.3.15 that echo the cs that has been used for output, it will be possible to have this stamped in the pdf header.  I have a 'proof that it works' layout drafted up, and one day I will make a modular set of layouts to make it readily useable
 
The first file embedded on this page http://therion.speleo.sk/wiki/doku.php/templates
(ie http://therion.speleo.sk/wiki/lib/exe/fetch.php/templates:thconfig-templatesimple.txt)
shows how I use it in a th-config file.
 
Bruce

From: therion-bounces@speleo.sk [mailto:therion-bounces@speleo.sk] On Behalf Of Footleg
Sent: Saturday, 20 September 2014 4:13 a.m.
To: List for Therion users
Subject: Re: [Therion] Combining caves using different co-ordinate systems

I had put one cave using a cs and gps fix coordinates into one survey block in a .th file. Then a second cave using a different cs and gps fix coordinates in a second .th file.

My project then uses a master.th file which inputs both of these. If I run my config on either of the individual files then they appear at the correct place on Google Earth from an output kml file. If I include both caves then they appear in different places. The one I include first is in the correct place. If I swap the order they appear in the master.th file then this confirms the first one is alway put in the correct place, and the second is positioned using the cs of the first one.

The same in a lox model file (except I cannot confirm which is correct, but I see them in different places when I give them the same entrance location to test this).

Footleg

On 19 Sep 2014 14:33, "Andrew Atkinson" <andrew@wotcc.org.uk> wrote:

You can use cs in config file, either inside layout, or outside layout (most likely what you want to do) to control the output co-ordinate system, which I think is what you are trying to do?

Andrew

On 19 Sep 2014 14:00, "Footleg" <drfootleg@gmail.com> wrote:
I have been converting a project from an old datum to a more recent
one following the issuing of new maps using the newer datum. So all my
entrance GPS coordinates changed. I updated the cs statements in my
survey data files and my KML output is still in the correct place. So
all was looking good. Then I discovered that if I include a survey in
the old cs with one using the new cs then Therion appears to treat
both caves as having the same cs. Whichever cs is used by the first
included cave is applied to subsequent caves too. Is there a way to
deal with this?

Footleg
_______________________________________________
Therion mailing list
Therion@speleo.sk
http://mailman.speleo.sk/mailman/listinfo/therion

_______________________________________________
Therion mailing list
Therion@speleo.sk
http://mailman.speleo.sk/mailman/listinfo/therion