[Therion] A puzzle in Loch

Bill Gee bgee at campercaver.net
Wed Aug 16 15:20:09 CEST 2023


Wow, thanks Tarquin!  And Ollie, too.  I tried a few things, but the 
spike still exists in the .lox file.

I changed the connecting shots from 3 to BU1 and BU2 from splay to 
duplicate.  After recompiling, the spike is still there.

Next I change the shot from BU1 to BU2 so that it has no LRUD data. 
Recompile, and the spike is still there.  Hmmm....  I think I need to 
change most of the splay shots to duplicate, but that seems to not be 
the reason for this anomaly.

I double-verified the LRUD data for stations BU3, BU4 and BU5.  The 
passage makes a sharp turn at BU4, and a bad left measurement would go 
about the same direction as the spike.  The LRUD data for those stations 
is all correct.

One thing that strikes me is that the spike is not horizontal.  It 
angles upward by about 20 degrees.  If it were the result of bad LRUD 
interpretation, I would expect it to be horizontal.

I included the data line in my extract.  It is indeed the same line as 
you deduced by looking at the data.  Therion does produce warnings about 
"forwards and backwards bearing readings do not match".  There are about 
20 of these in the therion.log file.  I have checked every one against 
the original notes.  It really is off.  Most are under ten degrees.  The 
worst is 17 degrees off.

Most of this cave was surveyed with DistoX2 devices.  The compass and 
inclination readings go to two decimal places (0.01 degree). I do not 
believe for a New York minute that the DistoX2 is really that accurate. 
I suspect it is no better than a Suunto.  The readings in the book are 
rounded to the nearest tenth, and I count on Therion averaging the two 
shots to get a somewhat better consensus.

In this cave, especially around the tour trail, compass readings are 
difficult because there is so much metal.  Hand rails, electric lighting 
fixtures, rebar in the concrete, support pillars and framing for the 
overlook platform ...  It is a magnetic mess!  We did the best we could, 
but there is just a lot of noise in the data.

Some time ago I asked about reverse tape readings.  That is a fairly new 
feature in Therion.  For this cave we did not record any reverse 
distance data.  The distances are in Imperial measurements to the 
nearest 0.1 foot.  The DistoX2 displays to the nearest 0.01 foot and we 
round that to 0.1 before writing it down.

I found a photo of the area we are looking at.

https://campercaver.net/MiscFiles/P1100940-Small.jpg

This shows the stairs leading up to the overlook.  Station 3 is just out 
of sight on the left of the photo.  BU1 and BU2 are out of the photo at 
upper right.  The BU side passage is reached by getting onto the rock 
ledge on the right, then coming toward the camera.  The passage goes off 
to the right of this photo.  There is an electric light fixture just 
about dead center in the photo.  If you levitate about 8 feet up from 
that fixture and look right, that is the passage entrance.

===============
Bill Gee

On 8/16/23 02:18, Tarquin Wilton-Jones via Therion wrote:
>> Hmmm.....   Maybe I am not using the splay flag correctly?  Is there
>> another flag that might be more appropriate?
> 
> flags duplicate = a leg that must not count towards the length. It can
> count towards the vertical range, if it is the highest or lowest point
> in the cave. Use this when you have had to survey twice (or more) down
> the same passage. Also use it if you are going from a station in a
> passage whose length was already included with the legs going down the
> passage itself, and the new legs are used to reach the start of a side
> passage, where the main passage is really wide (such as a station on the
> left wall of a 50m wide passage, and you need to get to a side passage
> on the right wall). This is particularly useful when you have a side
> passage you did not have a useful fixed station for, and the nearest
> fixed station is a long way down the passage, so you will need to
> resurvey from that fixed station, back down the passage you already
> surveyed, to reach the side passage. As Olly said, this is the one you want.
> 
> #survey from fixed station to AB side passage
> flags duplicate
> megacairn AB1 12.34 27.5 1.5
> AB1 AB2 8.97 42.93 -21.5 #AB2 is start of side passage
> flags not duplicate
> 
> flags splay = a shot from a station towards a wall, ceiling, floor or
> other solid object, such as a stalagmite. Therion uses these to build
> the walls in the .lox file. Normally, you use anonymous splays like this
> ("-" instead of a station name):
> AB1 - 1.234 12.5 -17.5
> However, sometimes *rarely* you might want to name a splay, such as a
> splay that hits something important like a named formation. Imagine you
> have a stalagmite called "Medusa", you might do this to tell Therion "I
> know this looks like a station, but it's a named splay":
> 
> flags splay
> AB1 medusa 1.234 12.5 -17.5
> flags not splay
> 
> flags surface = a leg or splay that is on the surface, not in the cave.
> This will not be used for length, depth, or wall generation. It
> basically tells Therion to ignore it for cave statistics, and for wall
> generation. This might be used for the legs that go from the cave
> entrance to your fixed location on the surface, or for splays that go
> from a surface station to a surface feature, such as a cliff face
> outside the cave.
> 
> station 2 "Alpha Cave" entrance
> flags surface
> fixedpoint 1 12.345 27.5 1.5
> #splays around the outside of the entrance
> 1	-	0.555	237.50	-37.67	
> 1	-	0.882	258.89	-16.29	
> 1	-	0.895	296.19	-14.58	
> 1	-	1.198	320.46	-3.11	
> 1	2	1.266	62.55	-31.87
> flags not surface
> 
>>    BU1    BU2    11.6      53.1   84.4    228.4   -84.9   16.8 2.2 1.1 11.6
> 
> Oh, this is interesting. I was wondering if you had used "UP" to get
> from BU1 to BU2. In such a case, LRUDs are quite useless, as a leg has
> no direction, so Therion has no information about which direction is
> "left" or "right", so it might have had to make things up and got it
> wrong. (Splays are vastly superior to LRUDs, because each one has an
> explicit direction, so you can point to all walls of a pitch, not just 2
> of them.)
> 
> I had expected maybe it was generating walls without any idea where to
> point them, but that seems not to be the case, since you have an actual
> direction for that leg.
> 
> So it seems the most likely cause for your problem is indeed the "flags
> splay" where you meant "flags duplicate".
> 
> However, I would need more of the data forming that part of the cave to
> be sure, since when I tried to compile your snippet, it did not show the
> problem. Maybe it only shows up when you have both the old and new data.
> 
> I also have never seen data in this format, and am curious as to what
> your "data" command looks like. It looks like it should be this:
> 
> data normal from to tape compass clino backcompass backclino left right
> up down
> But Therion won't let me use that, because the compass and backcompass
> are not perfectly opposite. I am wondering if it gets confused when you
> ask it to use all the data columns (I told it to ignore the backcompass).
> 
> Equally curious about what your measurement device is, since you are
> getting only 10 cm accuracy (suggesting a measuring tape) and yet .1
> degree precision compass and clino, which seems incredibly precise for
> manual tools, especially considering the forward and backward devices
> seem to disagree with each other by as much as 2.1 degrees (compass) and
> 1.1 degrees (clino). There is a huge 4.7 degree disagreement on the
> BU1-BU2 leg. (Because the devices are being used nearly vertically, so
> compass error is increased. Plumbed legs would normally be recommended
> when using manual devices.) Whenever I have seen high quality data from
> manual devices, the compass and clino have been given to 0.5 degree
> precision, because that is the reading limit of the devices.
> _______________________________________________
> Therion mailing list
> Therion at speleo.sk
> https://mailman.speleo.sk/listinfo/therion


More information about the Therion mailing list