DistoX2, SAP5 and BRIC4

[email protected]
Draft - March 29, 2021

Cave surveying workflow

The cave surveying process has several stages
1) data acquisition (measuring devices)
2) preliminary data organization and sketching (cave notebo ok)
3) final data processing,
4) map drafting,
5) cave data storage and management.

The first two steps are carried out in the cave, the last three at home.
Data acquisition is done with Suuntos, DistoX, SAP, BRIC4,
CaveSni per, or other devices. The second step is done with paper
and pencil notebo ok, or "paperless", with a program such as Auriga,
PocketTopo, TopoDroid, etc. For the other steps cavers use cave PC
programs (Compass, VisualTopo, Survex, Therion, etc.) as well as
programs not specifically written for cave surveying (Illustrator
Inkscape, AutoCAD and similar, QGIS, ...) .
The measuring device records the measurements: distance, azimuth,
inclination. It would be go od if it also sets the stations, but this
operation is still not automatic underground. It would require to know
the position from which the reading is taken and that of the point that
is sho oted to. In princi ple this could be done with a "perfect"
odometry. In practice the station assignment must be done by the user
(SAP, BRIC, CaveSni per) or is ignored and left to the second-stage
(DistoX, Suuntos) .

Readings, notes and sketches used to be recorded in the cave

notebo ok. Nowadays the use of programs for handheld devices is
becoming common. Auriga, an electronic notebo ok, has been around
for almost 20 years, and has evolved from organizing the survey
data, with the addition of sketching functions and the integration with
the electronic instruments.
The DistoX has been released with PocketTopo, a windows program
that manages the data, and allows to draw sketches. Several Android
programs are now available to work with the DistoX.
CaveSni per comes with its own PC program.
SAP has its program, PonyTrainer, which can download the surveys
from the device. The data are in the format of Survex.
BRIC saves the data on the internal SD card, in CSV text files. These
can be transferred to a PC attaching the BRIC as an external memory
device, like a USB key.

For the third stage dozens of PC programs have been created. The
development of most of them stopped after a short time. Some turned
out to be quite successful and are currently used and maintained:
Compass, Survex, VisualTopo, CaveRender, ... Other programs used
in the U.S. are Walls, and, to lesser extent, WinKarst and OnStation.
In Europe TopoRobot is still very popular. The original program does
not seem maintained, but it has influenced several other programs.

Therion is the worldwide reference for drawing cave maps. However

most people find Illustrator/Inkscape more intuitive. Others use
AutoCAD, usually bacause of the personal technical background.
There are other map drafting programs (cSurvey, Tunnel, TopoCalc'r,
GHDraw, ..., CaveRender) . Either the programs of the second stage
exports directly for these, or specific "converter" programs or plugin
have been created.

For the last stage, people used Cave Kadasters. Nowadays these are
being replaced by web repository, online Karst Information Database,
etc. These are becoming more and more integrated with geographical
information system servers.
DistoX
The DistoX2 has been
around for over ten years
and it has become the
reference standard for an
electronic integrated cave
survey instrument. It is well
known, it has been
described by B. Heeb
[refs] , and there exists a
fair number of papers that
describe its features and test the accuracy (of which only a few are
listed below) .

The DistoX (based on Leica Disto A3) weight about 150 g, and has
size 135x45x31 mm. The DistoX2 (based on Leica Disto X310) weights
about 150 g. Its size is 55x31x122 mm.

The DistoX interfaces via blueto oth with an application in order to

download the survey data. The applications that supports DistoX are
PocketTopo, TopoDroid, Auriga, CaveSurvey, and SexyTopo.

The DistoX calibration requires to take 14 groups of four shots each,

distributed in all directions. The calibration data are downloaded and
the calibration coefficients are computed and uploaded to the device.
The application that supports the calibration of the DistoX are
PocketTopo, TopoDroid, Auriga, and SexyTopo.

From version 2, the DistoX firmware can be upgraded via blueto oth.
The applications that support firmware upgrade are PocketTopo and
TopoDroid.
Website: https://paperless.bheeb.ch/

[1] B. Heeb, "Paperless caving - An electronic cave surveying system", Proc.

Verco ors 2008 - Spelunca Memories 33 (2008) 130-133
[2] B. Heeb, "An all-in-one electronic cave surveying device", CREG Journal 72
(2009) 8-10
[3] B. Heeb, "The next generation of DistoX cave surveying instruments", CREG
Journal 88 (2014) ...
[4] A.A. Ortiz, J.G. Sanchez, "Realizacion de una poligonal cerrada en el Sistema
Cueva del Agua. Karst en Yeso de Sorbas. Almeria. Precision del DistoX", Comision
de Topografia Espeleologica de la F.A.E. 16 de octubre de 2010
[5] D. Ballesteros et al., "Tape-compass-clinometer, DistoX or total station, what is
the best method to elaborate a cave survey? A case study in El Pindal Cave, Spain",
Int Congr. Geomorphology, Paris 2013
[6] L. Redovnikovic, M. Ivkovic, V. Cetl, I. Sambunjak, "Testing DistoX device for
measuring in the unfavourable conditions", 6th Int. Conf. Engineering Surveying,
Prague 2014, p. 269-274
[7] K. Almeida-Warren, D.R. Braun, S. Carvalho, "The DistoX2: a methodological
solution to archeological mapping in po orely accessible environments", J.
Archeological Science, Report 35 (2021) 13 p.
[8] U. Del Vecchio, T. Santagata, N. Damiano, "Rilievo 3D in grotta, comparazione
di diverse metodologie", III Convegno Campania Speleologica, Napoli 2017 67-76
[9] M. Perme, "Kako in zakaj umerjati DistoX", Glas Podzemlja 2020, 61-65
Shetland Attack Pony
The Shetland Attack Pony (SAP) is an early
project for an integrated electronic device
for cave surveying, like the DistoX and
CaveSni per (and the no-longer developed
DUSI) . Initially (2007) it measured only
azimuth and inclination, however the latest
version, SAP5, appeared in 2020,
measures also the distance and there is a
model that has blueto oth and can
immediately transfer the data to a connected handheld device (usually
Android) .

The case seems a bit flimsy compared to that of the DistoX. It is

sealed with silicon, so it is not clear how much it can resists water and
dust (lime) .
It weights less than 100 g, and has size 90x65x27 mm (not
considering the button which is 10 mm high) . On the rear there is a
holder for a wrist strap. Under it is the micro-B USB socket. The
socket is not protected against water and dust, and this can be a
problem in the cave.
The SAP is recharged through the USB connection. If it is connected
when it is off, it will start recharging. If it is connected when it is on it
does not recharge, however if it is
turned on while recharging it will
continue to recharge.

The SAP has a single-button user

interface and a small display. A double
press of the button turns the device on or off. It has a timeout to turns
off by itself if not used for a while. When the SAP is turned on the
laser goes on. The display shows the battery charge, the time and the
units.
A long press of the button, until the displays blibs, starts a reading.
One can hold the button down until the reading is complete. If there a
device connected via blueto oth, the shot is tranferred to that.
Readings that are not immediately transferred, will not be transferred
later.
The user can cho ose to store or discard the reading on the SAP itself.
By default a reading is not stored. Therefore if the user takes
immediately a second reading, the previous one gets lost (unless it has
been transmitted to a connected device) . The shots stored in the
memory can be inspected with the PC program PonyTrainer, which
connects to the SAP over USB. PonyTrainer is available for Windows
and for Linux. It accesses the stored data and convert them into
Survex format.

The display shows azimuth, inclination, distance and extension, either

one at a time (large setting) or all together (compact setting) . Tilting
the device it cycles thru the readings and three options:
- Store
- Discard
- Main menu
Pressing the button while a reading value is displayed, starts a new
measure. The other three cases are for actions or the menus.

The Store submenus lets you set the stations and save the reading in
the SAP memory
- A -> B, eg, 3->4
- A ->
- B -> A
- B ->
- CUSTOM lets cho ose stations AA->BB
- Back (goes back to Store)

Discard discards the reading and goes into "measure" mode. The Main
Menu has several submenus
 - Settings
- Measure, which goes back to measuring
- Calibrate
- Visualize
- Info
- Off, which turns off the device

With "Settings" the user can set units (Metric, Imperial) , style
(Cartesian, Polar, Grad) . display mode (Compact, Large) . timeout (30
s, 60 s, 2 min, 5 min, 10 min) , and date and time.

Calibrate is used to for sensors calibration, as well to set the distance

zero.
Visualize displays a plot of the midline, and
it is still under development.
Info shows lots of data and infos:
- Raw M and G: XYZ
- M and G
- Name compass clino
- Field and di p (at this point the laser turns
on)
- Stddev of magnetic and gravity values:
XYZ
- time, date, battery, temperature
- hw fw data-format

The axes of the SAP are as in the picture. Y aligned to the laser, and
Z upward.

Calibration
The old model of SAP required a calibration course consisting of 12 or
more stations around a central master station, and the truth values of
the directions of the calibration stations with respect to the master
station. Therefore, the SAP precision was limited by the accuracy of
these directions. Four readings were necessary for each direction at
four orientations of the SAP display: up, right, down, and left. After
having taken the readings of the calibration shots, these were
downloaded to a PC program, supplied also with the true values. The
program computed the calibration and uploaded it to the SAP.

The 2020 models do not need a calibration course, and the

calibration is computed directly by the SAP. However the process is
somewhat awkward. To start the calibration the user goes to "Main
menu" on the device, then "Calibration", "Sensors", and confirms. At
this point the SAP display says to put the device on an inclined surface
and rotate 90 degrees at each beep. Then the display goes black
while the user places the SAP on the surface are rotates it at each
beep. Then the SAP display says to place the device on a horizontal
surface face up and turn it upward, face down, downward at each
beep. Again it goes black while the user follows its instructions.
Lastly the SAT says to take 8 readings at a distance point, rotating
about 45 degrees each time. As usual it beeps after each reading. It
can do this for more than one point. Finally it says "processing" and
spits out the result: a percentage, whether or not the calibration is
go od, and whether it has been stored.
The user has to be alert to follow the SAP instruction timely. He has
better have spotted the two surfaces and the two target points in
advance, because the SAP does not give much time to cho ose them.

In conclusion it seems that overall the SAP suffers from a device-

centric design (similar to the CaveSni per) : a shot consists of readings
and stations. Taking readings is as simple as pressing a button, but the
station assignment requires going thru menus and making a choice.
The DistoX avoids this by leaving it to the program in the second
stage.
The data transfer protocol is rather simplistic. The fact that the
readings are lost when the SAP is out of reach from the Android is
likely to be a problem in certain surveys.
The single-button interface is limited, and the "tilt" to cycle thru the
options is awkward to say the least: one may accidentally get to the
next option without willing to do that.
The choice of having the calibration procedure driven by the SAP is
very bad. The user must follow the timings given by the device.
Furthermore, the screen does not provide any feedback during most
of the steps. Finally, the SAP does not offer a choice whether to store
a calibration, or discard, or acquire further data.
Hopefully its usability will be improved in the future.

Website: https://www.shetlandattackpony.co.uk/
Github: https://github.com/furbrain/SAP5

[] Ph. Underwo od, Calibrating a combined electronic compass/clinometer, Compass

Point 37, 2007, 5-7
[] Ph. Underwo od, A combined electronic compass/clinometer, CREG Journal 66,
2007, 12-14
BRIC4
BRIC4 (Blueto oth Ruggerized
Integrated Cartographer,
version 4) is enclosed in a
shock- and water-pro of Pelican
case. The BRIC4 weights
about 350 g, and has size
150x88x45 mm, not
considering the case hinge,
lock, and stri p holders, and
the BRIC4 tail.
There is a single button (on the
bottom side) for normal
operations and four internal
buttons to adjust the device settings, and to calibrate it. Inside the
cave there is a mini-B USB socket that is used to recharge the device
battery as well as to connect it to a PC to downlad files. A mini-B USB
connector is not very common nowadays - it is an old connector that
you might find at home if you have an old digital camera. The BRIC4
can be charged from the PC. While charging the main screen shows
the battery charge with a flickering percent symbol.
When connected to the PC, the BRIC4 acts as external SD-card. The
filesystem ro ot folder contains:
- calibration folder
- data, a folder with the survey data in CSV format.
- Sn.txt, a text file containing the BRIC serial number.
- "System Volume Information'
There is one survey file for each survey day. The "calibration" folder
contains the calibration files, CSV, report (text file) , and raw (binary
file) . Maintenance files are uploaded to the ro ot folder.

To turn the BRIC on press the button three times in quick sequence.
To turn it off press and hold the button for 2 seconds. It turns off by
itself after a minute of
inactivity.
The laser is not immediately
switched on when the device
is turned on. To start it press
the button once. To take a
reading: press the button
again (with the laser on) .

The four inner buttons are

the menu (and navigation) .
When the display is in normal mode the buttons are
Menu, Light-Up, Light-Down, Menu
When the display is in menu-mode they are
Enter, Up, Down, Back

Pressing any internal button, when the BRIC is off, turns it on.

Main menu
Options
Distance units m / ft
Temperature units C / F
Shot delay 0s
Charge current 500 mA
Error sensitivity 1 deg
Backlight color white, red, blue, green, purple, cyan
Backlight level 0
Error info
Measurement 48 (index)
Comp2 High 14.6866
Mag delta ax2 51.809 %
Calibration
Display the report of the current calibration.
Lo op test. A lo op closure test carrued out by the device.
 Quick Azimuth calibration
Full Azimuth+Clino calibration
Range finder: distance calibration
Set Clock (format: yyyy mm dd HH MM SS)
Blueto oth info (name, MAC, status, etc.)
Debug menu
Sensor raw data
Backlight manual
Charger info
Reprocess full calibration
Reprocess azimuth calibration
Firmware update

The normal mode displays five rows of data:

index (last four digits) , distance, azimuth, clino
and a bottom row with
time, battery_level, temperature, blueto oth (if connected)

The index starts with an 'E' if the BRIC4 detected and error. The
BRIC$ as redundant sensors both for acceleration and for magnetic
field. It signals an error if there is a discrepancy in the sensor
readings.

The data protocol transfers primary shot data (date, time, distance,
azimuth and inclination) , additional data (index, roll, di p) , and error
data. If the client does not subscribe to additional and error data, only
the primary data are transferred. When the client is not in reach (or
not connected) the data are stored and transferred upon regaining the
connection.
The transferred data are automatically acknowledged by the blueto oth
stack. No acknowledgment is expected from the client. This leaves the
possibility that
a data does not reach the client and is lost.
Calibration
The calibration of the BRIC4 is quite similar to that of the DistoX.
However the computation is done by the device, and there is no need
for an external program. The user must take at least 14 groups of four
shots in directions as for the DistoX. Within each group the device
must be rotated about 90 degrees at every shot. The BRIC4 displays
the number of groups and the number of shots in the last group (with
four circles that gets filled as the four shots are taken) . It is possible to
take more than 14 groups.
After taking the shots the calibration coefficients can be computed and
a report is displayed.

Calibrating the BRIC4 feels very like calibrating the DistoX. I have not
yet tested what happens if a shot is taken badly.

Firmware update
The BRIC4 has two firmwares. The system firmware is updated from a
PC with Windows (and maybe also Linux) . The Blueto oth firmware is
transferred to the BRIC4 sdcard and updated from the menu of the
device.

Website: https://www.caveexploration.org/gear/bric4
Accuracy test
A DistoX2, a SAP5, and a BRIC4, have been calibrated at the same
place and time.

The DistoX has been calibrated with 22 groups of four data each (for
a total of 88 shots) . The calibration has been computed with
TopoDroid, using the "TopoDroid" policy which compares the shots
within each group, for all the groups and not only for the first four.
The data distribution shows a complete (100%) cover of the directions.
The average error is 0.13 degrees and the original delta is 0.356. The
error standard deviation is 0.10 degrees and the maximum error 0.54
degrees.

The SAP5 calibration has been done following the SAP instructions: two
groups of four shots each, on an inclined surface and a horizontal
surface, followed by two groups of eight shots each aiming at two
targets with an inclination of roughly 45 degrees. The calibration file
can be retrieved from the device using the PonyTrainer. The file (in
json format) contains the shots and the calibration coefficients,
"accel": 1.0089497, 0.0124098, 0.0008267,
-0.0136654, 0.9943023, -0.0028016,
0.0007578, 0.0050781, 1.0007210,
-0.0140548, -0.0232518, 0.0323913
"mag": 0.0191062, 0.0000952, 0.0000050,
-0.0003655, 0.0191116, -0.0000852,
0.0000080, -0.0000473, 0.0194873,
-10.7116508, 48.5108490, 30.87334061
"laser_offset": 0.0900000

The BRIC4 has been calibrated with 18 groups of four shots, The
azimuth standard deviation was 0.106 degrees, the inclination standard
deviation 0,079 degrees. The accelerometers delta percent x=0.006,
y=0.007, z=0.007. The magnitude percent error A1=0.013 and
A2=0.015. The magnetometers delta x=0.043, y=0.074, z=0.047,
with magnitude percent errors M1=0.167, and M2=0.162. The
rangefinder has a calibration of 0.18 m.

After calibrating the three devices a simple test has been performed,
measuring the same two legs with the three devices. The legs are
Distance Azimuth Inclination
6.16 m 188.0 2.9
9.07 m 96.6 -40.9
For every device each leg has been measured with four shots with
orientations at 90 degrees. The maximum difference between leg
shots for each device are
device leg-1 leg2
DistoX 0.4 0.3 azimuth
0.2 0.2 inclination
SAP5 0.5 0.5 azimuth
0.2 0.2 inclination
BRIC4 0.2 0.3 azimuth
0.2 0.2 inclination

The following images show the plan view differences among the three
devices. The suffixes are 'd' DistoX2, 's' SAP5, and 'b' BRIC4. The
side of the square is 0.1 m.
The three points for the horizontal leg are within a couple of
centimeters, which is well below the accuracy of setting the
instruments on the station and aiming at the target.
The points of the inclined leg are distant about 4 cm from one
another. However the target was farther away (9 m, instead of 6 m)
and not as clearly marked, as it was the top of a barrel.

A second test was carried out a week later. Although the test time
was after sunset, I have experienced difficulties with the SAP. It often
failed to get the distance with the laser. I do not know whether this
was because of the eccessive daylight or because of the rugosity of
the aimed surface.

Six legs were measured with the DistoX and the BRIC5. Only the first
four could be measured with the SAP5, and in some case not all four
orientations.
Distance Azimuth Inclination
3.62 m 208.9 9.4
6.87 m 301.4 26.1
6.20 m 220.8 10.1
7.51 m 234.5 12.8
10.64 m 84.1 -53.3
7.15 m 23.0 -29.1

The variation among the angles read by the DistoX stayed inside 0.4
degrees for the azimuth and 0.2 degrees for the inclinations. For the
BRIC4 both the azimuth and the inclination variations stayed within 0.2
degrees. The SAP5 inclination variations were within 0.2 degrees, but
the azimuth variations were almost 1 degree.

A graphical comparison shows that the leg endpoints for the DistoX
and the BRIC4 remain within a distance of 2 cm from each other. This
is quite satisfactory considering that the target could be aimed with an
accuracy not better than a couple of centimeters. On the other hand
the SAP5 endpoints were as far as 15 cm from the DistoX/BRIC4
points. This is an relative error of 0.025 on a 6 m shot. The accuracy
of the SAP5 is therefore 1 degree.

Conclusions
There are many factors that must be taken into account for an
electronic integrated device for cave surveying. Cave surveying
presents many different challenges depending on the variety of cave
conditions.

In order to compare the three devices I consider eight points:

1 - handyness
2 - robustness (resistance to shocks and water)
3 - transportability (weight and size)
4 - usability
5 - calibration
6 - accuracy
7 - integration with cave surveying software
8 - cost
For the first point, the DistoX2 is definitely the winner. To this extent
the main advantage of the DistoX2 lies in its case which is produced
by a top quality company. The BRIC4 is enclosed in a Pelican box
which is a bit bulky. The SAP is small and easily handed.
For the second point, the best choice is the BRIC4. The DistoX2 is
water resistant only to splashes. The SAP5 is definitely bad because
the USB socket is exposed on the outside.
As for weight and dimensions DistoX2 and SAP5 are optimal. However
a protection case must be expected in most caves. Considering this
they are not much better than the BRIC4.
Considering usability, the DistoX2 is optimal (again because of its
industrial case) . However the presence of many buttons is confusing at
times, and one may happen to press the wrong button. The SAP5
button is bit clumsy, expecially because one must hold it pressed a
second or two. The BRIC4 button is not as smo oth as those of the
DistoX2, but it offer the advantage of a single button interface. Both
the SAP5 and the BRIC4 can be operated with gloves, contrary to the
DistoX,
The BRIC4 comes with an extension tail with a rubber ti p which is very
handy to precisely place it on the station. Similar extension tails have
been designed for the DistoX2. Both the SAP5 and the BRIC4 do not
have a thread to attach the device on the tri pod.
Fot the calibration the BRIC4 is the best because everything is done on
the device and there is no need of an external software. Also the
SAP5 calibration is done on the device, however the procedure,
driven by the device instead of the user is distressing.
The DistoX has a precision suitable for UIS grade 6 (Ph. Hauselmann,
UIS mapping grades, Int. J. Speleol. ?) , when properly calibrated. The
test shows that both the SAP5 and the BRIC4 are suitable for accurate
cave surveying. However the SAP5 can achieve a UIS grade 5 survey
(1 degree errors) . The BRIC4 attains a survey quality comparable to
that of the DistoX, if not slightly better.
All the three devices are well integrated with cave surveying
applications. The DistoX2 has its own program, PocketTopo for
Windows, and is supported by all Android apps and by Auriga. The
SAP5 is supported by TopoDroid and SexyTopo. The BRIC4 is
supported by TopoDroid and CaveSurvey. Both the SAP5 and the
BRIC4 store the survey data on the device in formats that can be
easily transferred to a PC. For the SAP5 one need its program
PonyTrainer, which retrieves the survey data in Survex format. For the
BRIC4 the device memory is visible as external sdcard, and the
surveys are CSV files.
The SAP5, with a price below 200 euros, is an interesting option for
surveys of small caves, especially for people who survey with the
paper notebo ok. The cost of the DistoX2 varies, as one need to buy
the DistoX board, the non-magnetic battery, and the Leica Disto, and
assemble the device. It can be estinated around 400 euros. Finally the
BRIC4 costs 850 US $.

The chart below summarizes these considerations using a scale 0-5 for
each factors. DistoX: red, SAP5: cyan, BRIC4: blue. It reflects only my
personal opinion, and it might change over time as the devices get
improved.