Remote Location Site Camera Project
The photograph above shows
the inner workings of the RLSC.
This
article is in 8 Parts:
Contents
Part 1 History ‑ Planning Considerations
Part 2 The Design
Part 3 Switch design
Part 4 Construction
details
Part 5 Electrical
Design
Part
7 Some Fun with Film - Recovery
and Results
Return to Welcome Page of
this 40th Anniversary Edition
Part 1 History ‑ Planning Considerations
by Lloyd N.
Robinson
During the
course of the Illawarra Speleological Society's (I.S.S.) 1982 caving
expeditions to the northern areas of the Northern Territory and Western
Australia (I.S.S. Newsletter, vol. 2, no. 4), it was raised by Joe Jennings
that "it would be interesting to witness the flooding of the northern
caves during the tropical wet season".
The above
comment initiated the following ambitious project, which was branded by some as
a useless exercise and praised by others as having value.
To the
eventual project team the task of taking photographs of rising cave flood
waters in the hostile north while possibly relaxing at home, in pleasant surroundings,
in front of the television, three‑and‑half thousand road kilometres
away, was viewed as a challenge. At the very least, the reliability or
otherwise of the equipment used would be ascertained, which could be of benefit
to others contemplating similar projects under trying conditions.
The society
had contemplated running an expedition to coincide with a wet season. However,
after discussions with local residents and the study of local rainfall
registrations it was determined that one could spend four weeks in the wet
season and see no rain at all, let alone a substantial fall. This being
considered it was decided to take the idea no further.
The maximum
height of water flowing into a cave entrance has been monitored over a number
of wet seasons. The variation between the highest and lowest levels recorded
was considerable, with the lowest maximum being 0.3 metres compared to the
highest maximum of 2.7 metres.
It was
decided that rather than subjecting ourselves to a wet season expedition, a
remote location site camera (rlsc) arrangement could be installed in a
suitable location within a dark cave to photograph rising water levels at
predetermined heights.
Health problems resulting from the 1984 expedition brought a halt to our northern adventures for a few years. During this period the project team gave occasional thought as to how the project should be tackled, or whether it should be abandoned.
A rival project, in the form of a caving robot, was also in the pipeline. Its purpose is to cross a sensitive cave floor to investigate an extension and secondly, to proceed beyond a formation obstruction in a small cave. A collection of bits and pieces were gathered over time for both projects ‑ it's amazing what people will part with when it fails to work.
We opted for the rlsc project in spite of its greater possibility of failure. Its rival was held in abeyance as it was considered our construction efforts could be wasted due to alternate routes being found to bypass the obstacles or others could come on the scene and not be deterred by the sensitive formations in pushing onwards.
Progress was hindered when the I.S.S. was placed 'on hold' for a number of years due to liability scares that landed too close for the comfort of club members. During this period the authors of this series carried on in a private capacity.
One drawback of this particular project is the need to run trips over successive years. Past experiences were not encouraging for mobilising expeditions in successive years. Also, it is a long journey to recover a film that may not even be there to collect.
The major problem forecast was the long period of idleness of the rlsc had to endure. Our northern trips are run around the middle of the year at which time the rlsc would be installed and ready to operate. It would need to survive a waiting period of 6 to 8 months before any water flow arrived to set the rlsc into action, the final two months being the harshest during the buildup to the wet season.
If the rlsc functioned as planned and photographs were taken over the ensuing wet season, it would then become redundant, save for protecting the film. The final months before recovery would be far from ideal for exposed, unprocessed film, well beyond film manufacturers recommendations.
Due to the possibility of a super flood 'once in 1 000, 100, 10, or whatever years,' which would wipe out the project, plus the greater possibility of vandalism, a project 'done on the cheap' was favoured. Therefore, names such as Nikon, Pentax, Cannon, etc. did not reach the planning stage.
The high failure rate of camera and video equipment in northern areas steered us away from the automatic 'point and shoot' cameras, a decision supported by camera repair technicians.
For our needs the lighting medium had to be positioned some distance away from the camera to avoid the problems created by moisture laden air and/or water spray. Cameras using their built‑in electronic flash would not be suitable under such conditions.
From the onset we planned a once only attempt, trusting the wet season involved would be normal (ie. not too dry or too wet). As it was expected that the rlsc would be in no condition for further use after one year in the hostile environment considerable thought was directed towards its simplification.
Planning proceeded at a snails pace over a number of years with many schemes and ideas coming forward. The rlsc installation would have four or five main components in separate locations with connecting electrical wiring as:‑
(a) power supply (expendable or rechargeable batteries in an enclosure);
(b) solar panel (if rechargeable batteries used);
(c) lighting in the form of expendable flash bulbs, electronic flash or quartz floodlight(s);
(d) camera and control equipment in an enclosure;
(e) the triggering devices.
The final item demanded the most attention as it had to reliably set the system into action at different water levels, and from the time of installation it would be electrically alive. Of all the main components this one had to be installed within easy reach of anyone entering the cave. One could imagine the words of an inquisitive visitor sighting this item: "what on earth is this for?"
Did I hear someone say "we will use high voltage for this part of the circuit."
Part 2 – The Design
by Lloyd N. Robinson
The long period of idleness and intended location had an influence on the design of the RLSC. A number of components (ie. camera shutters and electronic flash units) have a habit of becoming inoperative if not used regularly, or exposed to trying conditions.
We did not want to be faced with complicated fault location or repairs on site; there would be enough problems with installation. Our thoughts were centred more towards the approach of a blacksmith rather than that of a watchmaker.
The favoured camera was one from the 1950/60's era. These cameras of metal construction are easier to modify than the modern plastic cameras.
A Tamron-auto SE 35 mm rangefinder camera with a fixed citizen-PE f2.8, 35 mm lens was chosen. Its between lens leaf shutter was not working on the high speeds, the reason for it being discarded by its previous owner. This fault was of no concern to us as we intended to jam the shutter open.
Before the camera was modified, it was loaded with 1000 ASA colour negative film and taken to Blackheath. On a dark damp night it was set up to photograph a scene a little upstream of the Horse Falls that roughly resembled the proposed subject. A number of photos were taken using PF1 flash bulbs and various electronic flash units, one flash for each photo. Also a few photos were taken using battery powered floodlights. All results were encouraging.
As a first construction move the camera was modified so that a gutsy motor drive could be fitted to overcome any problems with cranky film cassettes. The stroke lever film advance was discarded as being over-complicated to operate by means of a motor drive. The clutch and drive sprocket were freed from the auto chain and left to freewheel with the stroke lever being replaced with a small wheel and knob to act as a hand winder. The film transport would now work in reverse similar to some modern automatic cameras. On loading the film is hand cranked onto the take-up spool.
The motor drive is fitted to the camera's original rewind mechanism to wind the film into the cassette, one turn per photo. With this arrangement the last photo on the film becomes the first; the first three photos slightly overlap each other. As the film builds up on the cassette spool the gap between each photo increases thus reducing the number of photos per film; not a problem for this project. A sound reason for this method is that any photos taken end up within the cassette.
At least three test photos would be taken on-site to check the operation of the RLSC after installation. If thought necessary these test photos could be cut free and processed first so that processing times could be altered for the photos that mattered.
To house the camera a robust enclosure was considered the best option complete with a light tight sliding door driven by a second gutsy geared motor to briefly expose the camera lens to the cave atmosphere for each photo. We had in mind the possibility of an insect building a mud nest against the sliding door, hence the need for a strong drive to free the door. A glass shield in front of the door was considered but not employed as it was felt it could become dirty.
For lighting we opted for electronic flash. The problems involved in rotating expendable flash bulbs posed too many complications. Likewise, floodlights were by-passed due to the considered risk to the solar panel from vandalism and cyclonic weather
The means of triggering the RLSC gave rise to a considerable amount of brainstorming. Only one photo was required from each selected water level. This meant that as each trigger did it's job it had to become neutralised, as we intended to run a two wire system to which all triggers would be connected. To achieve this we opted for sacrificing each wooden float by allowing it to float downstream
Requests were made to include a time and date record on each photo. Some research was carried out on this matter, but we decided to leave same for another occasion.
While the RLSC was under construction an unscheduled northern trip eventuated which allowed time for some caving for the small party involved. The opportunity was taken to carry out a test run using the single shot sheet film camera, as used in the Gunbarrel Aven of Wyanbene cave (I.S.S. Newsletter vol. 1 no. 12), in a small cave that had a wet season water rise of approximately 0.5 metres.
This camera has a solenoid operated shutter. The camera was placed in a plastic two litre ice-cream container with a small window; it was positioned in the container by using small stones and a silica gel bag.
The flash was a PF100 flash bulb in the base of a two litre ice-cream container with a backing of foil as a reflector; the lot being housed in an oven bag. A 6 volt lantern battery also found it's way into yet another ice-cream container. A small stake, stuck in the watercourse, supported a tilt type float switch that would activate the shutter solenoid. The shutter was set on one second and once activated also fired the flash and disconnected the battery to avoid solenoid burnout.
PART 3 – SWITCH DESIGN
by Dave Dicker
This article follows on from Part II of the project, and outlines the reasoning behind the final design of the triggering switches
After internal modifications were carried out to the camera, consideration was given to the mechanical operation of the equipment. A number of criteria were arrived at:
a) The film wind on was to be electrically driven through a gearbox.
b) The "shutter" was to be an electrically opened door in the camera container.
c) Four or five triggering switches were to be used. The switches were to operate once, then be disabled.
d) An electronic flash was to be used, the flash being housed in a waterproof enclosure, and positioned a short distance from the camera.
As mentioned in Part II of this series, we now had a working camera prepared for the project. The next, and most critical job, was to design the triggering switches.
There were a number of potential problems which were unique to the switches. They were the only part of the project which would be totally exposed to the elements, thus they would be the most likely part to fail due to the long waiting period. By the nature of the project, the switches had to be installed in the main stream passage, thus increasing the risk of vandalism by casual visitors. As mentioned in Part I of this article, the switches would be electrically live from the date of installation, an open circuit failure would render the project inoperative, and a short circuit failure would keep the equipment running until the batteries ran flat.
After much discussion, some design criteria were arrived at for the switches:
a) They had to be as small and inconspicuous as possible
b) They had to be simple and reliable
c) They had to trigger once, then disarm (To positively disarm the switch, the triggering element had to be totally removed from the switch)
d) They had to be unaffected by dust and water
e) They had to be installed in an unobvious location, yet had to be exposed to the main flow during the wet season.
At this stage, the serious brainstorming got under way. Our first idea was a tube pivoted in the middle, with a float mounted at one end. As the flood rose, the tube would come just over the horizontal, and send a ball bearing rolling down the tube to trigger a micro switch. This idea was thought to have too many moving parts, and thus, too high a risk of malfunction. Another idea was to mount a sensitive pressure switch to the top of an inverted blind tube. This would eliminate all moving parts, but the problem of disarming the switch would have been difficult to solve The idea of an electric detonator was suggested - admittedly it was around the campfire at Wyanbene, so one doubts the seriousness of the idea!
After considerable thought, we decided that a reed switch would be one of the simplest and most reliable devices available. A reed switch is operated by passing a magnet across the front of the switch. The magnet has a range of approximately 12mm, so distance from the
switch is not critical, and the unit does not have to be ultra precise A cheap commercially available switch was found at Tandy's. This particular model was designed as an alarm switch in a home security system, mounted in a door or window frame. They measure 6mm diameter by 35mm long and came complete with magnets. The final design incorporated the magnet mounted in a hollow wooden float which ran in a copper tube, the float having an "umbrella " head to prevent any dirt entering the tube and jamming the float The reed switch was silasticed into a second tube as shown in Fig. 1. As the flood water rises, the float rises in the tube, triggers the switch, then floats away, thus disarming that particular switch,
Bench testing showed that the unit was 100% reliable, regardless of how quickly the float was withdrawn from the tube. A potential problem would occur if the floodwaters rose slowly - the float could be within range of the switch for some time, thus taking more than one photo. We decided that this was an acceptable risk. Thus all design parameters were met, and another element of the project completed. Five of these switches were made. They were to be set at different levels in the cave.
This left four elements of the mechanical design to be sorted out, film transport, the "shutter", equipment enclosure and flash arrangement.
PART 4 – CONSTRUCTION DETAILS
by Dave Dicker
The project was now gaining some momentum. We were some 12 months away from running a trip to the Kimberley, and there was still a considerable amount of work to do. We had gathered together a few elements of the project which worked well under controlled conditions. The next stage was to give some thought to the film re-wind, and "shutter" opening mechanism. These had to be completed and mounted before the enclosure dimensions could be determined. As stated in part III, these functions were to be carried out by means of electric motors driving through reduction gearboxes.
Commercially available small motors run at high speed (8000rpm +) and are fairly expensive. Eventually, a dead cassette player became available. It was quickly dissected and the motor removed before the owner changed his mind. The motor was very quiet and smooth-running and ran at 2000rpm. Eventually, another dead cassette player was located and its motor appropriated.
The next challenge was to come up with a satisfactory gearbox. Commercially available, hobby gearboxes were looked at, but had some disadvantages: they were very expensive, they were bulky and didn't achieve a low enough ratio - especially for the film transport. Industrial quality small gears were also looked at, but again, they were expensive. The next step was to visit the local hobby shop, and there we found what we wanted. They stocked a large range of "Meccano" products, including gears. To keep the size to a minimum, a worm / compound reduction was chosen, consisting of a single start worm driving a 60 tooth gear. A 19-tooth gear was soldered to the 60 tooth gear, and this drove another 60 tooth gear. This gave an overall reduction ratio of 189: 1 in a final package of 20mm x 50mm x 84mm. One gearbox was built to this design, the motor attached and the assembly mounted to the camera. A film was loaded, and the system hooked up to a 6V battery. The motor was struggling a bit towards the end of the film. This could be disasterous if the battery charge was getting low near the end of the exercise, so the original gearbox was relegated to the "shutter" function, and a new gearbox built. The new design had a worm driving a 76-tooth gear with a 19-tooth gear mounted to the same shaft which drove a 96 tooth gear. This gave a final ratio of 384: 1. The new gearbox was mounted on the camera and tested as previously, with satisfactory results. Various micro switch brackets and actuators were attached to the unit, and from a mechanical point of view, the camera was ready for action.
A start could now be made on building the camera enclosure. Due to possible problems with sticking shutters, the camera shutter was locked in the open position. This meant that a lightproof box was required with a sliding door in front of the camera. The box was made from wooden panels held together with aluminium angles at the edges. This made it easy to remove individual panels to gain access to the machinery. Wood was chosen because condensation tends not to form on it, and it is easy to screw any extra bits and pieces to it. The sliding door ran in an aluminium channel and was protected from drips by a visor on the outside of the box.
The door was opened by means of a 19 tooth sprocket mounted on the output shaft of the
189: 1 gearbox, driving a chain type rack mounted on top of the door. In operation, this arrangement was satisfactory, the door opening slowly and smoothly, thus eliminating any over - run. The inside of the enclosure and all components were painted flat black to minimise the risk of reflections, and the outside was painted a mottled grey / brown / black. A further wooden box was made to hold the two 6V lantern batteries, one to power the camera equipment, and the other to power the flash. This box was totally waterproof. The flash was housed in a separate box having a perspex front. This was done so that the flash could be mounted a short distance from the camera, thus giving better depth to the photos. The flash box was also totally waterproof. Where electrical wiring came out of enclosures, the holes were thoroughly silasticed.
This completed the mechanical side of the project. In the meantime the detailed sequence of events of the equipment had been established.
Part 5
Electrical Design
by Lloyd
Robinson
The last thing
we needed with the RLSC project was having to carry more than a
cheap multi-meter for on-site fault location. The circuit was designed with
this in mind. A number of extras could have been easily added. A third timer to
delay the re-arming of the RLSC between photographs was tempting, to counter
the water level remaining stationary, and failing to defeat a float switch,
thus causing repeat photographs at one level. Apart from having something else
to malfunction, we had been advised that flooding of the intended cave site
came and went quickly, more akin to flushing a toilet.
The tropics
are harsh on electrical equipment. We were happy to run the RLSC from six-volt
expendable lantern batteries, a voltage not likely to be worried by wide
ranging weather conditions.
Relays with
five-volt D.C. coils were chosen to overcome voltage drop problems. These
relays still functioned if the voltage fell well below their rated voltage. For
the timers, we could not go past the hand made timers the author developed to
survive the harsh conditions on mobile face machines in underground coalmines.
A relay to
fire the electronic flash unit was mounted with the flash unit in its sealed
enclosure to contain the high voltage of the flash unit within the enclosure.
As depicted in the wiring diagram, separate power supplies are used. It was
felt that the initial switch-on high currents drawn by the flash unit after
each firing, and more so for the first time after an idle period, could
interfere with the control sequence due to voltage drop if the RLSC was
operated from one battery.
Although we
had rough plans as to the mounting positions of the RLSC modules in three
possible locations, it could be a different matter when it came to
installation. The length of twin core cable connecting the triggering switches
to the camera / control enclosure was the concern. Testing determined we could
maintain reliable operation with up to 400 metres of the selected cable.
Surely, we would not be likely to separate these items by anything remotely
approaching this distance.
Three
single pole changeover limit switches were mounted as:
a) one to
changeover as the sliding door fully opened. (DOL)
b) a second
to changeover as the sliding door fully closed. (DCL)
c) a third
to changeover when in contact with a short cam fixed to the film wind last
stage gear wheel.
A red light
emitting diode (LED) was mounted on the front of the camera / control enclosure
and connected across TSR2 relay coil. It is necessary for timer 2 to complete
its timing cycle during the film winding operation. The LED’s purpose is to
assist with last minute site adjustments to the timing cycle of timer 2, when
test photographs are taken on the film that remains behind to record the
flooding. (Tropical conditions, and variable 35mm film cassettes are causes for
adjustment.) Three different coloured banana plugs were fitted for easy removal
of the camera and winder; a luxury item.
Apart from
the motors and timers, the electrical components used are the cheap variety as
found in supermarket type electrical / electronic stores. The assembled RLSC
had plenty of testing as it attracted a lot of interest from engineering types,
pro photographers and camera clubs, all wanting to witness the contraption go
through its paces.
The wiring
diagram may not be easy for all to follow. To assist, a step-by-step procedure
is detailed to take one through the order of events when the first flush of
water activates the lowest float switch. We must assume that the RLSC has
survived the long wait in stand-by mode and avoided vandals.
1) The RLSC is in stand-by mode with its
circuit ‘live’, but no battery drain, except for insignificant leakage on humid
days. The sliding door is closed, and the cam on the FWM gear wheel is in
contact with the WML limit switch. (The wiring diagram is depicted in this
state.)
2) The first flush of water activates the
lowest float switch.
3) The FPR relay coil is energised, the
relay retains itself and powers the timer via FPR 1, and also powers the flash
unit via contact FPR2.
4) After a delay of 70 seconds, TSR 1
relay coil is energised and starts the door motor via contact TSR 1. As the
door starts to open, the DCL micro switch changes over.
5) When the door fully opens, the DOL
micro switch changes over resulting in:
a) The FFR
relay is energised, firing the flash via contact FFR.
b) Timer 2
is energised.
c) The FWM
motor is briefly powered, causing the WML limit switch to change over.
6) Timer 2 immediately energises TSR 2
relay which:
a) Removes
power from relay FPR and timer 1
b)
Energises relay DMR
7) The DMR relay reverses the power to the
door motor. As the door starts to close, the DOL limit switch changes over.
8) As the door closes, the DCL limit
switch changes over, de-energising relay DMR, stopping the door motor in the
process.
9) Power is now supplied to the film wind
motor via limit switches DCL and WML. It proceeds to wind the film cassette one
turn.
10) During step 9), timer 2 times itself out
and rearms the RLSC.
11) The film winder stops when the cam
strikes the WML limit switch.
12)
Provided
the first float switch has been neutralised, the RLSC stands by until rising
waters activate the next float switch to repeat the process all over again.
Part 6 The Installation
BY Dave Dicker
As stated
in Part V of this article, we now had an operational and fully tested unit. On
a previous visit to the area, a number of possible installation sites were
noted. These had to satisfy a number of criteria:
1. The
site had to be reasonably easy to access for installation purposes.
2. It
had to be inconspicuous from the main stream passage to minimise the risk of vandalism.
3. It
had to be close to the main stream passage to facilitate the connection of the
float switches.
4. It
had to provide a good view of the stream passage, preferably with some feature
in view which could be used to gauge the depth of the water.
One of the
main difficulties in locating and setting up the camera and switches was to
find a site which suited the different maximum levels to which the water may
rise over different wet seasons. On
previous visits, we had monitored the maximum height of the previous year’s
flood near the entrance of the cave. This was achieved by measuring the height
of flood debris on a tree in the main Stream passage. Before leaving the area,
all the old flood debris was removed. We found that the maximum water level varied
from 0.3 metres to 3.0 metres. Judging by higher debris in the cave, there were
times when the level exceeded 3.0 metres. It must also be borne in mind that we
were mounting the equipment in a much narrower part of the cave, where the
floodwaters rise much higher than where we were measuring at the entrance. In
view of all the time and effort spent on the project to date, we wanted to get
some results, regardless of whether the wet season presented us with a low
maximum level or an extra high maximum level. The problem was discussed at
length, and the decision was made to mount the lower switch within 0.3 metres
of the lower part of the floor, and the highest 3 metres above the floor, with
the other three switches equally spaced between.
A
Relevant Side-Track
As
mentioned in Part II of these articles, on the 1992 Kimberley trip, a “One
Shot” RLSC camera was installed in another cave in the area. This was done
because the main camera wasn’t quite ready for the 1992 trip, and we may have
learnt something useful from the results. The camera itself was the Mk2 unit
used in the Gunbarrel Aven in Wyanbene. The equipment was mounted on a gravel
bank and was triggered by a lever type float switch. When recovering the
equipment in 1993, the site was approached with caution, using a single LED for
lighting. (incase the shutter was still open). The equipment had evidently
operated, the shutter was closed, and the flash had fired. The sheet of film
was removed, and the camera removed from the cave. We were encouraged, as the
equipment had operated. It must be noted that the decision had been made at an
early stage not to develop film in the field as it was too messy and risky to
the film, and the risk of pollution with the photographic chemicals was great.
Back
to the main plot
A number of
possible locations had been looked at during the 1992 trip, and bearing in mind
all the parameters, the final choice was in a dark part of the cave, on the top
of a large pile of collapsed boulders which divided the main stream passage.
The mounting position was some 7 metres above the lowest part of the floor. The
occasional super flood would inundate the camera - this was a risk we had to
take. There was a narrow fissure between two boulders just below the mounting
site which was ideal for mounting the switches. There was also another large
boulder in view which had a distinctive white mark on it - possibly the regular
flood level. This was useful as a marker to gauge the depth of water.
All open
wiring was carefully concealed and anchored to prevent accidental dislodgement
by floodwater, or intentional dislodgement by vandals. The switches were
mounted in the fissure by drilling small holes in the boulder and mounting the
switches using plastic plugs and self-tapping screws. As a final thought, the
lever switch from the “one shot” camera was mounted above the 3 metre float
switch and was connected in such a manner that if it was triggered, the camera
would take continuous photos until the film ran out or the batteries flattened.
The camera
enclosure was mounted near the edge of the boulder, facing upstream, using
various pieces of aluminium strips to anchor it. These were screwed to the
camera enclosure and mounted to the boulder in the same manner as the switches
were mounted. The final aligning of the camera was achieved by removing the
rear enclosure panel, opening the sliding door and viewing through the camera
itself, using a piece of greaseproof lunch paper in a similar way to which a
ground glass screen is used on a plate camera. Final adjustment was carried out
by bending the aluminium mounting frame. The flash enclosure was mounted about
two metres from the camera. This was mounted in the same way as the camera and
aligned by eye. The waterproof battery box was wedged under a nearby fissure
between the boulders.
When
everything was assembled and connected, the camera was loaded with an unexposed
roll of Kodak Ektar 1000 ASA film (24 exp). The equipment was now loaded and
alive. Some members of the group were placed in the scene, and one of the
switches was activated. The camera was monitored closely - the equipment worked
perfectly, and at that stage, two datum photos were taken. The site was
camouflaged as best as possible. A large log was jammed in the opening of the
switch mounting fissure, and various flood debris used to disguise the camera
and flash. Some days later, before we left the area, another datum photo was
taken. The equipment was then at the mercy of the next wet season, and
hopefully, the remaining shots would be taken under wetter conditions. We left
the area full of hope, but not knowing whether the test shots were successful.
However, if once installed and left to do its job and someone comes along and disturbs it, well that's a different matter.
PART 7 Some
Fun with Film - Recovery and Results
by Lloyd Robinson
From our 1993 northern trip we possessed the recovered single sheet of
black and white cut sheet film. As was frequently done with the Wyanbene
Gunbarrel Aven photographs (I.S.S. Newsletter Vol. 1 No 12), a small corner
section of the film was cut off from its parent and processed first as per the
film’s processing instructions. The result was good, there had been exposure
and so there should have been with a PF 60 expendable flash bulb as lighting.
The parent film was processed; the author caught a brief glimpse of the cave
scene in the wash water before most of the emulsion parted from the film base
and quickly broke up in the wash water. A second sheet of cut film from the
same box was exposed and processed with no problems. This put a damper on the
project; we could hear the words of photographic technical folk “I told you
so.”
In spite of the setback, project members took keen interest in weather
reports during the 1993/94 tropical wet season.
A small, two-vehicle trip was planned to recover the RLSC in mid 1994. Near departure time, a trip member fell to a serious medical problem, which left him in no condition for any trips. In spite of careful containment, we were not keen to leave the project’s batteries beyond two years.
In April 1995, the Robinsons linked up with Miles and Rhonwen Pierce of V.S.A. at Ceduna on a caving / sight seeing trip with no particular goal in mind. Six weeks later we arrived in Broome. We decided to continue onwards and recover the RLSC. As the trailer still carried a spare float switch and floats from the aborted trip, with some difficulty we purchased a set of RLSC batteries in case the unit still functioned.
At the cave entrance, we noted the gauge indicated a maximum water level of three metres had occurred since our last visit two years ago. On reaching the RLSC unit we noted it’s door was closed, all the switch floats were missing and the lever switch was jammed in the run position by debris. The wiring connecting the float switches to the camera/control box had been mangled where they ran under a slab of rock, upon which the box was mounted. In this part of the cave a stream rise of over five meters was needed to do that.
The batteries had not leaked, but some terminals had corroded and come adrift. With new batteries the system operated with ease, indicating the film was completely within the cassette. We had a unit that still functioned! On opening the camera box, we found the interior had survived in good condition, except for a few spots of fungus on the rear of the camera. We decided to reload the RLSC and to this end we removed the camera and silica gel bag to dry them out at camp.
One float switch was damaged by debris. With higher flooding in mind, we repositioned the lower float switches higher up and the lever switch at a much higher level. The wiring was repaired, and the camera loaded with a 36-exposure roll of Ektapress 1600 film. When all was ready three test photos were taken before leaving the RLSC to the elements.
One problem faced by continuing this project is the fact the RLSC is rearmed without knowing the results of the recovered film, which in this case was posted home to be placed under refrigeration.
Much discussion ensured as to the processing of the film. We decided not to take it to the one-hour photo labs. David Brooks L.A.P.S., S.S.A.P.S., a long time member of the local camera club, who processes his own colour films, volunteered to develop the film. We decided to cut off the test photos and process them first. On the film manufactures advice they were given an extra twenty-five percent development time. The result, three over-processed negatives. David’s film cut had been perfectly placed. The remaining film was processed as normal.
The film was taken to a one-hour photo lab for printing, but the tonal range of the negatives was beyond the range of their machines. Best results were achieved by printing on monochrome paper. The damaged float switch had caused a number of photos to be taken at the one level. We later learnt the film used was being discontinued. With only 20-exposure film available the RLSC had run out of film when the peak action commenced. We were pleased that the system worked to plan and produced photographs!
By examining the photos we can assume all bar the last photo depict the
1993/94 wet season, with the last taken a year later. The film’s final photo
shows water flow with light debris being carried. The earlier photos depict
rising calm waters.
In mid July 1996 the author was again on site, having been able to combine work with C.S.S.’s Northern Territory caving trip. This time the door on the main RLSC unit was found half open, with the box showing signs of having been struck by a solid object (possibly a projecting piece of a floating log). The floats plus one float switch were missing, and the lever switch jammed by debris in the run position. While they had not leaked, the batteries were in a sorry state with all terminals eroded off - too long on the store shelf. Of more concern was the electronic flash unit found under the camera/control box and directed to one side of the intended subject. It could not have reached that position by itself. Obviously an interested party had sighted the RLSC and presumably tried to determine its use, and in the process pulled the flash unit from its mounting. The big question, did this occur before or after the wet season?
The total RSLC equipment was recovered. The missing float switch was found two years later.
A week later we arrived in Perth and had the film developed at a custom lab. The only joy on examining the film was the fact the RLSC had worked - the interference had occurred before the wet season. We had three good test photos and the faintest signs of exposure for the rest of the film - no use for its intended purpose. The reader might lament, like we frequently have, "If only the dislodged flash had been directed towards the subject".
One now must ask, "What has been gained from the exercise?" The answer to this question comes in three parts:
Scientific: Virtually nothing of what Joe Jennings had in mind.
Photographic: Modern negative colour films can take a lot of abuse, far more than what the film manufacturer's claim. These films have been designed to stand the abuses handed out by the general shutterbug (ie. left in overheated parked cars, long periods in the camera, etc.).
Engineering: It is possible to develop cheap, fairly complex equipment to stand the rigours of time under harsh conditions with a long period of idleness and still function to produce results.
2002 and the RSLC was placed in a part of the Bullita
Cave system in a newly found drain that removes all the water from a part of
this massive system. It will be interesting to see the results this time!