Umbillical Video Camera

INTRODUCTION

This page is a guide to constructing and using an umbillical underwater video system. It is just one of the entries in Shonas diving pages. Click on one of the items below to access detailed information on that subject. Any errors or omissions should be sent to the author for inclusion.

Assembled unit
Assembled unit with umbillical camera head


Background
Requirements
Technology Available
Construction
Operation
Technical Specs
Technology problems and Modifications
Performance
Summary


Background

I've been doing some stunning dives over the last few years, and some of these dives are almost too spectacular to believe. This led me to think about carrying a video to keep a permanent record of those dives, but with the bulk of a video and associated lights to carry around, I wouldn't be able to do any 'work' on those dives. Some of the dives were very pretty scenic dives, some were diving intact wrecks in extremely good vis. Whatever the case, most of these trully 'stunning' dives were in the most unexpected and inhospitable conditions.

Bike in 20m
Bike in 20m

Basil Bang
Box of fuses on the Basil - littlehampton

Mast
Mast on shipwreck

Diver in front of boilers
Diver in front of boilers

I've dived with buddies with 'traditional' video systems, and the dive is vrey much orientated towards shooting video. Problem is that these cameras only every get carried on the odd occasion when the vis is known to be good and the sea conditions are perfect - i.e. not very often. What I needed was a video which I was able to carry on any dive, but not prevent me from doing any work whilst underwater.

Almost all of the existing video offerings are of a hand-carried housing with associated lights and cables. These are much too fragile to chuck into the bottom of the RIB in a force 4 and drive 12 miles to the dive site. What was needed was something that was bomb-proof, and hands-free. The idea of an umbillical video was born.


Requirements

In order to be able to carry the video on every dive, it had to be robust and virtually maintenance free. I started to draw up a list of requirements:

1) Robust. Must be able to be treated like all other dive kit.
2) Hands-free. I don't want to have to hold it, thus freeing up both hands for working.
3) Reasonable quality footage. I would like to be able to replay a dive and be able to sketch the details of the particular wreck afterwards. I'm not too bothered about 'pretty' dives, so colour not a necessity.
4) Must be capable of videoing the entire bottom section of a dive, say 40 mins max.
5) Must be able to operate in all depths to which I'm likely to dive.

The biggest problem was how to record the dive footage. Memory chips were still not big enough to be able to dump the footage to solid state, and underwater hard-drive recorders were not available at that time. I investigated chopping down a domestic VHS-format VCR, but even with the audio system and transformer removed, it was still too large to house. I looked at the Sony range of video walkmans - both analogue and digital, but they have a large square footprint and are not that easy to house. I finally loooked at the domestic video cameras available in the UK and decided that this was the way forward. If done carefully, I could always use the camera out of the housing for family events etc.


Technology Available

I needed to find a camcorder which could record from the analog-in port. This would allow me to attach a seperate 'lipstick' type camera which could be helmet mounted. However, camcorders manufactured for the EU market, have an odd quirk in that their ability to record from the line-in ports are disabled for tax reasons (something to do with camcorders and VCRs being taxed seperately). I had to either pay a premium for an analog-in enabled machine, buy one from a non-EU source, or buy one in the UK that could be 'nobbled' to re-enable the analog-in.

The Sony range of Digital8 camcorders available at that time were suitable candidates, are were priced quite competitively. These had the analog-in circuitry and could be 'nobbled' to re-enable it quite easily either using a digital 'widgit' or a serial interface fro your PC. The advantage of the 'widgit' was that it also provided a 'start' and 'stop' button for recording.

There was a good selection of 'lipstick' type cameras available, but most of the colour ones were of poor resolution. I ended up opting for a 'drain' camera which worked with infra-red light. It would work in very low light conditions and had its own infra-red light source built-in. This camera came in a 'waterproof' housing with 50m of cable and would run off a 12v power supply with minimal power consumption.


Construction

Having purchased the camcorder, I set about recording it's dimensions to determine the smallest diameter housing that I could get away with. With the Sony TVR120E, I could just get it into a 5.5" diameter cylindrical tube. I would also have to house other circuits within the housing:

a) CamCorder and power supply
b) Power-supply for external camera
c) Switches and fuses
d) Flood alert system

I sourced some 6" OD ali tube with 1/4" wall and went about getting it machined with an internal thread and step at both ends (Thanks Steve). The endcaps were turned out of 1" slices off a 6" ali bar with one blank, and the other drilled and tapped to take a pg11 cable gland. The 'blank' was drilled with two 'key' holes to enable tightening it. The entire 'housing' was then anodised and an 'O'-ring fitted in each end-cap to provide a watertight seal.

Housing
Housing with endcap

Housing
Housing with endcap and spanner

Head unit
Drain camera in modified housing

I wasn't very happy about the 'waterproof' claims of the original housing for the drain camera. I decided to have the back of the housing re-made out of brass and get it chromed (Thanks again Steve). This was tapped for a pg11 cable gland again. The front of the housing has a very thin glass window - which deosn't seem as if it will survive 50m let alone deeper. Testing the camera around the house, I found that the infra-red LEDs situated around the lens provided too much back-scatter is dusty enviroments (like under the sofa). I've disabled these LEDs and rely on ambient light instead.

I set about building a prototype 'slug' in MDF which would hold the camcorder and other internal parts securely in the housing without things rattling around inside. When the dimensions of this were confirmed, I built the current 'slug' using nylon chopping boards, purchased from Woolworths. These are easily machined and should be relatively waterproof. The idea was to build the slug with watertight bulkheads between each section, to minimise damage should a leak occur. It would be a relatively easy task to add 'o'-rings to the bulkheads to achieve this.

Power for the drain camera is in the from of 10 AA rechargeable cells of 1.2v each. A primitve leak detector circuit was built and potted in the lid of the slug. This has a warning LED and a siren which sounds when the leak sensors start to get damp. The siren is audible when underwater and is powered from the 12v supply. All that was needed was a couple of BNC panel-mounted sockets and plugs to allow the camcorder to be easily connected to the slug. All of these parts came from Maplin at minimal cost.

The 'widget' is velcro'd to the top of the slug and plugs into a socket there. The lid of the slug also holds an on/off switch for the 12v power supply, a BNC video out socket (unused) and a panel-mounted fuse for the 12v supply.

Slug
Slug for main housing

Slug
Camcorder in slug

Slug
Sliding slug into housing

Slug
Slug fully inserted with control panel visible

The bottom of the slug contains the BNC socket for video input and the 12v power supply socket for the external camera. A compartment above this bulkhead contains the sensor array for the leak detection system. This is simply strips of veri-board aranged around the edge of the bulkhead, covering all 360 degrees.

Housing Parts:

6" OD ali tube
2 x 6" ali end-caps
PG11 cable gland

Slug Parts:

3 x white nylon chopping boards (Woolworths)
6mm Stainless steel threaded rod (B&Q)
6mm ID anodised aluminium tube (B&Q)
6mm nyloc nuts (Screwfix)
4 x Brushed Ali door handles (B&Q)
10 AA cell battery holder and wiring (Maplin)
10 x AA rechargeable cells (NiMH)
Booted dolly switch, fuse holder, wiring, sockets (Maplin)
Leak detector circuit with LED and siren (home-made)

Camera Parts:

PG11 cable gland
4-core umbillical cable (central heating controller cable)
Brass billet to make new housing back
Cable protector

The 'O'-rings on the bulkheads were found to be un-necessary as the slug fits like a glove into the housing, so much so that in hot weather, you have to compress the air in the bottom of the housing when getting the slug in!


Operation

The mode of operation is as follows:

1) Charge camcorder battery before dive and assemble all components in housing.
2) Throw housing in back of boat. Don't worry about people putting cylinders on it.
3) When kitting up, open housing, pull slug partly out and put camera into playback mode (dry hands), before sliding slug home. From the top of the slug, switch on the 12v supply and press the button on the widget to start recording.
4) Screw lid onto housing with key and attach to person somewhere (I usually carry it as I would a stage cylinder)
5) Fit drain camera to housing on helmet and route cable around body.
6) Dive
7) After dive, open housing and press button on widget to stop recording. Switch off 12v supply. This can all be done with wet hands. There's no need to turn the camcorder off as its battery will almost be exhausted.

I've carried the housing in a variety of manners from belly-mounted, side-slung and back mounted. Even when belly mounted, the housing doesn't make its presence felt. One unexpected bonus of a system like this is that I'm able to remove the head unit and poke it into holes in wrecks to examine areas into which I cannot see. Of course, I can't see the results at the time, but it makes interesting viewing afterwards.

Head unit housing
Head unit housing above RHS work light

As regarding lighting, the camera works well when there is only a little ambient light. Shining a dive lamp into the frame will 'burn' out the centre of the picture. Problem is, that I need a lamp for me to be able to see. My current solution is to carry a 'video' lamp, which is a low-powered lantern with a scrumpled-up tin-foil diffuser over the reflector.


Technical Specs

Camcorder: Sony TVR120E Digital8.
Housing: Bespoke 6" OD aluminium 'drum'.
Head Unit: Infra-red drain camera.
Umbillical cable: 4-core with sealed plug from 'Dev Pein'.


Technical Problems and Modifications

The system worked well from the start. It took a bit of time to get the head unit pointing in the right direction, as the helmet sits at different attitudes depending on whether it is in water or in air.

I have replaced the umbillical cable with a quick-release cable which plugs into a pg11 socket on the base of the main housing. This will allow me to swap head units whilst on site from infra-red to colour.

I have suffered from camcorder battery problems which are a trait of the InfoLithium battrey technology used by these units. The only solution is to replace the battery every 24 months or so.

I've got round the limitations of tape duration by rewinding the tape to the start of the ascent part of the first dive before starting the second dive. This way, I can get two good 40 minute bottom times on one tape.

As the picture is stored in digital8 format, I can use lower-quality tapes which are modified to run with digital8. Typically, I can get 8mm or Hi-8 tapes from eBay for a nominal sum. By drilling a hole in the back of the tape, I can fool the camcorder into thinking that it's a digital8 tape.


Performance

In terms of the original requirements, the system is robust enough to keep on the floor of the boat without worrying about people standing on it. It's totally hands-free operation with a fixed focus and no lighting requirements. With a large battery I can get two dives out of it on a 90 minute tape.

The system performs even better than expected. In very low light conditions, the camera switches into a low-res mode which looks a little grainy, but in normal ambient light conditions, the picture is broadcast quality. I've sent footage to the BBC for a documentary which they were making, and they were more than happy to use it.

Diver in 50m at NDC, in Chepstow
Diver in 50m at NDC, in Chepstow

Spidge on Waitara at 60m
Spidge on Waitara at 60m

Ascent
Ascent after Waitara dive

The system is quite compact, and isn't noticeable whilst on the dive. buddies don't even realise I've got a video system with me, mistaking the housing for a canister lamp.

In terms of depth rating, although the housing can dive deeper than I ever will, the weak point is going to be the glass front to the head-unit camera. So far it's been to 60m with no problems, but I wouldn't want to push it past this point without replacing it with something stronger.

Summary

The system has performed well for a few years now. One bonus is that I capture an entire dive, from kitting up to sitting down on the boat at the end of the dive. This means that I get footage of people falling out of the boat, jumping in without fins, fluffing their ascents etc etc.

At the time (2001), the technology used was the best I could get within my limited budget. Technology has progressed since then with the advent of DVD camcorders and hand-held MPEG video devices. A memory chip capable of recording a 40 minute dive can't be far away, thus reducing the size of the recording device, and the resulting housing.

Saying that, my system is still in use 4 years later, and still giving very good results. There is still nothing comparable on the market.