I LIKE TO DIVE DEEP. Its rewarding to do something difficult and return safely. But in February 2003, I almost did my last dive.
This 260m dive was intended as a practice for an even deeper one. The ascent plan was aggressive timewise, but I had confidence in my chosen, commercially available, deeper for longer decompression algorithm. As it turned out, this schedule proved woefully inadequate, and my injuries will probably take a lifetime to recover from fully.
My doctors advised me against diving again, but at 34 a career of 10 years and nearly 3000 dives was far too difficult to discard. If I could not return to the sport I loved, the inevitable depression would be harder to endure than any physical injuries.
Over the months my health improved, and I became proficient at dive-table design. I worked out a plan for a 320m dive which I felt dealt with all the weaknesses of the earlier one.
Building dive plans is not rocket science. All the information needed is freely available and was tested long before most of us were born. I worked with a knowledgeable computer programmer on software that incorporated data tested by commercial divers in the past; information from recent unsuccessful deep dives; and matrices to avoid counter-diffusion problems (caused by the different rates of diffusion of inert gases).
A dive below 300m needs a rapid descent. The resulting HPNS (High Pressure Nervous Syndrome) can be minimised by using a high END (Equivalent Nitrogen Depth) value - in my case more than 70m.
I would keep oxygen high (ppO2 was 1.6-plus) because the exposure was short. Keeping the helium as low as possible in the bottom mix has many benefits, and makes it easier to derive the next trimix decompression gas.
On the dive my nitrogen spike would be down deep, when critical tensions were not yet high. The maximum-depth gases would be largely helium with 20% nitrogen and around 5% oxygen.
The 140m deco gas would go up to 10% nitrogen with a rise of 6% oxygen. Subsequent ascent gases to 9m would retain the same helium content, increasing oxygen at the expense of nitrogen.
At 9m, heliox would be used for fast decompression.
I would manage oxygen toxicity by keeping the ppO2 low from 6m. No ppO2 breaks were needed - I consider them potential suicide on a trimix/heliox dive. Trimix or heliox ppO2 breaks are as bad.
Avoiding such breaks is difficult from a pulmonary toxicity point of view, and breathing heliox on open circuit for hours is hard, as the body wastes so much energy heating the gas. But I chose heliox to avoid counter-diffusion problems.
To combat the enormous dehydration due to immersion diuresis on this almost-seven hour dive, I would also have to drink 2-3 litres every hour, again difficult.
The OTU (Oxygen Tolerance Unit) and CNS (Central Nervous System) count on this dive would be high, but managed by not chasing 1.6 ppO2.
As it turned out, a drop in vital capacity would occur for 25 days after my dive, but this may have been lung fatigue from breathing unheated, unhydrated gas for long periods, or pulmonary toxicity.
Many extreme explorers seem to believe that strength or fitness rather than technique will somehow break down brick walls. Previous deep-dive attempts seem to have followed a familiar pattern: drop to maximum depth like the US dollar and then ascend like a jet fighter on afterburners.
I always thought this wrong. Rapid ascents store up problems for later, yet too long spent in deep water could cost me as dearly as it had before.
It now seems that stopping deep for short periods, typically less than 30 seconds, can have dramatic benefits on overall decompression, though too many stops below 200m add heavily to the burden. I chose a relatively slow ascent rate from the bottom of around 18m/min. The first stop would be at around 250m, with the next four within 18-20m of each other.
I planned and planned, fine-tuning the gas mixtures. Ten months after my accident, I felt confident and fit enough to make the 320m attempt.
The tanks for me and 14 support divers took three full days to fill with almost 60cu m of helium. This was completed just two days before the dive - too close to deadline, as the process was as stressful as loading the diveboat during a squall.
I had had insufficient sleep and, although I normally feel quite optimistic, would have made an excuse to postpone the dive, had gale-force winds not done the job for me.
I felt incredible relief, but postponing was expensive. I could afford only one more attempt and would then have to shelve the project.
The next window of opportunity with low tidal movement was a fortnight away - a week before Christmas. I didnt relish the thought of Christmas dinner in a chamber, and still had nagging doubts, but I decided to go for it.
As the days passed, the storm clouds cleared in my mind and I managed several practice dives in less-than-ideal conditions.

Mark Ellyatt prepares for his big moment
The night before the dive I went to bed early, mindful of the windy conditions, but woke to blue skies and calm seas. I felt as carefree as a 6m support diver!
The journey to the site, which lies 35 miles off Phuket in Thailand, on the edge of the continental shelf at 450m-plus, took four hours over smooth seas.
The dive and rescue boats came courtesy of Scuba Cat, for which I was working. The support divers were those who had been great in February, when they had held me on the line as I convulsed, vomiting, for three hours because of counter-diffusion- induced DCS.
The descent went very smoothly. Darkness fell at 180m, but nowadays with the right lights its possible almost to guarantee operation to 300m.
My head-mounted lights shone onto the weighted line as I checked the metre markers against my depth gauges. By 250m I still felt in charge, my lowish helium content adding the sort of confidence that would help avoid helium tremors and other HPNS symptoms.
My drysuit was keeping the icy water out. I noticed shakes in my body and hands, but couldnt say whether it was the cold or helium-induced tremors.
The regulators had been chosen for their heavyweight all-metal construction, helping to maximise internal temperatures. The DFC (Dynamic Flow Control) system is great at managing high gas flows, with its smooth operation as opposed to venturi flow support.
And no environmental seal means less intermediate pressure amplification with the ensuing high-pressure seat instability. Water temperature at depth was 3-4C and the gas flows enormous, but I experienced no free-flow or stutter.
By 280m I started to grip the line tighter to slow down. I checked my back gas contents gauge and at 300m found that I had already reached my turnaround pressure. I didnt seem overly concerned, and that concerned me!
At 310m I was 20 bar behind schedule and a minute over my planned drop time. I looked down and saw the ghostlike image of some kind of large hydroid.
I scanned from left to right to check for visual abnormalities - and to check my distance from the jelly-like visitor. These little checks suggested that I was stalling into a complacent mind-lock, which set the narcosis alarm bells ringing. The hydroid looked to be more than 2m long and set to hit my descent line!
I was deep enough. I grabbed the line marker at 313m and headed for 249m, the first deep stop. I ascended hand over hand, eyes shut, not wanting to see the contents gauge. With twin 20 litre tanks, you have a virtually inexhaustible supply. Even when they equalise with the surrounding water pressure at 26 bar you simply head up 5m, and because of the huge volumes involved you can breathe again and again.
The deeper stops came and went without drama because of the slow ascent rate. The first gas switch was my only concern, but with the minimal time at depth, the risk was acceptable.
At 90m Sveinung handed me a 15 litre tank of trimix 14/56. Phil gave me a similar tank at 75m, as did Khun Gai at 60m.
With members of the support team at the Scuba Cat dive centre
By 10m I had started to shiver, perhaps due to the high helium content in my drysuit-inflation gas. At 9m the long heliox stops began, gas supplied from the surface through long regulator hoses.
A drama occurred two hours later, when the tank I was breathing was changed at the surface. I still had the regulator in place when I felt my tongue being sucked through the second stage and could see the hose collapsing to the surface due to the vacuum! I pulled my now-much-longer tongue free and cursed the topside monkeys. Very unpleasant, but it could have been far worse.
During the stops, I ate fun-size Mars Bars and banana pieces. I knew the support divers were having their own lunch, because chicken bones were raining down around me and small fish closing in.
I got out of the water after six hours 36 minutes. I had gone deeper than any solo diver before me and surfaced under my own strength.
I did sustain a shoulder bend, though this resolved itself. And doctors tell me that after the age of 40, as a result of such dives, I can look forward to suffering from osteo-necrosis. But for the time being I was exhausted but very happy.
  • Mark Ellyatt can be contacted at mark@