That's when the fun started...

Everyone agreed that a day off was in order.

This meant Ash heading in a straight line with no wavering for a Mcdonalds.

Rick, Mark and Robbie also headed into Rijeka while Rich and I stayed at home drying out and repairing pretty much everything we own.

The next day we were ready to go again, with Rich and I aiming for one last push at the end of Ash's line.

Robbie helps Chris kit up

The whole thing was made harder by the fact that we now had another 207 metres to cover at around 50m depth. It was going to be a chilly dive.

We arrived at the sump pool in good time and I fitted the assumed faulty MAV (manual addition valve) after a service.

After some fettling it now appeared to be working. It was doing everything it was supposed to in pre-dive checks and giving me all the right numbers. It was good to go.

Robbie helped us kit up an I cannot understate the importance and value of having a diver who knows what they are doing, helping you get ready. Little things like dropping dry gloves into the water would mean dive over. Robbie was superb and a huge asset to the team. Things would definitely have been much harder without him.

Ash followed us down to 38 m depth with a Paralenz camera mounted on a tray with Light Motion video lights. He had been given a crash course in videography which was definitely not his thing, but he filmed us thus far and then handed the camera over to me and I continued filming Rich while Ash returned to base.

Rich and I continued until we met the start of Ash's line.

I was totally relaxed and enjoying myself. The visibility wasn't so great and we were moving a little slower than the pervious year as I didn't want to even get close to raising my breathing rate on the CCR.

I filmed for about 20 minutes, finally dropping the camera as the cave began to undulate significantly.

After what seemed an age, we came to the end of Ash's line and, contrary to what he had told us, the cave looked to trend immediately deeper.

I called Ash lots of things out loud and Rich began laying line while I surveyed behind, declaring we had 5 minutes run time more, regarding bail out strategy.

Rich laid another 36m of new line and the cave started to trend deeper, still ongoing, ending at 42m depth. He cut the line free and we packed up and headed home.

That's when the fun started.

Not long after we had begun to head home, I noticed my oxygen percentage was creeping dangerously high and it seemed to get worse the deeper we went. I repeatedly flushed the loop with 15/55 trimix to bring it down, but this only relieved the issue temporarily.

With only a 3 litre bottle of diluent (more than enough to do this dive several times over) I knew that flushing the loop every minute or so would mean running out of diluent - and I did.

It was inevitable that I would have to bail out.

Bailing out seems to be something that rebreather divers are afraid of and they don't seem to want to do it readily.

I had practised it many times and it was really no great drama.

We had 1.5 times the gas we needed to get home - each - plus bailout decompression gas each - so it was no big deal to simply switch to open circuit and dive home.

To this day I don't understand why CCR divers are so afraid of bailing out and if many had done so sooner, they might still be alive.

Chris carrying a Suex scooter on a KUBI scooter back pack, just before sump 2. Image: Mark Burkey

The biggest problem was the camera and light arms. Every few metres it seemed to snag on something so I threw it at Rich who man handled it home.

I switched the oxygen off to help manage the loop volume and had a swift dive home, switching between bailouts and bailout deco gas and we completed a thankfully short 40 minute decompression.

Ash nonchalantly showed up at 6 metres completely unaware of the situation. How, I don't know, but he also seemed to miss the fact that I was no longer on my rebreather!

Not spotting much amiss he carried on down to 25m as instructed and cleaned up the messy stash of bottles at the bottom of the shaft.

I fiddled about trying to get his underwater ipod to work (I couldn't) and Rich caught up with me at 6 metres.

The cause of the oxygen issue had long been thought to be the MAV (manual addition valve, which injects oxygen into the breathing loop at the press of a button) but on closer inspection and a thorough service, evidence of an oxygen fire in the oxygen regulator is likely to be the culprit, causing the interstage pressure to become very unstable. This meant that too much oxygen was being forced into the loop at depth. A high PO2 (partial pressure of oxygen) is toxic to the human body and can lead to a seizure. Nobody has ever survived a seizure underwater.

The whole unit has been serviced and the offending regulator replaced.

Ash, Robbie, Rich, Christine, Rick, Mark