Cis Lunar Mk 6

Tino de Rijk

This is exactly why Narked@90 recently brought out there excellent (above water) cell-checker device - and even that one has of course the limitation that it checks BEFORE the dive, not DURING the dive.

But you could sort of argue that if you test your cell before the dive all the way till a PO2 of 2.5 or so, and it is still linear in the high range after extended exposure time, that it is then statistically not very likely that it will suddenly fall off during the couple of hours of a long dive so heavily that it isn't linear anymore in the 0.7-1.3 operational range.
I don't really like "statistical likelihood" (I still plan to e.g. beat the statistical average male maximum age in Holland....), but it's the best we've got.

The MKVI tests in-dive till 1.6 (6 meters), which is not that far from its high SP. So I am very, very curious in the secret solution Nigel still has in store for us to address that..!

However, on the positive side, in the MKVI this happens during EACH dive, while it is not so likely you will remove & test your cells before each dive and test them in a cell-checker; that is typically something you do before a week's diving or so (or in the rental market, as part of a disclaimer towards the renter). The two complement each other nicely.
"Plan for the worse, hope for the best"....

Ciao,

Tino.

dave t

Agreed Tino but I suspect Nigel's point would be hand held analyzers can be very in accurate

Dave

dave t

Again I agree and John and Brent's checker is a must have but and this is my original point about the cis, if you have manual control of your breather you can cell check as and when you want. If my setpoint is 1.2 and the cell has gone current limited at say 1.9 what do I care. adding a small amount of o2 to push my cells above my chosen setpoint when ever I want will always afirm that my cells are capable of holding my setpoint on that dive

Dave

Tino de Rijk

True, especially if the cell in your analyser is old - and so we have sort of a full circle, as we started this all to make cell validation better.....

In all practicality I would take in the manufacturer's statement as starting c.q. maximum likelihood point, but still double-check with my own analyser.

What I do (and AFAIK I learned that from you, Dave!) is use my Rebreather as a glorified 3-cell analyser, as follows:

• I calibrate my (in this case Inspiration) rebreather at home shortly BEFORE the trip, take note of the ambient pressure (barometer).
• when getting on-site abroad, I then say "calibrate? No", when the system wants to calibrate, as such effectively forcing the cells to use the calibration correction values stored in the unit from the previous calibration at home with a known purity oxygen.
• Next I expose the as yet unknown local O2 to it, and have all 3 inboard cells that were calibrated before at home to see what they tell me. I simply do that by switching to high setpoint at the surface, and let the solenoid inject for 1-2 minutes or so, as such giving the cells a good flushed "oxygen bath". You need to correct a bit for ambient pressure, but that's usually not that much.

Again: not "scientifically pure and 100% accurate", but in my experience better than using the local five guide's over-used and shot analyser with a 3-yar-old cell.....

Ciao,

Tino.

Tino de Rijk

Fully agree.

silent running

Hi Rich, (or Nigel), it occured to me that in order for the unit to do it's inwater checks to 6M, the diver would probably have to do a very slow descent and stop at 6M for a period. Is this right, and if so, what happens if the diver has to do a fast decsent, say in a current where getting down quickly can mean the difference bewtween finding the wreck/reef and being blown out into the blue? Would the unit force you to bail to OC if the descent was too fast for the auto val system to do it's checks? Thanks, -Andy

onetime

VERY interesting thread thus far. Some definite features to the MKVI that have potential to truly change the market.

One question for you though Rich, the argument against the manual O2 add strikes me as a bit odd. While the goal of making CCR as safe as OC is a noble one, the comparison, when it comes to O2 manual addition, doesn't fit since the death rate of manual CCR divers is virtually nil. It would seem, based on the 500 or so KISS units out there that the manual add hasn't proved to be a major risk.

Can you share any details on what was perceived as the risks that pulled the manual O2 add feature from the unit?

Thanks.

--Paul

deepreef

Hi All,

I'm on a VERY brief break from fish-nerd stuff. Nigel has covered most of the questions, but I'll jump in here very quickly:

The difference is that the sort of diver who dives the KISS is not the sort of diver who Poseidon targeting the MK-IV for.

I return again to "Richard's Reliable Rebreather Paradox": that a more reliable rebreather can be more dangerous, because reliability encourages complacency, and as we all know, complacency kills rebreather divers more than anything else.

By this metric, manual rebreathers are the safest, because the diver has no choice but to always be cognizant of the PO2. A KISS diver is, almost by definition, an excellent rebreather diver, and one who is almost immune to the complacency killer.

What has made being involved with the MK-VI development team so interesting to me, is the challenge of overcoming RRRP. Is it possible? If you'd asked me two years ago, I would have said "No". Obviously I'm biased, but after the design discussions we've had these past two years, I believe that the MK-VI is closer than any other rebreather ever has been. Close enough? Only the ongoing extensive testing will tell for sure, but early indications seem to be good.

The goal is to come up with a control system that is as consistently reliable as a well-trained KISS diver. If we can achieve that, then we've not only outrun the other guy, we've outrun the bear (reference to my earlier joke about two guys walking in the woods) -- which comes back to the point you made, and that many of us have noticed for more than 10 years: homebuilders and KISS divers just about the best track record for avoiding accidents.

So, in this context, we need to re-evaluate the cost/benefit of a manual O2 addition valve. The KISS diver obviously needs to have one, and clearly the benefits exceed the costs. This is because the KISS diver is intimately familiar with the O2 addition valve, and certainly knows how not to misuse it. But Poseidon's target is not this sort of diver. The target is the sort of diver who does recreational OC dives in shallow water. The goal is to have a CCR system that is as consistently reliable at keeping a diver alive as basic OC SCUBA. Based on an analysis, we've decided that the cost (in terms of potential safety hazard -- not in terms of money or design considerations) of a manual O2 addition valve in this paradigm outweighs the benefits.

Empirical data could certainly change this equation, and it may well be that even in the MK-VI diving paradigm, the benefits of a manual O2 addition option may exceed the costs. In this case, the monetary costs of adding one are trivial, and the design implementation costs of adding one are trivial. The reason the current design lacks the feature is not due to design constraints or to monetary costs -- it is due only to our prediction that in the target market and diving paradigm, the costs outweigh the benefits in terms of ensuring the consistent survival and good health of the diver.

As to the ASV (Active Sensor Validation) system, the points brought up by Tino and Dave are well understood by the MK-VI design team, and have been thought through in great detail. As Nigel alluded to, we're not in a position to discuss some of the stuff just yet -- but it's really not that complicated.

Things to keep in mind are:

1) Validation tests can be performed very quickly (limited mostly by the response time of the sensor, and the flush latency of the sensor chamber, both of which are measured in seconds);

2) It doesn't take many data points to be able to perform a high-reliable validation test (one at the surface and one at 6m [or 8m, or 10m, or whatever depth is found to be optimal] is enough to give you most of the information you need).

3) Incorrect gas concentrations (e.g. 94% O2) that are both small enough to be missed by the initial surface calibration as being out of bounds, but large enough to have some non-trivial impact on life-support (either hypoxia, hyperoxia, or DCS), *can* be detected with high confidence. The trick here, as I tried to point out earlier, is that there are multiple parameters that all need to align correctly, and they do not fail in ways that cancel each other out across the multiple sets of parameters. Put another way, you would need to have BOTH gases (O2 and Dil) wrong by exactly proportional amounts (highly unlikely), *AND* the depth sensor would have to fail in a way that exactly offset the simultaneous failure of an oxygen sensor. Now we're dabbling in the probability realm of shark attack and alien abduction.

3) What we have described in public so far is the SPORT unit. This means that we don't care WHY the numbers don't agree -- we just care that they DO disagree -- at which point the diver is sent safely to the surface on OC, where the system can later be interrogated for the cause of the problem.

4) 1.6/20m are just numbers. The point is that the system architecture is capable of MANY different validation tests, using whatever sets of numbers the empirical testing reveals to be optimal (in the context of various sensor failure modes, etc.)

So, the one point that Dave and Tino jointly brought up is the scenario where the O2 sensor is validated as linear out to X.X during initial descent, but then somehow becomes voltage-limited *during* the course of the dive while the diver is deeper than depth Y. As Nigel alluded to, we have thought this situation through. And as he also alluded to, we've not yet been cleared to describe the solution. But really, folks -- it's not all that complicated. All the information you need is in the whitepaper. The only advantage we've had is a lot more time to think through all the implications in greater detail.

And if ALL else fails, keep in mind that there is the second O2 sensor watching over the shoulder of the first O2 sensor.

Alas, I've blown my non-fish-nerd time budget for the week...

Aloha,
Rich

__________________
Richard L. Pyle, PhD
Database Coordinator for Natural Sciences
and Associate Zoologist in Ichthyology
Department of Natural Sciences, Bishop Museum
1525 Bernice St., Honolulu, HI 96817
email: deepreef@bishopmuseum.org
http://hbs.bishopmuseum.org/staff/pylerichard.html

Sutty

Surely if you have the initial calibration/validation data if you want to check that the primary cell is not current limited all you need is a tiny puff of O2 onto it at depth (eg when you want to add anyway) and check that the mV rises to well above the high set point predicted mV? A bit like a Kiss diver knowing his cells can read above "set-point" by watching the O2 blip above set point for a few seconds after adding O2.

__________________
Never forget that life is a finite resource.

Andy M

Did you not read the poseidon white paper?

The Cis Mk6 has three solenoids. It has O2 sensor test, dil sensor test, and O2 replenish solenoids - so yes, it puffs O2 -and- dil onto the primary cell during the dive. It can also inject dil as well as O2 to cater for O2 being too high so its a completely different control system to current RBs.

Absolutely fascinating and novel, this must have taken quite some time to model, program and test.

I'm interested to know what sensors are being used in this system.

__________________
Know your PO2 at all times