Argonaut Kraken exhaust loop flooding issue...?

[Couv]

This is why I keep on suggesting how to install the valve to have the best chance of being properly seated.

Agree 100% with Luis here. At least try the sealant treatment before doing any permanent modifications to the spider. Imagine if you take a couple of thousands off of the spider's periphery, would the difference in thickness between the center of the spider and the valve skirt's sealing surface make things worse?

 

[lexvil]

My Kraken has been through a lot, I started with a used one, 2015 model if memory serves, it had the spin off yoke so when I had a chance to pick up one of the newer ones (new) I did, I thought the more conventional yoke retention would be better, at times I miss the old style because it was easier to pack.

I put a lot of dives on my ‘17 before the exhaust issue showed up, I sent it back and Bryan who went through it and couldn’t duplicate the problem, as an aside mine did develop an IP issue where it would drift up and at 175 vent through the safe second. Sent it back for that and Bryan rebuilt it, I’ve lost track of the parts but don’t believe there was any odd wear on the HP seat, my recollection is it looked new. It did drift upward a few more times and I set the IP down to 130 which seems to have solved that.

I still had the periodic exhaust thing which never felt like a full flood but more of a 2” to 3” of water in the hose, once I got the exhaust flow started it cleared but would come back after several othe cycles, never bad enough to call a dive but disconcerting enough that I left it home on my recent Truk trip, I just didn’t want to deal with this at 180/200 feet.

My ‘17 has had other mishaps too which I don’t offer it as any evidence of any real issue with the design and function of the Kraken but do on the inherent toughness.
Last year I had completed two dives with the Kraken and as was the norm before getting a pee valve in my drysuit, took off the rig and left it on the tailgate of my truck, while walking away I heard a crash along with assorted gasps, kept going because nothing behind me was going to change. I had weights and all on the rig, all told probably 85 to 100 pounds, something had shifted and drug it off the tailgate landing on the Kraken, breaking off a few inches of the lip on the top can. When I got home I put the reg on a fresh tank and headed for my pool, functioned fine and no leaks, took it apart and nothing beyond the broken lip, later with closer inspection I saw the yoke and knob stem were slightly bent. I kept on diving it.

A few months later I noticed a crack in the bottom can (pressure side) I had already asked about replacement cans but never pulled to trigger, to shorten this up a bit I found toluene to be a solvent that worked on this plastic by experimenting on the broken lip, pieces went back and seem to be as strong as the original. I sealed the crack with toluene and that all works fine, a little unsightly but strong enough, I never hav and hint of moisture in the incoming air or in the mouth piece.

The exhaust issue is the only thing that bothers me and I never said anything because of the other things this reg has been through would cause doubt as to the reason.

The issue remains unrepeatable in any test I’ve tried but still crops up while diving.

 

[Fibonacci]

I have seen those in the molded piece and I am fairly sure I know why they are happening, but that is not important at this point.

It is a real shame that there is an imperfection, but the imperfection doesn’t seem to be affecting most units. I don’t know if the imperfection varies on the different cans. I don’t have a lot of experience with the tolerance changes in molded plastic parts. But I have never had an exhaust flood (I have hundreds of dives on my Argonauts) and there are hundreds of Argonauts out there that seem to be working fine.

The imperfection doesn’t have any sharp edges/ discontinuities. Therefore, even do it may not be perfectly flat, the soft conforming valve has a reasonable chance of forming a reliable seal. This is why I keep on suggesting how to install the valve to have the best chance of being properly seated.

This is not the end of this. I will be talking to Bryan about it, but in the mean time, please try what suggested.

Morning Luis

I tried pulling down on the arrow end of the valve using a haemostat poked down the exhaust horn to replicate what adding some silicone sealer would do. Didn't seem to improve things much... with a small steel ruler the distortion is easy to spot. I am hesitant to effectively permanently bond the exhaust valve in a place where I can't clean off the old silicone if I wanted to change the valve at some point in future.

Part of my job entails checking and signing off injection mouldings against part specs in conjunction with the supplier.
Moulded part dimensions can change quite radically with subtle changes in parameters.
A parameter report is usually included with first off-tool shots, and a dimensional report submitted for approval.
Moulding parameter reports typically include resin moisture levels, mould core and cavity temperature, pressure, molding speed, cycle time, injection machine barrel temperature, clamp pressure etc etc.

Certain dimensions or areas can be classified Significant Characteristic (SC) or Critical Characteristic (CC).
Typically SC's may include checking with that certain tolerances for length, parallelism or flatness have been met, and check marked with a yellow wax pencil or paint dot. Typical CC's would include safety related items like steering fasteners tightened to specified torque or correct dimensions for the airbag break through area on an instrument panel, these are usually barcoded against each vehicle VIN.

The exhaust valve seat area should have been flagged as a CC equivalent area and each moulding checked at the supplier on a jig to ensure it meets specified standards for flatness and sink marks or rejected. A new batch of resin will often have slightly different characteristics, so the moulding parameters will need fine tuning so the parts continue to meet specifications. So if Bryan is having these cans produced in small batches it is entirely possible dimensions could shift year to year. My Kraken is dated 2017, maybe cans before then are fine?

Believe me the suppliers will try to get away with everything they can if you don't keep onto them...
I once spent a dreary week in Tianjin, China (think of that drizzly polluted noodle bar street scene from Bladerunner, with added sleet) working with our doortrim supplier's toolmaker. Normally these parts are signed off in the country of assembly, at full run-at-rate volume. This is to prevent suppliers making special perfect parts for sign-off at very slow unrealistic volumes... but we were out of time to meet sea freight window.
Anyway they flew a injection moulding wizard over from Spain who performed miracles and confirmed that 'theoretically perfect' parts could be made from these tools.
However when the tools were shipped to the country of production, using different moulding machines, slightly different resins and non-specialist operators we never, ever saw perfect parts again

I do hope you speak to Bryan about this issue, I will also contact him regarding a new exhaust can as this one clearly has a manufacturing defect from new.

My 2017 Kraken is effectively an A$800 paperweight until this issue is resolved.

 

[lexvil]

I wonder if a metal cage can be retrofitted into the old can? My spider looks exactly like yours but at the moment will seal a very slight vacuum once I wet it slightly, when dry it will lose vacuum.

 

[Fibonacci]

I wonder if a metal cage can be retrofitted into the old can? My spider looks exactly like yours but at the moment will seal a very slight vacuum once I wet it slightly, when dry it will lose vacuum.

With the wonders of 20:20 hindsight a replaceable or moulded-in metal spider would probably have been a better execution and not have to force resin through such a narrow constriction. The problems then lie in trying to tune moulding parameters to suit a thin flat square spider without affecting the dimensions of the rest of the can. BUT the original design brief may have had different constraints...

So it is <possible> the tool could be modified to either allow a machined metal spider as a moulded-in insert or designed to be ultrasonically welded to a suitably modified spider boss. The slider used to form the inside of the exhaust horn and underside of the spider could be simplified.
But of course all this costs time and money... Bryan's call.

Modifying the existing exhaust can to accept a metal (Ti?) spider would be much more difficult... but I love an engineering challenge... let me work on it

 

[Luis H]

I am hesitant to effectively permanently bond the exhaust valve in a place where I can't clean off the old silicone if I wanted to change the valve at some point in future.

The clear aquarium silicone caulking I buy from my local hardware store works well but it doesn’t have a huge adhesion property. It is a great caulking, but when it comes to bonding it is not what I would call permanent.

I totally agree that it is important to be able to replace a damaged mushroom valve, and that is why I use and recommend silicone caulking. I have found it easy to remove the old silicone when I wanted to replace and try a new valve.


Added:
I inspected several of my cans tonight. They all have the same marks and from the front they look the same, but when I put a straight edge on them I can’t see light from under it. I also can’t feel anything other than flat smooth surface, but I don’t consider that a reliable test.

The last test I did on the three cans I checked was the light test. There is a perfect seal on these ones.

I think all of these cans are from the first production run, but I don’t have a clue of the production history of the cans.

 

[Fibonacci]

The clear aquarium silicone caulking I buy from my local hardware store works well but it doesn’t have a huge adhesion property. It is a great caulking, but when it comes to bonding it is not what I would call permanent.
~snip~
I think all of these cans are from the first production run, but I don’t have a clue of the production history of the cans.

OK I will try and hunt down some aquarium caulking.
Usually injection moulding production runs are identified via a small date/month tool insert in a non-visible area of the part.
Mold Date Inserts | DME

 

[Couv]

As someone who has worked in aviation maintenance for over 40 years, I can tell you it's a sad fact that all too often a part that has gone bad during use can be replaced by a part that is bad from stock. Try a known good valve in the DBE (rob it from one of the mouthpiece cage valves) and test again. To be triply sure, convict the suspect part by putting it into a known good position and see if you can duplicate the malfunction.

 

[Fibonacci]

Clearly I have too much time on my hands this fine Sunday... but it was fun playing about in 3D CAD

I reckon the OEM Kraken exhaust housing <could> be modified to accept a metal spider, if the original spider and lower support is machined out as the wall thickness is quite heavy in this area.
Part of it is unsupported so needed a flanged profile. Can just package an o-ring seal.
Not recommended

 

[Couv]

....or just a flat seal such as found on an AIR 2