No air integration in high-end and tech DCs . Why ?

[Jim Lapenta]

I am not going to pay for a bunch of transmitters. I am not going to pay a couple hundred bucks for an SPG. Which while the price MIGHT come down eventually it is not going to be cheap starting out. You have to redesign the case, create new tooling to make the case, test the hell out of, make it in different sizes that tech divers prefer, and insure that it can be quickly and easily replaced in the field with minimal tools should it prove necessary. So either you get another one (or a couple to have as spares) or you switch the offending gauge for a basic one that should have been on there in the first place.

Not to mention testing the hell out of the computer to make sure there are no bugs (see the x deep black threads on here) so that it is not doing stupid stuff that it should not.

Tech diving is about keeping it simple. As much as is humanly possible. With current technology, what you are proposing not only violates that principal but assaults it in a way that is obscene in my book.

 

[GrumpyOldGuy]

GrumpyOldGuy,

I think you misunderstood the proposal: having a digital transmitter and piezo included in an otherwise classical analog SPG hausing means that:
- you do not have to touch the computer any more when switching, as the gas switch is automatically visible by the DC and registered
- failure of the digital pressure gauge or xmit module does not affect the analog SPG (one device being electronic and the other mechanical by nature, they will not interfere with eachother)

Planning is justified during trainings, I suppose (never done tec dives), and the need for planning will not just vanish because some extra data is available on the DC & logs. Or am I overlooking something ?

--

For those who fear that passive radio com can reach only a few inches distance, and that it works only in the total absence of RF noise or any other devices, I invite them to brush up on radio fundamentals, and wonder: do you get connected to your neighbours wifi by accident because they live too close ? Are you unable to use yours when they switch theirs on ? Do you log your collegue out when you present your card to the office reader ? A number of toy electronics kit offer to build crude AM radios that are powered only by the antenna and they can catch stations that are 100s of miles away.

Yeah, I missed a little but not as much as you assumed. I have built crystal radios as a kid and am a licensed ham operator (not many of us left). I love to dick around with electronics, but just not so much while diving as water and electronics are a poor mix. I am not sure about how a passive receiver would work underwater, but I think that is besides the point. The DC needs a power supply so the receiver is not going to bother anything. As for radio signals over powering each other, we used to do it all the time. It was a constant problem with ham radios (picking up close by transmission on TV and such). But we are talking analog and pretty high power which is outside this subject.

You could add a transmitter to a analog gauge and make it about 5X more likely to fail. You would gain some redundancy (because even analog gauges stick at times), so its not all bad.
More parts almost always equal more failures. You need a power supply for a transmitter and these tend to be the weak link. Batteries leak acid or overheat, both a failure mode that can damage nearby parts.
Flooding is a catastrophic failure for electronics, not so for analog. You would need to keep the 2 parts completely separate and add access for battery replacement, its not just adding some parts inside the housing.
Lastly, traditional analog SPG's flop around on the end of a hose, they are essentially abused because you can. I have tanks and such bang into mine on a regular basis. The same SPG with electronics mounted on it might not be so forgiving. The standard electronic pressure transmitter is mounted to the 1st stage where is it spared most abuse (except idiots lifting the tank by the transmitter - nothing is foolproof).

 

[Dr. Lecter]

Please buy yourself a "high end" Suunto/ScubaPro AI "tech" computer and leave the rest of us in peace. TYIA.

 

[guyharrisonphoto]

Ideally, yes. But the ability for the computer to continue to function without the AI input is something that has to be engineered into it. This is a feature, and like any feature of any system, it is at least theoretically subject to failure. If poorly engineered, it is not impossible for a failure of the AI input to affect some other aspect of the computer. The only way to guarantee that a failure of one system will not affect the other is to separate the systems.[/QUOTE]

You are right at the theoretical level, but, in the real world, I have not heard of a single instance where the loss of the AI signal or failure of the transmitter actually caused entire computer failure or the loss of deco calculations. Such a catastrophic flaw would have been reported all over on every forum imaginable. Transmitter drop-outs are not the norm but they are not super uncommon either (the signal is most often regained within a few minutes) and total computer failure just does not happen as far as I have heard. Thus, I would not rate it a concern as opposed to the traditional causes of entire computer failure (floods, battery leaks, damage). In other words, I think the engineers have effectively done their job on this issue.

This is no to discount the special reasons many have stated above for why tech divers forego AI (cost, bulk, no need due to gas planning, no need for a computer at all), but the risk of AI causing deco calculation failure realistically is not one of them.

As one poster above noted, tech AI computers (I have a Galileo) can, if all tanks are equipped with transmitters, take you through a dive, and show the pressure of the tank you are currently breathing on your wrist for convenience. For divers that spend time planning, the extra minutes setting up the computer don't seem like much. You will also have spgs on each tank for manual back-up and confirmation. Whether a tech diver wants to do this is up to them, but losing deco calculations should not be a deciding factor.

 

[boulderjohn]

The biggest problem with AI in my opinion is that tech divers carry multiple gasses. Backgas, Deco gas, and sometimes More deco gas. Imagine having a decompression computer that has to be wirelessly connected to 4 different gas sources and having to differentiate between each one. It is more foolproof to use analog in this scenario, not to mention a hell of a lot cheaper and more reliable.

There was another similar thread less than a year ago, and someone did identify a computer that does identify multiple gases from multiple transmitters. I did not think there was one on the market yet, but apparently there is. I believe it is a Uwatec. In this case, you would have to tell the computer you had switched, as is true now with multi-gas computers.

A couple of years ago I was doing some dives on a boat that included a group that was testing a computer that not only tracked multiple gases from multiple transmitters, it knew when the diver switched. I don't know how--perhaps it responded when one tank stopped showing gas loss and another started showing gas loss. One of the divers was the computer's designing engineer, and the others were just testing it. It sounded like it was being reasonably successful from what I overheard. I have no clue what brand was involved.

So, if you get in the water with what you KNOW is enough gas, plus; and you do the dive you planned (which is part of the discipline of technical diving) then the SPG becomes just a DOUBLE-check on the accuracy of your planning, not a primarily driver of dive decision-making. You certainly wouldn't want to dive without one, but knowing your pressure on a continuous basis just isn't critical. For a tech diver, the $600 or so you'd spend on a transmitter translates into a couple of charter boat fees, or several tanks of high helium mixes, or a whole boatload of sob . . . money's better spent there, than on a fancy gadget that doesn't add much to your diving.

That's where I am. With my back gas, the gas I am doing for most of the working part of the dive, I really only need to check it occasionally for confirmation that things are going as planned, and part of that plan is having a whole lot more of that gas than I should need. As for deco gases, their spgs are right in front of me, so it is about as easy to check their levels as it is to look at a computer on my wrist.

 

[OkByMe]

As far as a dive computer used for tech diving, you have already noticed that analog SPGs are probably more bulletproof than AI. You also have to consider that AI transmitters have a tendancy to fail, and if you dont have an anolog backup, you are diving blind. Some Tech divers choose to use a high end deco computer, and have a wireless AI computer as their backup. This allows them to have digital access to their backgas pressure, while still having an analog guage. The biggest problem with AI in my opinion is that tech divers carry multiple gasses. Backgas, Deco gas, and sometimes More deco gas. Imagine having a decompression computer that has to be wirelessly connected to 4 different gas sources and having to differentiate between each one. It is more foolproof to use analog in this scenario, not to mention a hell of a lot cheaper and more reliable.

The Galileo Sol with TRIMIX will monitor up to 10 tanks, with or without transmitters, all with different gas mixes and will suggest gas changes. The smart transmitters will not time out (not using the gas) once the valve is turned on and will sync when that happens. You can turn off the tank valve if needed and the transmitter will keep sending, making that gas part of your dive supply. The computer will show you the battery level too. The only thing is... if you leave the tank, and swim away, you will lose connect and lose the gas pressure as part of the calculation for remaining dive time and gas availability, it will still be available for gas switching though.

Since I back mount and side mount, I have all my gas with me, a non issue. The transmitters are pricy too. I also use button gauges on the regulator... just in case. I always have a gas plan to setup the Sol with and have it with me if needed, if a transmitter fails me.

 

[Dr. Lecter]

I have not heard of a single instance where the loss of the AI signal or failure of the transmitter actually caused entire computer failure or the loss of deco calculations.

When your AI stops working, why are you so sure it's merely a transmitter connectivity issue and not a problem with the computer itself? Long story short, you can't know for sure and whether you want to call it super unlikely or not, any failure by any part of the computer renders the whole computer suspect.

Which is a great reason not the build the AI into actual tech computers at all, whether you decide to stick oversized, expensive, fragile transmitters on all your 1st stages or not.

The only value AI offers is ease of capturing gas consumption data. That's just not enough value add to justify the added cost and complexity/uncertainity they introduce into the system. You have a hardon for knowing your gas consumption on each dive, then make note of your starting and ending pressures. I tend to be too lazy to care that much about it, though

 

[Lorenzoid]

You are right at the theoretical level, but, in the real world, I have not heard of a single instance where the loss of the AI signal or failure of the transmitter actually caused entire computer failure or the loss of deco calculations. . . .

I absolutely agree with everything you said. For the most part, dive computers are well engineered, and they're not all THAT complex. That's why I said that the bit of engineering that keeps the computer functioning even if the AI link is lost, as straightforward as this bit may be, is only "theoretically" subject to failure.

I don't know what all the issues might be, and the engineers who design dive computers no doubt have looked into a lot of potential issues. But who can say whether, for example, they looked into what happens if some component doesn't just fail outright but fails in some kind of unexpected way, maybe injecting spurious signals or something that some other bit in the computer interprets in an unexpected way. Again, it is all quite unlikely, but theoretically possible. I'm just trying to make the point that the only way to absolutely ensure that one system doesn't affect another is to truly isolate them. The OP is suggesting not just integrating in the same way as most manufacturers but integrating in a novel way. Being a pioneer is not the way to ensure robustness. Rather, it seems to me that true tech-oriented computer manufacturers do their best NOT to break new ground and to keep things as simple and basic as possible. Put the money into making a simple computer as bulletproof as possible rather than adding features.

I suspect we will see more sophisticated computers, rebreathers, etc., that make diving a no-brainer, just like we will also see driverless cars. All in good time. I can't blame the relatively few dive computer manufacturers who have worked so hard to establish a reputation for bulletproof quality not wanting to push the envelope and potentially harm their reputation.

 

[hroark2112]

I have a T3 with AI that I use when diving open circuit. It's a nice toy. I check it from time to time to make sure I am where I should be on my tank.

I would not want it on my Petrel. It's big enough already and I want it to tell me exactly what I want. Nothing more. For rebreather diving, I don't need to know that I used 400# from my O2 and 200# from my diluent supply. Tell me my ppO2 and my NDL time, depth, dive time etc. AI would just annoy me.

 

[guyharrisonphoto]

Dr. Lechter. My AI has, thus far, always functioned properly, so has the computer. Before I got AI, I knew a lot of people that used it and none of them indicated that a loss of signal was a sign of computer failure or affected the deco calculations in the slightest. Nor has it ever been reported to do this on any forum that I have ever visited, and we all know that such a failure would go viral all over the globe if it was actually happening. So no, a loss of a signal or pressure reading would not make the whole computer suspect.

It seems evident that, because any AI computer works just fine as a regular computer without a transmitter and without showing pressure, that it's "regular" computer functions are not affected one way or another by the AI.