Deco cleared "on the go"?

[scubadada]

This question goes on and on, why doesn't it stop, so to speak. For a non-stop dive, surfacing is only controlled by the GF high. Once you have gone into deco, i.e. not able to make a direct ascent to the surface at or below the GF high, the GF low will control the depth of the 1st stop.

 

[doctormike]

e.g. on the dive in question there is a ceiling but no first stop since the ceiling clears by the time diver gets there. GF Lo never kicks in.

Second nitpick of the day. GF Lo does determine the depth of that ceiling, even if it clears by the time you get there. And your computer will generate a stop (depth and time) once that ceiling is generated. The stop length will be determined by your current tissue loading and your gradient factors. The computer makes the assumption that you will immediately ascend to just below the ceiling when calculating that stop length, but it is continually recalculating. So if you ascend slowly but not too slowly, the stop time may steadily drop and the ceiling may clear.

 

[Jay]

Second nitpick of the day. GF Lo does determine the depth of that ceiling, even if it clears by the time you get there. And your computer will generate a stop (depth and time) once that ceiling is generated. The stop length will be determined by your current tissue loading and your gradient factors. The computer makes the assumption that you will immediately ascend to just below the ceiling when calculating that stop length, but it is continually recalculating. So if you ascend slowly but not too slowly, the stop time may steadily drop and the ceiling may clear.

Yes, and it's only GFHi that effectively controls it (for all intents and purposes). I note you did note your comment as a nitpick, and for those that feel like more reading ... here's a recentish link:

On a NDL dive, which computers' NDLs are not affected by GFLo?

 

[dmaziuk]

The computer makes the assumption that you will immediately ascend to just below the ceiling when calculating that stop length, but it is continually recalculating.

Well, we'll have to ask OP whether his computer showed him NDL of 0 on the posted dive. I can tell you that mine didn't on the profile below. According to it I was not in deco at any point, regardless of what Subsurface may think, there was no stop and no gradient factors were involved in its calculation.

 

[doctormike]

Well, we'll have to ask OP whether his computer showed him NDL of 0 on the posted dive. I can tell you that mine didn't on the profile below. According to it I was not in deco at any point, regardless of what Subsurface may think, there was no stop and no gradient factors were involved in its calculation.

View attachment 520178

OK, so then there is a software problem either with your dive computer or with Subsurface, right? A ceiling indicates NDL<0, right? I mean, it's not a question of interpretation, it's just a definition. So I don't know what you mean when you say you weren't "in deco" if you really had a ceiling. I guess that you were saying that your DC didn't give you a ceiling during the diver, right?

You may have had a temporary deco obligation that cleared on the way up, but if a DC during a dive generates a ceiling, that means that the diver is past NDL. Maybe that ceiling on the graph was erroneously generated, I can't speak to that. Maybe you had different conservatism factors (or mix) entered into Subsurface than you had on your DC? I don't know if Subsurface generates that ceiling on its graph when the DC tells it that says there was a ceiling, or if it calculates the ceiling based on the raw data from the DC along with whatever you tell Subsurface about your mix and GFs.

 

[atdotde]

OK, so then there is a software problem either with your dive computer or with Subsurface, right? A ceiling indicates NDL<0, right? I mean, it's not a question of interpretation, it's just a definition. So I don't know what you mean when you say you weren't "in deco" if you really had a ceiling. I guess that you were saying that your DC didn't give you a ceiling during the diver, right?

Not necessarily. The ceiling shown is a statement about your current tissue loadings and not about your current depth ("With the current tissue loadings, the deco algorithm --- possibly Bühlmann possibly with gradient factors --- does not allow you to be above the indicated depth, 1.9m in the display"). But the NDL (as its cousin the TTS) might factor in an ascent rate and thus the off-gassing that your would do during that ascent.

And for the same reason, stating an NDL is somewhat bogus (as is the depth of the deepest stop) since it depends on the assumed ascent rate. You could well ascent slower and then never encounter a mandatory stop because your tissues cleared on the way up. People should really not make such a distinction between non-stop dives and dives with a few minutes of mandatory stops. They are really not that different. Even on a non-stop dive you should not exceed the assumed ascent velocity and one or two minutes of formally existent stops (as far as your computer is concerned) often clear before you reach that depth on a somewhat decent profile. All the fuzz about different deco models and parameters is really about dives with a significant amount of mandatory stops. And yes, of course I can understand the point of recreational training agencies that don't want to deal with customers sueing them because they did not follow what they were taught.

 

[dmaziuk]

OK, so then there is a software problem either with your dive computer or with Subsurface, right? A ceiling indicates NDL<0, right?

What @atdotde said.

To add to confusion: another variable in the mix is the half-time of your leading tissue compartment. On my computer the fastest is 2.5 minutes whereas subsurface uses 4 or 5, I forget. What that means is the ceiling calculated by my computer would have been much more massive than that on the picture -- yet its calculated NDL was in upper single digits at the deepest point and already back to double-digits where the ceiling appears on the subsurface's plot.

That's why I asked a few posts back "where's the deco"? That transient ceiling would have been it if you were to instantly teleport to the safety stop depth at that point in the dive. Otherwise it's just like NDL: you don't like it, come up a metre and wait a couple of minutes.

 

[doctormike]

I'm not sure I understand the point that you guys are making. If you have a ceiling, then by definition you have exceeded the NDL based on whatever you have chosen for algorithm, conservatism factors, mix and profile. Yes, ascent rate will determine the actual ceiling you hit on ascent, so if you have a shallow ceiling and a brief stop generated by your computer and you ascend slower than 30 FPM, it will clear. If you ascend slower than that, it will get deeper and the stop will get longer. So while it is true that your deco obligation will either decrease OR increase if you do a slow ascent, I'm not sure I see why you are making this point.

If you are teaching a new diver about the problems with inadvertently generating a ceiling by overstaying at depth, I don't see how it helps to minimize the implications of that (getting back to the OP). A ceiling is a ceiling. It means that unlike a dive within NDLs, you no longer have the option of a standard ascent without incurring an increased risk of DCS. Just telling them to ascend slowly and the ceiling will probably go away is sort of like telling them just do the stop and the ceiling will go away. It's unplanned staged decompression by an untrained diver, and saying that's OK is normalization of deviance.

 

[jgttrey]

A ceiling is a ceiling. It means that unlike a dive within NDLs, you no longer have the option of a standard ascent without incurring an increased risk of DCS. Just telling them to ascend slowly and the ceiling will probably go away is sort of like telling them just do the stop and the ceiling will go away. It's unplanned staged decompression by an untrained diver, and saying that's OK is normalization of deviance.

@doctormike, I think we are having a terminological difference only regarding the meaning of "ceiling."

I totally agree that we should not encourage recreational divers to casually incur stops, hoping they will burn off during ascent. But, that's different than a ceiling. Where I disagree is that a ceiling "means you no longer have the option of a standard ascent...." I think that is the definition of a stop, or at least a slower-than-normal ascent in lieu of stop, not a ceiling.

A ceiling is simply the depth, at any given moment of the dive, at which your leading compartment would bump up against the limit of what is permitted in whatever model you are diving.

Many recreational dives will, perfectly within the "rules," generate a ceiling without generating a stop because a stop will only show up if the ascent rate used in the model is insufficient to clear the ceiling before you get there. We have ceilings all the time that clear during a normal ascent rate. A recreational diver would be oblivious to the existence of this theoretical ceiling unless it gets to the point that a normal ascent won't cure it. When a normal ascent won't clear it, then you have a stop.

For example, just playing with Subsurface at a 30/75 GF, a dive to 100' will generate a ceiling in only 10 minutes (the ceiling is 1 foot, but it's there). There's no deco stop because -- at a standard ascent rate -- that ceiling clears almost immediately and has been gone for almost 2 minutes by the time you get to it.

I think that discussion is useful for recreational divers, assuming they are interested enough to follow it. It is common for them to disregard ascent rate warnings on their computers -- if they even know what they are. Understanding that they may indeed have a ceiling and they could bust it if they don't honor the 30'/min ascent rate is useful knowledge -- it is the explanation for why the ascent rate warnings actually mean something.

That's why I dislike the term NDL, because it incorrectly implies that decompression is something that only exists when you have stops and therefore as long as you don't bust the NDL you're free to shoot to the surface. It would be more accurate if we called it a "No Stop Limit".

 

[KenGordon]

I'm not sure I understand the point that you guys are making. If you have a ceiling, then by definition you have exceeded the NDL based on whatever you have chosen for algorithm, conservatism factors, mix and profile. Yes, ascent rate will determine the actual ceiling you hit on ascent, so if you have a shallow ceiling and a brief stop generated by your computer and you ascend slower than 30 FPM, it will clear. If you ascend slower than that, it will get deeper and the stop will get longer. So while it is true that your deco obligation will either decrease OR increase if you do a slow ascent, I'm not sure I see why you are making this point.

If you are teaching a new diver about the problems with inadvertently generating a ceiling by overstaying at depth, I don't see how it helps to minimize the implications of that (getting back to the OP). A ceiling is a ceiling. It means that unlike a dive within NDLs, you no longer have the option of a standard ascent without incurring an increased risk of DCS. Just telling them to ascend slowly and the ceiling will probably go away is sort of like telling them just do the stop and the ceiling will go away. It's unplanned staged decompression by an untrained diver, and saying that's OK is normalization of deviance.

No disrespect, you are a fine poster, but you should try to be more humble about the nature of your understanding of how all this works. You have an imagined definition of NDL which may not be correct.

Personally I was baffled by the GF thing, especially how it chose the depths to interpolate between. I happened to have access to an expert to asked a bunch of questions, some of which I got convincing answers to. Then I wrote a load of code to implement the scheme and compared my results to various other plans. When it didn’t match I went back to the code and added a bunch of switches to control the ‘should it do it like this or like that?’ Implementation details.

One of those details is this business of ceiling. That leads into NDL.

You cannot reason how this stuff works from first principles. It is simply a choice made by someone writing code (or maybe the marketing person) for the particular computer or planner.

There is a lot of rubbish talked on these threads, mostly by people who have a somewhat deficient mental model of the systems and who are trying to figure out what is going on inside a black box by looking at a very small set of inputs and outputs.

I am not trying to be nasty saying people don’t understand this, it is quite complicated and unless the authors of the code involved publish a proper design document for what they wanted to build there is no chance of really knowing exactly what is supposed to be going on.