OSTC+ Tech simulations

OSTC's running hwOS sport or tech
Post Reply
Yves
Posts: 20
Joined: Monday 24. August 2020, 13:07

OSTC+ Tech simulations

Post by Yves »

Hi,
I have an OSTC+ Tech with SW Release 3.11. I wanted to get familiar with my new dive computer through simulations.
Then, I configured the deco model to ZH-L16 with different sat/desat factors 100/100 (default), 95/105, 90/110, 85/115, and for each of them I started the simulator at a series of depths : 12m, 15m, 20m, 25m, 30m, 39m, 45m, 51m to measure the duration before deco stops are displayed. Then, I have a security curve for each sat/desat setting.
For each depth, the above sat/desat factors give the same results, which in my view looks unappropriate. Every time a simulation is over, the computer automatically switches to the start screen where the current sat/desat factors in use are displayed. I discovered that, whatever the sat/desat settings that I selected in menu/dive settings/deco settings, the displayed S/D remains the same at S/D 110/90, which corresponds to the former default setting in the OSTC+ Sport ; that would explain why the simulation results are always the same. Then, I suspect that only the displayed S/D factors (110/90) is used when selecting the ZH-L16 deco model, and that the sat/desat settings are ineffective, which is not normal.
I did the same exercise with the GF deco model, for the same series of depths, with the GF factors 50/95, 30/90,30/85, 30/80, 20/75, 30/70, and the simulation results are coherent ; each time, the displayed GF setting was correct.
Would you be kind enough to check what is wrong when using the ZH-L16 deco model in the exercise above explained, and fix it. Indeed, I know that using the GF model is better recommended, nonetheless the OSTC computer should work properly even with ZH-L16.
Thanks for your help and support
User avatar
gr3yw0lf
Posts: 106
Joined: Thursday 28. November 2019, 22:18

Re: OSTC+ Tech simulations

Post by gr3yw0lf »

hi Yves
please take a look at this posting and update your ostc+
life is better in fins

OSTC Plus #18168
Yves
Posts: 20
Joined: Monday 24. August 2020, 13:07

Re: OSTC+ Tech simulations

Post by Yves »

Hi,
thanks a lot for your reply.
Actually, I discovered afterwards that there was a new FW Release that solved the problem.
I downloaded that new Release, and ran again my simulations. Now, it looks OK.
Sorry for the inconvenience.
Now this discussion is over.
Ralph
Posts: 708
Joined: Saturday 24. June 2017, 11:31

Re: OSTC+ Tech simulations

Post by Ralph »

May you share your results? As one of the OSTC programmers, i'd be curious to see if results are what i supposed them to be by independend testing... :-)
Yves
Posts: 20
Joined: Monday 24. August 2020, 13:07

Re: OSTC+ Tech simulations

Post by Yves »

Hi Ralph,
Indeed, no problems to share my simulations.
Actually, I wanted to gauge the dive duration at the definite depth before a deco stop is displayed. My reference was the MN90 deco table of the french navy.
The selected depths at stake were : 9m, 12m, 15m, 20m, 25m, 30m, 40m, 45m, 50m.
Since not all above depths were selectable on the OSTC, I fell back to: 12m, 15m, 20m, 24m, 30m, 39m, 45m, 51m. 9m relates to the max NDL.
2 deco models were used : ZH-L16 and ZH-L16+GF:
ZH-L16 (no GF) with the following sat/desat factors: 100/100, 95/105, 90/110, 85/115
ZH-L16+GF with the following GF paarameters: 50/95, 30/90, 30/85, 30/80, 20/75, 20/70, 30/70 (default by Shearwater)
OSTC default parameters are in bold.
The results come from the OSTC+ Tech simulation. I suppose the simulator is the same for all OSTC models.
A result in green depicts more conservative against the MN90 table. That means that OSTC shortens the duration before deco is displayed.
A result in red depicts less conservative/more progressive against the MN90 table. That means in OSTC, deco display starts later than the MN90 table.
Then, see attachment for the results.
Deco_security_curve_MN90vsOSTC.pdf
(415.05 KiB) Downloaded 234 times
PS: I do not know how to insert a drawing/image within the text here. Any clue?
Thanks Ralph for your feedback.
Yves
Posts: 20
Joined: Monday 24. August 2020, 13:07

Re: OSTC+ Tech simulations

Post by Yves »

Ralph? No feedback? :(
Ralph
Posts: 708
Joined: Saturday 24. June 2017, 11:31

Re: OSTC+ Tech simulations

Post by Ralph »

Ähm, sorry, too much other stuff can into way...

Did the results fit with your expectation? From my side, and some background information:

Saturation and Desaturation factors are written in sequence sat & desat, with sat >= 100%
and dest <= 100%. Technically, they speed up gas uptake and slow down gas removal. So e.g.
with sat = 110% and desat 100%, the NDL time should be reached in 100/110 = 91% of the
normal time. While in NDL, dsat has little to none effect - it is used while calculation of the
ascent, which is under no-stop condition (depth / 10 m) in minutes, so relatively short to make
a major Impact.

Regarding GF factors, while within NDL, GF high doesn't come into play. GF low sets the
supersaturation when reching the surface, so when again neglecting the ascent time, GF low
scales the NDL time (lower GF low, shorter NDL time) with some logarithmic touch the deeper
you are.

Is that what you also expected and/or got?

best regards,
Ralph
Yves
Posts: 20
Joined: Monday 24. August 2020, 13:07

Re: OSTC+ Tech simulations

Post by Yves »

Thanks Ralph,
I agree.
However, I received the new OSTC-GF-calculations document from Matthias, and, instead of determining the time without decostop as soon as the NDL=0 turns to decostop=3m, 1', I checked the Ceiling, and I used the simulator with the following parameters:
GF=30/85, depth=30m. I read the following results at 30m depth:
sat=0%, ceiling=0.0 metre (I assume), NDL = 0' meaning that I am already at 30m depth for several minutes, then:
sat=0%, ceiling=0.1, NDL=0': since Ceiling is under the surface, a decostop should apply shouldn't it?
sat=0%, ceiling=0.2, decostop=3m 1'
sat=0%, ceiling=1.1, decostop=3m 1'
sat=0%, ceiling=1.2, decostop=6m 1'
sat=0%, ceiling=1.5, decostop=3m 2' a little surprising: any idea?
sat=0%, ceiling=1.6, decostop=6m 1'
sat=0%, ceiling=2.6, decostop=6m 2'
sat=0%, ceiling=2.9, decostop=9m 1'
sat=0%, ceiling=3.6, decostop=9m 2'
sat=0%, ceiling=3.7, decostop=12m 1'
Then, I decided to ascend to the first decostop 12m, then 9m, then 6m, then 3m with these displays:
12m: sat=26%, ceiling=3.5, decostop=9m 2'
9m: sat=38%, ceiling=3.1, decostop=9m 1'
6m: sat=53%, ceiling=2.5, decostop=6m 3'
sat=46%, ceiling=1.8, decostop=6m 1'
3m: sat=45%, ceiling=1.7, decostop=3m 6'
sat=24%, ceiling=0.3, decostop=3m 1'
sat=20%, ceiling=0.1, NDL=0' means still wait before reaching the surface because of the ceiling >0.
sat=84%, ceiling=0.0, NDL=240' meaning probably that the saturation will be 84% in excess (=184%) at the surface.
I have 2 comments:
1) It looks like "Saturation" actually means "Oversaturation" during ascent. Please advise.
2) Since I understood that "Ceiling" means "the depth at which the first tissue will reach 100% saturation" and "as long as the Ceiling is displayed as 0, you can ascend directly to the water surface, you are within the no deco limit", I thought there was a close relation between Ceiling and Decostop. Please, explain.

Later, I was not sure about the relation between Ceiling and Decostop: a decostop is a ceiling isn't it:
With ceiling=0.1 I expected the first decostop to be displayed at 3m 1' to keep with traditional decostop depth.
If the ceiling=3.2, why the decostop is at 9m 2' and not at 6m?
Could you please clarify?
Thanks and regards
Ralph
Posts: 708
Joined: Saturday 24. June 2017, 11:31

Re: OSTC+ Tech simulations

Post by Ralph »

Hi,

you are good on track and actually already deep into the details of the deco calculation algorithm.

NDL x' is the "count-down" of remaining NDL time. With x = 0 you are riding on the edge of the NDL (in German language we call that "Nullzeitschrammeln").

It is correct that the OSTC still shows NDL for a short time even if the ceiling already started to be none-zero. The reason is that the OSTC calculates the ascent, with 10 m per minute, updating the tissue pressures all that time while ambient pressure and in consequence tissue pressures drop. And as long as the ceiling is 0 again at the end of the calculated ascent, the OSTC rates the dive as being stil in NDL. There is a similar effect at the end of the last deco stop, where it can happen that a remaining stop time of " ..' " (two dots) is shown: in this case technically there is still a stop obligation (ceiling > 0), but if ascending now with max 10m/min it will have gone way when the surface is reached. Soon after, the dive is shown as being in NDL again, even if there is a tiny ceiling still remaining.

Jump from 6m 1' to 3m 2': that's an artefact from how the GFs are defined. The GF low sticks with the first (deepest) stop. When the ceiling becomes 3.1 meter, the 6 m stop is introduced. At the same time, the GF low reference depth moves from 3 m (previous 1st stop) to 6m (current 1st stop). With the steep GF 30/85 ratio, the original 3 m stop resulted from the M-values adjusted down to 30% of their original value. Now with GF low ref depth at 6m, the adjusted M-values for 3 m are at 30% + (85% - 30%) / 2 = 57,5% of the original Bühlmann value, i.e. more supersaturation is allowed.
As the OSTC calculates stops with a resolution of full minutes, 1 minute at 6 m will clear the tissues by so far that the 3 m stop is not needed any more. Then the bespoke GF low reference depth shift comes into action, and in the next calulation round the 1st stop initially is placed at the depth of 6 meters (with GF low = 30% of M-values as limit), but then the OSTC discovers that going up to 3 meters will not bring the supersaturations above the 57,5% that are allowed there as of now, and thus discards the 6m stop in favour of going straight to 3m. This continues until the straight ascent to 3 meters starts to exeed the 57,5%, at which time the 6 m stop will re-appear. This pattern often repeats with deeper stops, too, especially when GF low and GF high are very different.

The ceiling is always calculated against the original M-values, that is why you have e.g. a stop at 6m (calculated with 30%) while the ceiling (calculated with 100%) is only 2.6 m, hence < 3m.

The sat% indeed is supersaturation, but "supersaturation" doen't fit the available screen space (as well, for a pitty, most divers don't know the difference anyhow...). So supersaturation percentage can only rise above 0% once the bottom depth has been left and ambient pressure falls below tissue pressure.

When you're sick of doing deco stops at 3m intervals, you can also ascent by not following the indicated stop depth, but by controlling (i.e. holding) a specific supersat percentage value. When doing so, you will apply the GFs by hand, stopping ascent for 1st stop when reaching your desired GF low %, and then progressively let rise the % value by a bit each stop until finally surfacing with supersat % at your desired GF high. The OSTC does exactly the same. In case of emergency, the ceiling always informs you where you could go to, bringing up your supersat to 100%. I personally love this custom view, as it gives all information needed to manage deco ascent, especially in abnormal conditions like overpopulated stop depths or emergency situations like low on air etc.

An oversaturation of 84% means your leading tissue is at a point of 84% way between tissue pressure = ambient pressure and the tissue's M-value. You can't really add 100% to convert it into "total"pressure" - it is simply not defined that way that this would work.

1) and 2) answered now?

A decostop is a kind of ceiling, just calculated with GFs and rounded to the next multiple of 3 meters. You can proof that by setting GF low = GF high to get rid of the sliding GF changes and then doing a deco ascent in the simulator. You will see you will be cleared for the next stop each time the ascent will bring your supersat up again to your set limit. With both GF set to 100%, this will coincident with the ceiling having counted down to the next stop depth. But do not forget: the OSTC figures in the ascent time, so it will let you start always a tiny bit of time ahead of the numbers becoming exactly equal.

All clear now?

:-) Ralph
Yves
Posts: 20
Joined: Monday 24. August 2020, 13:07

Re: OSTC+ Tech simulations

Post by Yves »

Hi Ralph,
Thanks a lot for your detailed explanations. I am getting closer to the heart of the OSTC :)
Yes, it is much clearer.
If you permit me, I still have some comments for your appreciation:

You: It is correct that the OSTC still shows NDL for a short time even if the ceiling already started to be none-zero. The reason is that the OSTC calculates the ascent, with 10 m per minute, updating the tissue pressures all that time while ambient pressure and in consequence tissue pressures drop. And as long as the ceiling is 0 again at the end of the calculated ascent, the OSTC rates the dive as being stil in NDL.
Me: Agree. This is the way the desaturation shall work: the ascent to the next upper deco stop can be engaged only if the computer has anticipated that the tissue saturation stays within the defined limit (sat/GF) at the end of the step ascent. Between deco stops, the speed of the ascent is highly recommended to be very slow, around 6m/min even less, so 10m/min is questionable.

You: There is a similar effect at the end of the last deco stop, where it can happen that a remaining stop time of " ..' " (two dots) is shown: in this case technically there is still a stop obligation (ceiling > 0), but if ascending now with max 10m/min it will have gone way when the surface is reached. Soon after, the dive is shown as being in NDL again, even if there is a tiny ceiling still remaining.
Me: Looks it is a matter of latency.
10m/min close to the surface still looks fast.


You: Jump from 6m 1' to 3m 2': that's an artefact from how the GFs are defined. The GF low sticks with the first (deepest) stop. When the ceiling becomes 3.1 meter, the 6 m stop is introduced.
Me: You mean 3.1 Ceiling parameter (!). In my simulation, when the Ceiling was 3.1, the deco stop indicated 9m 1’ (same as ceiling=2.9) because of the GF low of 30%. With this GF low, the 6m deco stop was displayed as the Ceiling was 1.6 only, which is far from 3.1.

You: At the same time, the GF low reference depth moves from 3 m (previous 1st stop) to 6m (current 1st stop). With the steep GF 30/85 ratio, the original 3 m stop resulted from the M-values adjusted down to 30% of their original value. Now with GF low ref depth at 6m, the adjusted M-values for 3 m are at 30% + (85% - 30%) / 2 = 57,5% of the original Bühlmann value, i.e. more supersaturation is allowed.
... then the OSTC discovers that going up to 3 meters will not bring the supersaturations above the 57,5% that are allowed there as of now, and thus discards the 6m stop in favour of going straight to 3m. This continues until the straight ascent to 3 meters starts to exeed the 57,5%, at which time the 6 m stop will re-appear. This pattern often repeats with deeper stops, too, especially when GF low and GF high are very different.

Me: This reasoning sounds good to me.
It seems that the “GF gradient” (GF high-GF low) plays up for a short while. This is probably due to the “sliding GF changes” isn’t it.
Question: why would the (anticipated) ascent to 3m exceed the 57.5% that WAS allowed at 3m?
My answer: because of the OSTC recalculation after some time at depth, and readjustment.
Correct?


You: The ceiling is always calculated against the original M-values, that is why you have e.g. a stop at 6m (calculated with 30%) while the ceiling (calculated with 100%) is only 2.6 m, hence < 3m.
Me: Understood. Perfect. I appreciate the OSTC displays the Bühlmann ceiling separate from the deco stop ceilings.

You: The sat% indeed is supersaturation, but "supersaturation" doen't fit the available screen space (as well, for a pitty, most divers don't know the difference anyhow...). So supersaturation percentage can only rise above 0% once the bottom depth has been left and ambient pressure falls below tissue pressure.
Me: I agree. This is what I guessed. I just wanted to read that from you too. We can read the Saturation parameter as “Saturation in excess of the ambient pressure”.

You: ... The OSTC does exactly the same. In case of emergency, the ceiling always informs you where you could go to, bringing up your supersat to 100%. I personally love this custom view, as it gives all information needed to manage deco ascent, especially in abnormal conditions like overpopulated stop depths or emergency situations like low on air etc.
Me: I totally agreed with you, I love that too. In the 30/85 case, it shall be possible to ascent as long as 30% oversaturation is not reached, and stop for some time to allow the oversat to go lower than 30% enough to resume the ascent till 30% oversat is reached again. When getting to the surface, the 85% oversat comes into play instead. Brilliant!
I am wondering if we could not do something similar with the Ceiling parameter. The difference I see is that the ceiling parameter works on 100% Bühlmann and does not take advantage of the GF tuning like the Oversaturation parameter. Is that right?


You: An oversaturation of 84% means your leading tissue is at a point of 84% way between tissue pressure = ambient pressure and the tissue's M-value. You can't really add 100% to convert it into "total"pressure" - it is simply not defined that way that this would work.
Me: Actually, since the body is always at 100% saturation at the surface with no dive, I just wanted to express that with an oversaturation of 84% (<85 GF high by the way), there is an addition of 84% to the “normal” 100%, so 100%+84%=184%, just to say that my leading tissue is oversaturated by 84%. This is nothing to do with a computer total saturation calculation.

You: A decostop is a kind of ceiling, just calculated with GFs and rounded to the next multiple of 3 meters. You can proof that by setting GF low = GF high to get rid of the sliding GF changes and then doing a deco ascent in the simulator. You will see you will be cleared for the next stop each time the ascent will bring your supersat up again to your set limit. With both GF set to 100%, this will coincident with the ceiling having counted down to the next stop depth. But do not forget: the OSTC figures in the ascent time, so it will let you start always a tiny bit of time ahead of the numbers becoming exactly equal.
Me: Problem: GF low = GF high = 100% is NOT POSSIBLE.
On the settings, GF low max value is 85% (from 10% to 85%), which is in contradiction to your user manual (10% to 100%). Why?
Anyway, I set both GFs at 85% to get rid of the sliding GF changes, and started the simulation:
There was no artefact.
Ceiling follows the deco depth rounded up to modulo 3m. Perfect.


I do appreciate your help and support ;)
Thanks and regards,
Yves
Ralph
Posts: 708
Joined: Saturday 24. June 2017, 11:31

Re: OSTC+ Tech simulations

Post by Ralph »

Hi Yves,

that's probably one of the longest posts we ever had! I'll try to catch the remaining open items only:

Ascents are calculated with 10m/min, because that's the design of the Bühlmann model. Sure the slower the better, so when you do 6m/min or even just 1m/min up from the last stop you'll be on the safe side. When being slower, in theory you could start the ascent a little bit earlier as you are loosing a bit more tissue pressure while under way. But in practise, that would be a couple of seconds only.

Question: why would the (anticipated) ascent to 3m exceed the 57.5% that WAS allowed at 3m?
My answer: because of the OSTC recalculation after some time at depth, and readjustment.
Correct?

Because time has gone by since, and the tissues have taken up more pressure since, demanding deeper stopping now.

Yep, the ceiling depth is always calculated with "100%" or pure Bühlmann as per design of the OSTC. So to say, it always displays "where the border line is". If you like to have a secondary stop depths & (super)sat display, you can activate the the alternative GF factors and set them to e.g. 100/100. During the dive, you can toggle back and forth between the GF and the aGF at any time, all calculations will update accordingly.

OK, i would just avoid speaking out loud that i'm on 184% saturation for fear someone calls DAN before i can explain what i mean... ;-)

Since not too long there's a change in the firmware, now preventing GF low set higher than GF high for two reasons: 1) it does not make sense, and 2) the OSTC would compute very odd things in that case. So you need to tune GF high up to 100% before you can let GF low follow.

cheers,
Ralph
Yves
Posts: 20
Joined: Monday 24. August 2020, 13:07

Re: OSTC+ Tech simulations

Post by Yves »

Hi Ralph,

It's all fine to me.
Thanks for your time and patience with me.

With best regards,
Yves
Post Reply