+ Reply to Thread
Page 1 of 3 1 2 3 LastLast
Results 1 to 10 of 24

Thread: Choosing Gradient Factors

  1. #1

    Choosing Gradient Factors

    I'm still using gradient factors for deco (Shearwater on Meg). When listening to what others have said, I've heard people using 10/90, 30/85, 10/100, 25/75, etc. ( Low GF / High GF ).

    I was wondering how you're supposed to choose. I recently found (actually was directed to) the GAP table feature which allows you to generate a bunch of RGBM profiles relatively quickly. So here's what I did.

    1. I generated a bunch of RGBM profiles for CCR. The profiles varied by depth and bottom time. All were first dives (none of this applies to repetitive or multi-day profiles). But then again, no adjustments are made for that in the GF model I dive anyway.

    2. Model the RGBM profiles and observe the GF at the 1st stop and at the surface.

    3. For any given 1st stop depth, there is a range of GFs observed. For example, for all the profiles that generated a 1st stop at 70ft, the GF upon reaching that 1st stop depth varies. Therefore, at each first stop depth I took the minimum GF observed and then graphed the results (see "RGBM GFs.pdf").

    4. I did the same thing for the surfacing GF and for various levels of helium used.

    The results were enlightening to me and will change the way I choose the GF setting on my dives. In particular, I've become a more conservative on my high GF. Hearing what others were using I thought maybe I'd up the high setting. But, at this point my dives are 150-170ft max (many less) and to date my deepest 1st stop has been at 80ft. In those lower 1st stop depths (especially when helium is used), the RGBM would not seem to recommend the surfacing GFs of 90+. Absent any other objective way of setting the GFs, I think I can stay more conservative when my dives are producing 1st stops under 100ft.



    Anyway, hope this helps someone else too.
    Attached Files
    Last edited by UWSojourner; 19th December 2005 at 19:26. Reason: Added graph to attachment

  2. #2
    RBW Member RonMicjan has a reputation beyond repute RonMicjan has a reputation beyond repute RonMicjan has a reputation beyond repute RonMicjan has a reputation beyond repute RonMicjan has a reputation beyond repute RonMicjan has a reputation beyond repute RonMicjan has a reputation beyond repute RonMicjan has a reputation beyond repute RonMicjan has a reputation beyond repute RonMicjan has a reputation beyond repute RonMicjan has a reputation beyond repute RonMicjan's Avatar
    Join Date
    Feb 2005
    Location
    Pacific NorthWest USA
    Posts
    1,238
    Meg

    CCR Dolphin

    Re: Choosing Gradient Factors

    So you have overlaid a bunch of RGBM profiles over buhlmann profiles and discovered that you need more options in your gradient factors up to and including 10/250? One could always buy an Explorer...:D

    It took me a bit to absorb the graphs (ok more than a bit, and a couple phone calls to my favorite mathematician) but it seems that one might not just choose one GF for all diving, but use different GF's for realitivly shallow dives and another one for deeper long deco stuff.

    could you further profile a range of preferred GF's based on how deep and how long the dive is and gas content? I am curious how you automated GAP to provide an information dump of that magnitude. Have some green!!

  3. #3

    Re: Choosing Gradient Factors

    Quote Originally Posted by RonMicjan
    So you have overlaid a bunch of RGBM profiles over buhlmann profiles and discovered that you need more options in your gradient factors up to and including 10/250?
    Yes, the idea was to see if RGBM might "critique" the gradient factors I was choosing. But no, I'd not recommend 10/250, or anything over x/100 since 100 is kind of the Buhlmann limit by definition.

    Quote Originally Posted by RonMicjan
    One could always buy an Explorer...:D
    I knew someone would ask that :o . One the other hand, I just wanted to get a feel for what GFs might be appropriate (for me) since I'm diving the Shearwater currently.

    Quote Originally Posted by RonMicjan
    ... it seems that one might not just choose one GF for all diving, but use different GF's for realitivly shallow dives and another one for deeper long deco stuff.
    You said it better and more succinctly than I did. But that was my conclusion. For me, 10/100 is too aggressive for the type of diving I'm doing right now, but for those guys going deeper for longer, it seems that it might be ok. At least I think I'd be comfortable using it on those types of dives.

    Quote Originally Posted by RonMicjan
    could you further profile a range of preferred GF's based on how deep and how long the dive is and gas content?
    Sure, I'll post later what I'd consider suggestions based on some of the data I saw. But no guarantees ... so use at your own risk.

    Quote Originally Posted by RonMicjan
    I am curious how you automated GAP to provide an information dump of that magnitude.
    Well, GAP is pretty amazing at being able to generate a bunch of tables relatively quickly (thanks Joe Radomski for pointing out that feature). Once you get the profiles, you just translate those into something your own GF program can use and feed them through overnight and grab the data you're interested in - low GF, surfacing GF, tissue pressures, etc., generated by the RGBM profile.

    If, for example, the bubble model starts a profile at GF=0.17 and ends at GF=0.78 I figure using 10/100 might be too aggressive for that type of diving . At the very least, it seems to be saying that you might be developing bubbles that aren't worked off by a longer shallow stop. Maybe its not an exact analysis, but I think its a better hint at how one might go about choosing your GF settings than a coin flip :) .

    I'll post that table later.

  4. #4
    Evolving 01RMB is just really nice 01RMB is just really nice 01RMB is just really nice 01RMB is just really nice 01RMB is just really nice 01RMB is just really nice 01RMB is just really nice 01RMB is just really nice 01RMB is just really nice 01RMB is just really nice 01RMB's Avatar
    Join Date
    May 2005
    Location
    Brisbane, Australia
    Posts
    198
    Evolution

    Re: Choosing Gradient Factors

    This looks great. I like the analysis you have done.

    I have tended to use 15/85 when diving trimix and 30/90 when on air but I think the comparison against RGBM helps select a good range of values for a particular dive. I have found using a computer with GF can be limiting sometimes and when comparing it against RGBM using GAP I have wondered whether a different set of GF values would be better suited.

    Your analysis seems to indicate some way of coming up with a guide. A much better idea than throwing a hand full of bones on the ground and trying to figure out what the GF values should be!

  5. #5
    RBW Member RonMicjan has a reputation beyond repute RonMicjan has a reputation beyond repute RonMicjan has a reputation beyond repute RonMicjan has a reputation beyond repute RonMicjan has a reputation beyond repute RonMicjan has a reputation beyond repute RonMicjan has a reputation beyond repute RonMicjan has a reputation beyond repute RonMicjan has a reputation beyond repute RonMicjan has a reputation beyond repute RonMicjan has a reputation beyond repute RonMicjan's Avatar
    Join Date
    Feb 2005
    Location
    Pacific NorthWest USA
    Posts
    1,238
    Meg

    CCR Dolphin

    Re: Choosing Gradient Factors

    If RGBM actually works, and according to Mr's Winke and Radomski, it does, why not surface with a GF of greater than 100? Last year at Zero Gravity I showed up diving an explorer and during the trip I won a VR3 in a contest, so I was diving both side by side. I had always dived the Explorer on RGBM, for those who dont know, the Explorer has 3 settings for RGBM conservatism (10 algorithms in all), cold water/work, medium and warm water/no work, I did many CCR dives to extreme depths on the proper setting and never saw any symptoms of DCI. Diving the units side by side I always followed the most conservative one for deep stops etc. The explorer always cleared me out of the water sooner than the VR3, on long deep dives sometimes as much as 20-30 minutes sooner. I think this says a lot about the RGBM algorithm, especially since I know Joe has taken it to over 500 feet with only small amounts of brain damage (but Joe has been a little different ever since I have known him).:D

    So we have at least two different pathways to the surface, bubble like heck on the way up and then hang till they are gone, or prevent the bubbles from becomming more than a flyspeck, and skipping all that extra hang time. I wonder why more computer manufacturers are not using RGBM? I know its an extra cost, but if it works?

  6. #6

    Re: Choosing Gradient Factors

    Quote Originally Posted by RonMicjan
    If RGBM actually works, and according to Mr's Winke and Radomski, it does, why not surface with a GF of greater than 100?
    If you are using RGBM for deep long dives then you ARE surfacing with a Buhlmann model GF>100, sometimes substantially greater. Those charts I posted show that.

    However, I wouldn't feel comfortable setting my Buhlmann GF algorithm above 100 since the RGBM relies on a particular ascent pattern to reduce bubbling and allow shorter shallow stops. If I don't KNOW I'm producing a similar ascent profile, I wouldn't want to taunt Buhlmann, role the dice, and surface. When I want the advantage of the shorter shallow stops, I'll get a computer that implements RGBM.

    This little thought experiment was really trying to determine when I might be comfortable setting the High GF up to 100, or 90, etc., not when I might be comfortable going over 100, which for me is never without a bubble model assisting me.

    Quote Originally Posted by RonMicjan
    I wonder why more computer manufacturers are not using RGBM? I know its an extra cost, but if it works?
    Extra cost, extra complexity. Some may not fully trust the sometimes radically shorter profiles. Other than that I don't really know.
    Last edited by UWSojourner; 23rd December 2005 at 19:18.

  7. #7
    Moderator jradomski has a reputation beyond repute jradomski has a reputation beyond repute jradomski has a reputation beyond repute jradomski has a reputation beyond repute jradomski has a reputation beyond repute jradomski has a reputation beyond repute jradomski has a reputation beyond repute jradomski has a reputation beyond repute jradomski has a reputation beyond repute jradomski has a reputation beyond repute jradomski has a reputation beyond repute jradomski's Avatar
    Join Date
    Mar 2005
    Location
    "Da" Bronx
    Posts
    4,116
    Hammehead & rEvo & Custom

    KISS,Optima,Inspiration,Meg

    Re: Choosing Gradient Factors

    Quote Originally Posted by UWSojourner
    If you are using RGBM for deep long dives then you ARE surfacing with a Buhlmann model GF>100, sometimes substantially greater. Those charts I posted show that.

    However, I wouldn't feel comfortable setting my Buhlmann GF algorithm above 100 since the RGBM relies on a particular ascent pattern to reduce bubbling and allow shorter shallow stops. If I don't KNOW I'm producing a similar ascent profile, I wouldn't want to taunt Buhlmann, role the dice, and surface. When I want the advantage of the shorter shallow stops, I'll get a computer that implements RGBM.

    This little thought experiment was really trying to determine when I might be comfortable setting the High GF up to 100, or 90, etc., not when I might be comfortable going over 100, which for me is never without a bubble model assisting me.


    Extra cost, extra complexity. Some may not fully trust the sometimes radically shorter profiles. Other than that I don't really know.
    What did you use to generate your GF numbers they don't look quite right?
    (I didn't examine that closely so I might be missing something, just by observation)



    100% free webcam site! | Awesome chicks and it is absolutely free! | Watch free live sex cam - easy as 1-2-3
    Joe Radomski
    CCR Trimix Instructor Trainer
    ANDI Instructor Trainer Director #10

    All posts are personal opinions and DO NOT reflect any affiliated agency unless specifically stated.

  8. #8

    Re: Choosing Gradient Factors

    Quote Originally Posted by jradomski
    What did you use to generate your GF numbers they don't look quite right?
    (I didn't examine that closely so I might be missing something, just by observation)
    Here's the process:

    1. Generate GAP RGBM profile.
    2. Plug the RGBM stops into my GF program.
    3. Record the 1st stops and LoGF and surfacing GF.

    All the numbers in the graphs are not single profile gradient factors. Rather, for a range of dives producing a particular first stop depth, they represent the minimum of the Low GFs produced or the minimum of the High GFs produced. The values will be dependent on the profiles used of course and whether or not my GF program is correct. However, my program reproduces the GAP GF profiles pretty closely except in certain limited circumstances (that I discussed with Kees but finally had to agree to disagree about the issue).

    Anyway, let me know if something seems fundamentally wrong so I can correct it if necessary.

  9. #9

    Re: Choosing Gradient Factors

    Quote Originally Posted by jradomski
    What did you use to generate your GF numbers they don't look quite right?
    Ok. Talked with jradomski. He seems to be ok with the methodology at this point (right Joe?).

    The issue is that he believes RGBM is too conservative in the "light deco" area. That is, when only let's say 10 minutes or less of deco is required, then Joe thinks RGBM is too conservative.

    I can't comment on whether its too conservative. But what I'll do is show the same kind of analysis based on less than 10 minutes of deco and 10 minutes+ of deco. Might be a few days.

  10. #10
    Moderator jradomski has a reputation beyond repute jradomski has a reputation beyond repute jradomski has a reputation beyond repute jradomski has a reputation beyond repute jradomski has a reputation beyond repute jradomski has a reputation beyond repute jradomski has a reputation beyond repute jradomski has a reputation beyond repute jradomski has a reputation beyond repute jradomski has a reputation beyond repute jradomski has a reputation beyond repute jradomski's Avatar
    Join Date
    Mar 2005
    Location
    "Da" Bronx
    Posts
    4,116
    Hammehead & rEvo & Custom

    KISS,Optima,Inspiration,Meg

    Re: Choosing Gradient Factors

    Quote Originally Posted by UWSojourner
    Ok. Talked with jradomski. He seems to be ok with the methodology at this point (right Joe?).

    The issue is that he believes RGBM is too conservative in the "light deco" area. That is, when only let's say 10 minutes or less of deco is required, then Joe thinks RGBM is too conservative.

    I can't comment on whether its too conservative. But what I'll do is show the same kind of analysis based on less than 10 minutes of deco and 10 minutes+ of deco. Might be a few days.
    Yes I agree with the methodology..

    To clarify for others the comment about RGBM being a bit too conservative for "light" deco, is that The RGBM model tends to enter deco much earlier than traditional models and will skew the results slightly.. This is especially true in a Nitrogen only dive near the traditional no deco limits.. The "10 minute" suggestion should bring the realtive safety of the two models about the same time.. Also it should be noted the comparisons done were based on a conservative RGBM setting as well..



    100% free webcam site! | Awesome chicks and it is absolutely free! | Watch free live sex cam - easy as 1-2-3
    Joe Radomski
    CCR Trimix Instructor Trainer
    ANDI Instructor Trainer Director #10

    All posts are personal opinions and DO NOT reflect any affiliated agency unless specifically stated.

+ Reply to Thread
Page 1 of 3 1 2 3 LastLast

Posting Permissions

  • You may not post new threads
  • You may not post replies
  • You may not post attachments
  • You may not edit your posts