Mailing List Archive Mailing List Archive
Hi, Peter and Tracy! Hopefully the article is appended to this e-mail. As i said, it's an Ami-Pro file. Your comments and criticism are most welcome, either public or private. I can send anything else i write if you are interested?? -- Julian Bottomer[ver] 4 [sty] [files] [charset] 82 ANSI (Windows, IBM CP 1252) [revisions] 0 [prn] HP DeskJet 500 Printer [port] LPT1: [lang] 2 [desc] 815829645 1 815823865 91 7 34 196 0 0 1 [fopts] 0 1 0 0 [lnopts] 2 Body Text 1 [docopts] 5 2 [GramStyle] [tag] Body Text 2 [fnt] Courier New 240 0 32768 [algn] 1 1 0 0 0 [spc] 36 504 1 0 0 1 100 [brk] 4 [line] 8 0 1 0 1 1 1 10 10 1 [spec] 0 0 0 1 1 0 16 0 0 [nfmt] 272 1 2 . , </c> Body Text 0 0 [tag] Body Single 3 [fnt] Courier New 240 0 32768 [algn] 1 1 0 0 0 [spc] 33 273 1 0 0 1 100 [brk] 4 [line] 8 0 1 0 1 1 1 10 10 1 [spec] 0 0 0 1 1 0 0 0 0 [nfmt] 272 1 2 . , </c> Body Single 0 0 [tag] Header 4 [fnt] Courier New 240 0 49152 [algn] 1 1 0 0 0 [spc] 33 273 1 0 0 1 100 [brk] 4 [line] 8 0 1 0 1 1 1 10 10 1 [spec] 0 0 0 1 1 0 0 0 0 [nfmt] 272 1 2 . , </c> Header 0 0 [tag] Footer 5 [fnt] Courier New 240 0 49152 [algn] 1 1 0 0 0 [spc] 33 273 1 0 0 1 100 [brk] 4 [line] 8 0 1 0 1 1 1 10 10 1 [spec] 0 0 0 1 1 0 0 0 0 [nfmt] 272 1 2 . , </c> Footer 0 0 [tag] Table Text 6 [fnt] Courier New 240 0 32768 [algn] 1 1 0 0 0 [spc] 36 504 1 0 0 1 100 [brk] 4 [line] 8 0 1 0 1 1 1 10 10 1 [spec] 0 0 0 1 1 0 0 0 0 [nfmt] 272 1 2 . , </c> Table Text 0 0 [frm] 1 1573060 1800 1800 10021 3113 0 1 1 1 0 0 0 0 0 0 0 16777215 3 0 0 0 0 [frmlay] 2816 8221 1 0 0 1 1800 0 0 2 0 1 1 0 1 1944 10296 0 [tbl] 4 2 508 0 4003 86 1 43 43 [h] 0 254 86 2 0 2 0 1 508 86 2 0 1 0 2 254 86 2 0 2 0 3 254 86 2 0 2 0 [e] [data] 0 0 0 0 0 0 0 1 0 0 0 0 @Body Single@Ju* Bottomer<:f> > 0 1 16 0 0 0 0 1 0 0 0 0 @Body Single@<+A><:f,,><:X3,0;NumWords>34<:X~3,0;NumWords> Words<:f> > 1 0 0 0 0 0 0 1 0 0 0 0 @Body Single@31* Highfield Road<:f> @Body Single@Bo* BH9 2SE,<:f> > 1 1 16 0 0 0 0 1 0 0 0 0 @Body Single@<+A> > 2 0 0 0 0 0 0 1 0 0 0 0 @Body Single@+44 (0)1202 580065 (w)<:f> > 2 1 16 0 0 0 0 1 0 0 0 0 @Body Single@<+A></i> Julian Bottomer<:f> > 3 0 0 0 0 0 0 1 0 0 0 0 @Body Single@ > 3 1 0 0 0 0 0 1 0 0 0 0 @Body Single@ > [e] [tble] [lay] Standard 516 [rght] 16833 11908 1 1800 1800 1 1800 1800 0 1 0 1 0 2 1 1800 10105 12 1 720 1 1440 1 2160 1 2880 1 3600 1 4320 1 5040 1 5760 1 6480 1 7200 1 7920 1 8640 [hrght] [lyfrm] 1 11200 0 0 11908 1800 0 1 3 1 0 0 0 0 0 0 0 0 1 [frmlay] 1800 11908 1 1800 72 1 792 1800 0 1 0 1 1 0 1 1800 10105 2 2 4680 3 9360 [txt] > [frght] [lyfrm] 1 13248 0 15033 11908 16833 0 1 3 1 0 0 0 0 0 0 0 0 2 [frmlay] 16833 11908 1 1800 792 1 15105 1800 0 1 0 1 1 0 1 1800 10105 2 2 4680 3 9360 [txt] > [elay] [l1] 0 [pg] 7 11 328 55 96 0 0 0 65533 65535 Standard 65535 0 0 0 0 0 0 0 0 0 65535 0 0 65535 1 0 0 0 0 14 216 56 0 0 0 0 65533 65535 Standard 65535 0 0 0 0 0 0 0 0 0 65535 0 0 65535 1 0 0 0 0 16 453 56 96 0 0 0 65533 65535 Standard 65535 0 0 0 0 0 0 0 0 0 65535 0 0 65535 1 0 0 0 0 20 492 54 0 0 0 0 65533 65535 Standard 65535 0 0 0 0 0 0 0 0 0 65535 0 0 65535 1 0 0 0 0 22 329 49 96 0 0 0 65533 65535 Standard 65535 0 0 0 0 0 0 0 0 0 65535 0 0 65535 1 0 0 0 0 24 439 10 96 0 0 0 65533 65535 Standard 65535 0 0 0 0 0 0 0 0 0 65535 0 0 65535 1 0 0 0 0 29 0 8 1025 0 0 0 65533 65535 Standard 65535 0 0 0 0 0 0 0 0 0 65535 0 0 65535 1 0 0 0 0 [edoc] <:t0> <+B><:H<*-> Julian Bottomer<:f> - <:P12,2,> > @Body Single@<+B><:#254,8305> @Body Single@<+B><:#254,8305> <+B><:#508,8305> <+B><:#508,8305> <+B><++>Use and misuse of oxygen analysers<-+><:f> <+B><:#508,8305>by <+B><:#508,8305>Julian Bottomer<:f> During recent years the oxygen analyser has become a widely used piece of kit as far as the nitrox or technical diver is concerned. You would think that with the benefit of several years of availability of analysers that it would be straight-forward to get accurate results when trying to measure the extremely important oxygen content, but my experience shows that this is just not the case. Let me explain further:I work partly in a shop where a large part of the business is filling nitrox and trimix cylinders, and accurate analysis is obviously of great concern to me and to the customers that we supply, both for there safety and from the liabi lity point of view. I am fortunate in that i have access to a Servomex Paramagnetic Oxygen analyser when mixing, which is THE laboratory standard as far as oxygen analysis is concerned (the main reason that they are not used for recreational gas mixing is n othing other than the expense of the instrument-anything up to 5000 pounds!). The performance and stability is to a far higher standard than the commonly used hand-held galvanic type analysers costing in the region of 200-400 pounds and has enabled me to in vestigate the accuracy of mixes analysed after production with these types of hand-held analysers. <:#3556,8305>The main course for concern as i see it is that many people are not aware that ALL analysers are not percentage sensors but are PARTIAL PRESSURE sensors. This means in effect that if we calibrated our analyser to read 21.0% in a normal air environment and w e then took that same instrument to a depth of 10m in a recompression chamber it would read 42% ! <:#2540,8305>The significance of this as far as we are concerned is that when we connect our cylinder of gas to the analyser by whatever method is necessary, any pressure on the sensing face of the cell will give a falsely high reading. When nitrox was in its infancy in the UK i was shown how to use an analyser by holding the t-piece containing the cell in front of the pillar valve and turning the cylinder on to get a slight flow. It was only later experiments that later showed the apparen t oxygen to be dependant on the flow rate and hence the pressure applied to the cell! The next inovative method i came across was to push the cell into the mouthpiece of a regulator and depress the purge button to give a flow of gas to the cell. This also c auses problems for exactly the same reasons-the more you push the purge button, the higher the flow and pressure,the higher the apparent oxygen content. So is there a way to overcome these problems? One very simple way is to place the cell(or the complete instrument) into a plastic bag and flush the bag several times with the gas until the reading no-longer rises and is stable. It is obviously important not to pressurise the bag to avoid a false reading, and this can be achieved by making a number of SMALL pinholes to allow excess to be relieved. A very neat and more professional solution is available from John Lamb who manufactures and distributes the very neat and robust Vandagraph VN202 analyser. This device is in effect a male din-type connector fitted to a plastic t-piece with facility to conne ct the cell into it, and fitted with a long tail to prevent air entering. It has a restrictor to prevent dangerous over-pressurisation of the cell and possible damage. It works in the same way as the plastic bag method by just providing a small volume of ga s with zero excess pressure. You can see for yourself the effects of pressure by using the cylinder valve to give different flow rates across the sensor. It is more dramatic on some analysers and there associated sampling devices than others. <:#1016,8305>Having now looked at the use of the sampling device, let us now look further into the depths of accurate analysis. <:#2540,8305>If you read the instructions that came with the analyser (you did, didn't you?) you will generally find a list of do's and don'ts. Some are obvious, such as not getting the instrument wet or dropping it, and some are not so obvious such as the correct gas t o use to calibrate it. <:#2032,8305>On a recent blender course, someone expressed to me his concern about the purity of the oxygen and its suitability for its intended purpose as it only contained 98.5% oxygen instead of 100% as expected. Although i had no other means to test the gas with me, reading the manual for the analyser did somewhat enlighten him! The problem is caused by the linearity of the sensors and analysers-that is, if we calibrate it in air to 21.0% then putting the cell into pure oxygen we can only expect the reading to be between about 97% and 103% because of the accuracy of the analyser and its various parts. Just because a meter has a reading to within 0.1% does not mean that it is right, just that it is easier to read! So what we must do is to use known gases to calibrate the analyser which are not too far removed in oxygen content from the mix we are trying to measure. It practise, this commonly means using air if we have a mix up to 50% oxygen and 100% oxygen if we wish to measure above 50%. It is much more important to be accurate when mixing the more usual types of mix around 26-40% than when mixing 60% or 80% as a variation of 1 or 2% will make very little difference to either the partial pressure or the decompression sch edule. <:#4572,8305>Another important factor in accurate analysis is the correct calibration point for the instrument. Consider the following: we set the instrument to read 21.0% and then find that our cylinder contains 40.0% according to the analyser. If we had set our initia l calibration to be 21.1% then our 40% would now read about 40.2%. If we had calibrated it to 20.9% then it would now show our 40% to be about 39.8%, a difference of +/- 0.2% or a total difference of 0.4% depending on how we initially started. An initial error of 0.2% would become +/- 0.4% and so on-the error multiplies itself up as we go on further. After taking a reading i like to check that the sensor returns to the correct calibration in air, as it is not difficult to move the calibration co ntrol by accident. A few minutes spent now can add greatly to your life expectancy! If you are in ANY doubt, do it again. So what are the ramifications of this possibly inaccurate analysis? Well basically, the nearer you are sailing to the wind in terms of oxygen partial pressure limits, the more account needs to be taken of possible errors. Although +/- 1% is an often quoted figure for the accuracy of a mix before recalculation needs to take place , i would tend to modify this to be +/- 1% of what is ACTUALLY IN THE CYLINDER not what the analyser tells you.Writing on a cylinder that the oxygen content is 35.6% and BELIEVING it is at best deluding yourself, at worst it can be dowright dangerous. Diving at what you think is 1.6 bar partial pressure of oxygen could actually be dangerously high if the above is not taken into account, another reason why at least one organistion recomm ends that you PLAN to dive at 1.4 maximum to give that bit more lee-way. Another practise that needs some comment is that of storing the cell in an airtight container in order to prolong its life. This is fine, provided that it is born in mind that when reactivated it can take several hours to stablise and this can be both an in convenience and another possible source of error. Although the cells are not cheap, it should not be beyond the means of the average diver to buy a new cell every couple of years if necessary. These hand-held analysers do give good results with careful use, so don't despair of the problems pointed out above, just be aware of them, and if you don't think that the person trying to sell you nitrox is doing it properly, you have the opportunity to te ll them now, and if they don't want to listen, remember that it is your right to take care of your own life by shopping elsewhere! <:#508,8305> <:#508,8305> <:#508,8305> <+B><:#508,8305>The End. > [Embedded] 00012975
Navigate by Author:
[Previous]
[Next]
[Author Search Index]
Navigate by Subject:
[Previous]
[Next]
[Subject Search Index]
[Send Reply] [Send Message with New Topic]
[Search Selection] [Mailing List Home] [Home]