2 point calibration. Why?

**UWSojourner** · 3rd July 2006, 05:54

First, everyone is doing 2 point calibration. You must have 2 points in order for the electronics to read PO2 over the range of PO2s we normally run on our RBs.

The question is "What 2 points?" Most RBs use (0,0) as 1 point assuming, with some theoretical basis, that without O2 the sensors won't respond and, therefore, will read 0 MVs. The second point is usually performed in as rich a PO2 as possible, normally something close to 100%. If it were possible (and easy) for me to calibrate at 1.3, I would. In general you want to calibrate as close to the PO2 you expect to run as possible.

The Apecs Meg calibrates not on (0,0) but on (Air MV, 21). This, theoretically, shows two points in actual observed MVs. Is it more accurate? I have no real data, but

1. In a perfect world it would produce identical results.

2. In a world of random errors, it would produce situations where random errors at 21% and 100% combine and produce a larger error range than simply the random error at 100% (combined with the 0,0 point which can't be far off).

3. In a world of diver induced error, well, its just another thing for us divers to screw up.

4. In a world where (0,0) is not the only observed value for 0% O2, then the two point calibration may pick it up and correct for it. My understanding from reading the Teledyne sensor spec is that they actually show a small range of readings for the sensors at 0% O2.

I use (0,0) as one of my points since I'm diving the Shearwater on the Meg. But if I was using another point in lieu of the (0,0) point, I'd perform a check to determine the implied intercept of the calibration equation. In other words, if the implied intercept is (2.5, 0), then my calibration equation implies that 0% O2 will generate 2.5 MVs. Probably not the best situation.

Here's how to perform the check.

1. Get your 2 points - ( AirMV, 21) and (HighO2MV, HighO2).
2. Calculate the implied intercept point ( MVAtZeroO2, 0 ) with the following formula:

MVAtZeroO2 = HighO2 - (HighO2-21)/(HighO2MV-AirMV) * HighO2MV

Example: Your 2 points are (9.6, 21) and (45.3, 98). Therefore, your implied zero intercept is (0.29, 0 ), or without any O2 you will generate 0.29 MVs.
This was calculated as 0.29 = 98 - (98-21)/(45.3-9.6) * 45.3
This seems reasonable (to me).

However, assume your 2 points are (8.6, 21) and (47.8, 98). Then you have an implied zero intercept of (4.1, 0). Somehow your calibration is telling you that you can generate 4.1MVs with no oxygen present. Probably not your best calibration effort. Maybe you want to do some more thinking before you dive.

Sorry, this got kinda long.

**wedivebc** · 3rd July 2006, 06:09

Quote: (Originally Posted by UWSojourner)

Sorry, this got kinda long.

Your posts are never too long

Thanks for the insight Kevin.

Gilles · 3rd July 2006, 14:11

[quote=Here's how to perform the check.

1. Get your 2 points - ( AirMV, 21) and (HighO2MV, HighO2).
2. Calculate the implied intercept point ( MVAtZeroO2, 0 ) with the following formula:

MVAtZeroO2 = HighO2 - (HighO2-21)/(HighO2MV-AirMV) * HighO2MV

Example: Your 2 points are (9.6, 21) and (45.3, 98). Therefore, your implied zero intercept is (0.29, 0 ), or without any O2 you will generate 0.29 MVs.
This was calculated as 0.29 = 98 - (98-21)/(45.3-9.6) * 45.3
This seems reasonable (to me).

However, assume your 2 points are (8.6, 21) and (47.8, 98). Then you have an implied zero intercept of (4.1, 0). Somehow your calibration is telling you that you can generate 4.1MVs with no oxygen present. Probably not your best calibration effort. Maybe you want to do some more thinking before you dive.

Sorry, this got kinda long.[/quote]

This works, but a bit complicated if your not an engineering/mathematical type.

Alternatively, you can look up the corresponding "rich gas" O2 reading you should get for each of your cell readings in air, in the attached table. Any under-shoots, corresponds to your zero O2 mv, and indicates possibility of current limitation.

I like the idea of placing the sensors under higher pressure. If they reach a proper mV @ 2 bar PO2, then you have reason for confidence!.

G

**UWSojourner** · 3rd July 2006, 17:20

Quote: (Originally Posted by sabgia)

This is a bit complicated if your not an engineering/mathematical type.

Other types? Here?

RonMicjan · 3rd July 2006, 17:48

Quote: (Originally Posted by UWSojourner)

Other types? Here?

uh, yup, that would be me, Im a lover not a mathematician.

thanks for bringing it down to my level Kev!

On the original question. The process of meg calibration is not calibrate to one gas then check to see the other is correct. (although this is good for the KISS Rebreather's) You actually tell the meg when its at air, then again to tell it its at 100% (or whatever gas you are using) then you save the data. Keep in mind that if you lie to your meg, it will lie to you, garbage in, garbage out. this is a reason to occasionally check your mv outputs during a dive and reference that to the PO2 the unit is reading. If you consistantly know that 65mv = 1.2 po2, then the day you look at mv during a dive and see 30mv, you might want to abort that dive and recheck calibration.

mortenkjerulff · 3rd July 2006, 18:13

Quote: (Originally Posted by RonMicjan)

On the original question. The process of meg calibration is not calibrate to one gas then check to see the other is correct. (although this is good for the KISS Rebreather's) You actually tell the meg when its at air, then again to tell it its at 100% (or whatever gas you are using) then you save the data. Keep in mind that if you lie to your meg, it will lie to you, garbage in, garbage out. this is a reason to occasionally check your mv outputs during a dive and reference that to the PO2 the unit is reading. If you consistantly know that 65mv = 1.2 po2, then the day you look at mv during a dive and see 30mv, you might want to abort that dive and recheck calibration.

Hi Ron.

Could you elaborate on this. If you have 3 known points at 0%, 21% and say 100% (and the third point is not just to cross check), what does the Meg software do with these points ?

Morten

RonMicjan · 3rd July 2006, 18:58

just like what kevin described, it maps all the points between the cal points and then projects out beyond those points, like drawing a line on a graph. The cal points define the lines slope and rise. one could assume that 0 mv=0.0 po2 as the first point, but what the meg does is peg the first point at air 7-13mv (depends on the individual sensor) = .209 PO2 The meg still only uses two points, not three. It is getting an actual value, not just assuming that it is zero. ISC has done testing on sensors and they are actually quite linear right up to the saturation point, so either method is plenty accurate.

**UWSojourner** · 3rd July 2006, 19:29

Quote: (Originally Posted by RonMicjan)

uh, yup, that would be me, Im a lover not a mathematician.

I didn't know it, but you're also a poet as I see from one of your other posts.

Quote: (Originally Posted by RonMicjan)

If you consistantly know that 65mv = 1.2 po2, then the day you look at mv during a dive and see 30mv, you might want to abort that dive and recheck calibration.

I like this as a check of last resort, but if you knew that your sensors generated about 65mv in a 1.2PO2 environment, then you could check your calibration before you get in the water. I've attached a spreadsheet if anyone wants to play around to see 2-point-air vs 2-point-intercept calibration.

The calculated item "MVs Needed to Generate of PO2 of" would show you whether your 65mv will display what you expect. Anyway, hope the spreadsheet makes sense.

mortenkjerulff · 3rd July 2006, 19:48

Quote: (Originally Posted by RonMicjan)

just like what kevin described, it maps all the points between the cal points and then projects out beyond those points, like drawing a line on a graph. The cal points define the lines slope and rise. one could assume that 0 mv=0.0 po2 as the first point, but what the meg does is peg the first point at air 7-13mv (depends on the individual sensor) = .209 PO2 The meg still only uses two points, not three. It is getting an actual value, not just assuming that it is zero. ISC has done testing on sensors and they are actually quite linear right up to the saturation point, so either method is plenty accurate.

Thanks.
Sorry Kevin, I did not read your post before now, somehow

. It explains things very well for me.

RonMicjan · 3rd July 2006, 20:01

Quote: (Originally Posted by UWSojourner)

I didn't know it, but you're also a poet as I see from one of your other posts.

getting mixed reviews on that one..

patti has taken to calling me "Longfellow", must be my shoe size...

3rd July 2006, 05:54	#11 (permalink)
UWSojourner Pacific Northwest Current Rebreather/s: Megalodon Other Rebreather/s: Join Date: Feb 2005 Location: Portland Oregon Posts: 558	Re: 2 point calibration. Why? First, everyone is doing 2 point calibration. You must have 2 points in order for the electronics to read PO2 over the range of PO2s we normally run on our RBs. The question is "What 2 points?" Most RBs use (0,0) as 1 point assuming, with some theoretical basis, that without O2 the sensors won't respond and, therefore, will read 0 MVs. The second point is usually performed in as rich a PO2 as possible, normally something close to 100%. If it were possible (and easy) for me to calibrate at 1.3, I would. In general you want to calibrate as close to the PO2 you expect to run as possible. The Apecs Meg calibrates not on (0,0) but on (Air MV, 21). This, theoretically, shows two points in actual observed MVs. Is it more accurate? I have no real data, but 1. In a perfect world it would produce identical results. 2. In a world of random errors, it would produce situations where random errors at 21% and 100% combine and produce a larger error range than simply the random error at 100% (combined with the 0,0 point which can't be far off). 3. In a world of diver induced error, well, its just another thing for us divers to screw up. 4. In a world where (0,0) is not the only observed value for 0% O2, then the two point calibration may pick it up and correct for it. My understanding from reading the Teledyne sensor spec is that they actually show a small range of readings for the sensors at 0% O2. I use (0,0) as one of my points since I'm diving the Shearwater on the Meg. But if I was using another point in lieu of the (0,0) point, I'd perform a check to determine the implied intercept of the calibration equation. In other words, if the implied intercept is (2.5, 0), then my calibration equation implies that 0% O2 will generate 2.5 MVs. Probably not the best situation. Here's how to perform the check. 1. Get your 2 points - ( AirMV, 21) and (HighO2MV, HighO2). 2. Calculate the implied intercept point ( MVAtZeroO2, 0 ) with the following formula: MVAtZeroO2 = HighO2 - (HighO2-21)/(HighO2MV-AirMV) * HighO2MV Example: Your 2 points are (9.6, 21) and (45.3, 98). Therefore, your implied zero intercept is (0.29, 0 ), or without any O2 you will generate 0.29 MVs. This was calculated as 0.29 = 98 - (98-21)/(45.3-9.6) * 45.3 This seems reasonable (to me). However, assume your 2 points are (8.6, 21) and (47.8, 98). Then you have an implied zero intercept of (4.1, 0). Somehow your calibration is telling you that you can generate 4.1MVs with no oxygen present. Probably not your best calibration effort. Maybe you want to do some more thinking before you dive. Sorry, this got kinda long. Last edited by UWSojourner : 3rd July 2006 at 16:24.
(Offline)

3rd July 2006, 06:09	#12 (permalink)
wedivebc Dave Tomblin Current Rebreather/s: Megalodon Other CCR Other Rebreather/s: Not Bought Yet Inspiration Classic Sport Kiss Other CCR Join Date: Feb 2005 Location: Vancouver Island BC Canada Posts: 1,508	Re: 2 point calibration. Why? Quote: (Originally Posted by UWSojourner) Sorry, this got kinda long. Your posts are never too long Thanks for the insight Kevin. __________________ Cheers, Dave.... Man is the only animal burdened with the knowledge he will eventually die
(Offline)

3rd July 2006, 17:20	#14 (permalink)
UWSojourner Pacific Northwest Current Rebreather/s: Megalodon Other Rebreather/s: Join Date: Feb 2005 Location: Portland Oregon Posts: 558	Re: 2 point calibration. Why? Quote: (Originally Posted by sabgia) This is a bit complicated if your not an engineering/mathematical type. Other types? Here?
(Offline)

3rd July 2006, 17:48	#15 (permalink)
RonMicjan Cap Ron scourge of the NW Current Rebreather/s: Megalodon Dolphin Home Build Other Rebreather/s: Sport Kiss Classic Kiss Dolphin Home Build Join Date: Feb 2005 Location: Cape Disapointment, The Graveyard of the Pacific Posts: 1,120	Re: 2 point calibration. Why? Quote: (Originally Posted by UWSojourner) Other types? Here? uh, yup, that would be me, Im a lover not a mathematician. thanks for bringing it down to my level Kev! On the original question. The process of meg calibration is not calibrate to one gas then check to see the other is correct. (although this is good for the KISS Rebreather's) You actually tell the meg when its at air, then again to tell it its at 100% (or whatever gas you are using) then you save the data. Keep in mind that if you lie to your meg, it will lie to you, garbage in, garbage out. this is a reason to occasionally check your mv outputs during a dive and reference that to the PO2 the unit is reading. If you consistantly know that 65mv = 1.2 po2, then the day you look at mv during a dive and see 30mv, you might want to abort that dive and recheck calibration. __________________ Marine rescue, towing and salvage. Interfering with natural selection since 1983. www.tmishop.com Diving bits etc. www.seatowpdx.com The Summer Job
(Offline)

3rd July 2006, 18:13	#16 (permalink)
mortenkjerulff Customise Me! Current Rebreather/s: Megalodon Other Rebreather/s: Not Bought Yet Join Date: Mar 2005 Location: Denmark Posts: 151	Re: 2 point calibration. Why? Quote: (Originally Posted by RonMicjan) On the original question. The process of meg calibration is not calibrate to one gas then check to see the other is correct. (although this is good for the KISS Rebreather's) You actually tell the meg when its at air, then again to tell it its at 100% (or whatever gas you are using) then you save the data. Keep in mind that if you lie to your meg, it will lie to you, garbage in, garbage out. this is a reason to occasionally check your mv outputs during a dive and reference that to the PO2 the unit is reading. If you consistantly know that 65mv = 1.2 po2, then the day you look at mv during a dive and see 30mv, you might want to abort that dive and recheck calibration. Hi Ron. Could you elaborate on this. If you have 3 known points at 0%, 21% and say 100% (and the third point is not just to cross check), what does the Meg software do with these points ? Morten
(Offline)

3rd July 2006, 18:58	#17 (permalink)
RonMicjan Cap Ron scourge of the NW Current Rebreather/s: Megalodon Dolphin Home Build Other Rebreather/s: Sport Kiss Classic Kiss Dolphin Home Build Join Date: Feb 2005 Location: Cape Disapointment, The Graveyard of the Pacific Posts: 1,120	Re: 2 point calibration. Why? just like what kevin described, it maps all the points between the cal points and then projects out beyond those points, like drawing a line on a graph. The cal points define the lines slope and rise. one could assume that 0 mv=0.0 po2 as the first point, but what the meg does is peg the first point at air 7-13mv (depends on the individual sensor) = .209 PO2 The meg still only uses two points, not three. It is getting an actual value, not just assuming that it is zero. ISC has done testing on sensors and they are actually quite linear right up to the saturation point, so either method is plenty accurate. __________________ Marine rescue, towing and salvage. Interfering with natural selection since 1983. www.tmishop.com Diving bits etc. www.seatowpdx.com The Summer Job
(Offline)

3rd July 2006, 19:48	#19 (permalink)
mortenkjerulff Customise Me! Current Rebreather/s: Megalodon Other Rebreather/s: Not Bought Yet Join Date: Mar 2005 Location: Denmark Posts: 151	Re: 2 point calibration. Why? Quote: (Originally Posted by RonMicjan) just like what kevin described, it maps all the points between the cal points and then projects out beyond those points, like drawing a line on a graph. The cal points define the lines slope and rise. one could assume that 0 mv=0.0 po2 as the first point, but what the meg does is peg the first point at air 7-13mv (depends on the individual sensor) = .209 PO2 The meg still only uses two points, not three. It is getting an actual value, not just assuming that it is zero. ISC has done testing on sensors and they are actually quite linear right up to the saturation point, so either method is plenty accurate. Thanks. Sorry Kevin, I did not read your post before now, somehow. It explains things very well for me.
(Offline)

3rd July 2006, 20:01	#20 (permalink)
RonMicjan Cap Ron scourge of the NW Current Rebreather/s: Megalodon Dolphin Home Build Other Rebreather/s: Sport Kiss Classic Kiss Dolphin Home Build Join Date: Feb 2005 Location: Cape Disapointment, The Graveyard of the Pacific Posts: 1,120	Re: 2 point calibration. Why? Quote: (Originally Posted by UWSojourner) I didn't know it, but you're also a poet as I see from one of your other posts. getting mixed reviews on that one.. patti has taken to calling me "Longfellow", must be my shoe size... __________________ Marine rescue, towing and salvage. Interfering with natural selection since 1983. www.tmishop.com Diving bits etc. www.seatowpdx.com The Summer Job
(Offline)

Thread Tools
Show Printable Version Email this Page
Display Modes
Linear Mode Switch to Hybrid Mode Switch to Threaded Mode