
I have had several emails from new users over the last few weeks concerning
calibration.
The O2 sensors are rated by Teledyne as 10 mV in air +- 3 mV. So a working
sensor should be between 7 and 13 mV in air.
Since the computer is expecting about 98% O2 in the loop allowing for water
vapor and imperfect flushing, that means that the mV should be between 33 and 61
in O2 at one atmosphere.
It means that if the mV is not between 33 and 61
mV, then it is not possible that you have a good calibration. So the Shearwater
will fail that sensor and just flash 0.00. (If a sensor is voted out after
calibrating correctly, it still shows the PPO2, but it flashes.)
Once a sensor has failed calibration, it can't come back into the system
until there is a successful calibration. So if you unplug the failed sensor and
plug in a new one, it will still flash 0.00 until you calibrate. Basically, a
sensor that has failed calibration cannot spontaneously become suitable for the
voting logic to consider.
There are lots of reasons for this situation,
but here are a few. Low mV can be a dead sensor. It is not even linear to .98.
Or it can be a poor flush.
High mV are usually caused by leaving the O2
on. The KISS valve continues to inject, increasing the pressure and therefore
the mV. It can also be from using a bad sensor. Even though it is unusual, some
sensors fail high. There are also some sensors on the market that will fit in
the Teledyne hole but are just not compatible. They tend to fail high.
The millivolt display function is very useful for troubleshooting sensor
problems. At any time you can push both buttons at the same time, and the
display will switch from PPO2 to millivolts.
There is one more issue we
have come across. The K1D sensor allows the user to put the Molex connector on
backwards. The KISS displays will just show a negative number. The Shearwater,
and any other active electronics, won't work with the polarity reversed. It
won't hurt anything, it just won't work.
Although stops aren't always exactly to the minute, this is by design. If you
clear to the next ceiling after 10 seconds, it lets you go. It also takes a
finite amount of time to calculate the ceiling, so it's possible for a stop to
take a few seconds extra while the algorithm is calculating.
The TTS
prediction is usually within a minute. With 5 hours of deco, it's still within a
few minutes.
First, a bit of an overview of how the computer does this
stuff.
There are two things going on all the time.
First, the
compartment loading function uses the actual PPO2 and the actual depth to
calculate the changes in compartment gas loading.
Second, the TTS
function predicts the length of time at the current stop and the total TTS. It
does this by using the current PPO2 and the profile. It always assumes that you
are going to ascend smoothly at 10 metres per minute to the current ceiling, and
then stay at that ceiling until it clears. If you have more than one CC gas
programmed, it assumes that you are going to switch to that gas at an
appropriate depth.
Anything that you do on the dive that is off profile
will change the TTS and the time at the current stop prediction.
Some
things have very little effect. Small changes in PPO2 on the bottom have very
little effect. Small changes in PPO2 on shallow stops make a big difference.
Being a metre off the stop at deep stops makes little difference.
Being 3 metres off the stop at shallow stops makes a big difference.
Here
are a few things that could cause that sort of problem.
1) Programming a CC gas and then not switching to it.
The gas will flash in that case. For example, I got a call last
week about a 1 minute stop that took 4 minutes. He had 99% programmed as a CC
gas. So here's what the computer assumes. You have told it you are going to
switch by programming the gas in CC. If you were using it as bail out, you would
have programmed it as an OC gas. The computer assumes that you are going to
switch to O2 in the next second.
But you don't switch, so the compartment
loading routines continue to offgas with your actual PPO2. The next TTS
calculation, again assumes that you are going to switch to O2 in the next second
but you don't switch and on and on....
2) Following the
profile of another computer.
For example, if the GF is showing a
30m ceiling, and you start doing stops at 50m, the Pursuit will add deco time.
For the TTS to match, the profile has to match.
3) Not
being at the stop.
If you have a 3m stop of 20 minutes and
you stop at 6m, it will take significantly longer. This is similar to
number 1. The computer assumes that in the next second you are going to ascend
at 10m per minute to 3m and complete your deco there.
However
that compartment loading function is always calculating with your actual depth.
So the prediction is always wrong.
By the way, there's nothing wrong with
doing that. I will often do my 3m stop at 6m feet. The computer won't
penalize you. But the time at stop estimate will always be too short for the
above reasons.
The Pursuit computer will show all three cells by default. If you connect a cell only to cell 2 on the computer then it will ignore cell 1 and 3 only after you have calibrated the unit. This works in the other way, where you may have all three cells connected but it is only showing one cell in the middle of the screen. simply calibrate the computer in oxugen and all three will appear.
Only SAFT LS14500 batteries. No other battery should be used. Other 3.6v lithium batteries may cause unexpected switch off due to poor performing batteries or even result in damage to yoru computer. You change the battery by unscrewing the battery cap with a large coin and gently withdrawing the battery holder with a wire tool or similar.
The Pursuit was designed and built to exceed the requirements of the most demanding technical diver.
Simple, elegant solutions to the most common problems.
Personal service you can count on.
The Shearwater Advantage.
Copyright Narkedat90 2007. All rights reserved