Current State Of Rebreather Electronics

[RainPilot]

Thanks again for the info, but I don’t understand, if the sss is such an improvement, why even bother with galvanic sensors at all? Why are they being so tentative with this new technology?

Cost. The sensor will be substantially more expensive to purchase at first (same price as galvanic over a ten year period though so a wash over lifetime cost)

Also, if you do manage to break one then a a galvanic backup on that big dive trip is a lot cheaper and simpler to obtain in the boonies.

 

[silent running]

Cost. The sensor will be substantially more expensive to purchase at first (same price as galvanic over a ten year period though so a wash over lifetime cost)

Also, if you do manage to break one then a a galvanic backup on that big dive trip is a lot cheaper and simpler to obtain in the boonies.

I see, thanks.

 

[tbone1004]

Thanks again for the info, but I don’t understand, if the sss is such an improvement, why even bother with galvanic sensors at all? Why are they being so tentative with this new technology?

A lot of divers are still woogity with them since none of the "big boys" have been able to test them on extreme dives yet. The initial concept was to get a few thousand hours with the SSS and a pair of galvanics since they're a known bad technology, but at least known, then move around to 2x sss and one galvanic, then just 2x sss. I think cost will come down with scale, but they're going to have to get some of the production quantity issues sorted out before they get adopted by DiveCAN and the rest of the market. Of note, these are optical sensors, so they'll have to go into the DiveCAN with their own modules, and if you put them in something like a KISS, it would have to be with a dedicated computer.

 

[silent running]

I have asked for current draw requirements on the Poseidon sss in several forums, but haven’t seen them anywhere yet. I’m pretty sure they’re probably enough to drain a 9v in short order as the light element is likely on all the time, never mind the computational power required to process the light output signal. If the idea is running 2 for safety/validation then we must plan for battery draw from 2, which will be significant and require a larger battery than previously used in any ECCR, not a good thing given the physics of water pressure and electricity.

My point is, the reliability of the sensors in a CCR loop is only one hurdle in their implementation. Galvanic sensors are faulty but their faults are well known, and perhaps most importantly, they don’t require high current to give the diver PO2 info. An sss will require full battery voltage to do the same basic job as a galvanic sensor, and a 1ATA battery to function. Which failure is more likely: Higher voltage under 10ATA or MV? MV readings seem to have no problem transmitting in 99% humidity at the ambient pressure of the loop. We are all used to living with the failure modes of galvanic sensors, but if the alternative requires higher voltages to operate under 100’s of psi of sea water, is the additional accuracy worth it?

I think it’s a legitimate question...

 

[njdiverjoe]

I have asked for current draw requirements on the Poseidon sss in several forums, but haven’t seen them anywhere yet. I’m pretty sure they’re probably enough to drain a 9v in short order as the light element is likely on all the time, never mind the computational power required to process the light output signal. If the idea is running 2 for safety/validation then we must plan for battery draw from 2, which will be significant and require a larger battery than previously used in any ECCR, not a good thing given the physics of water pressure and electricity.

My point is, the reliability of the sensors in a CCR loop is only one hurdle in their implementation. Galvanic sensors are faulty but their faults are well known, and perhaps most importantly, they don’t require high current to give the diver PO2 info. An sss will require full battery voltage to do the same basic job as a galvanic sensor, and a 1ATA battery to function. Which failure is more likely: Higher voltage under 10ATA or MV? MV readings seem to have no problem transmitting in 99% humidity at the ambient pressure of the loop. We are all used to living with the failure modes of galvanic sensors, but if the alternative requires higher voltages to operate under 100’s of psi of sea water, is the additional accuracy worth it?

I think it’s a legitimate question...

I respectfully disagree and would be surprised if high current or high voltage is required to operate the new "solid state" O2 sensors.

Other than some changes to the chemistry of the dye to accommodate really high pO2s, I don't believe the sensors are much of a departure from the current optical oxygen sensors. At the end of the day the sensor consists of an LED and a photodiode. Neither components requires high voltage or current.

To give us an idea we can check out other optical oxygen sensors we can a datasheet for (see for example https://www.sstsensing.com/wp-content/uploads/2018/01/DS0144rev2_LOX-02-S.pdf ). On that one it expects a 5V input and draws something like 7.5mA to 20mA, depending on how fast you want to sample the sensor. That means one could operate continuously on a single 9V for 45 to 120 hours.

 

[tbone1004]

I'm with @njdiverjoe the light that is always on is a LED.... It's driven off of the internal battery on the M28 in the CPOD which is a tiny li-ion rechargeable, and gets 25 hours. On a 9v it would get at least that much, it will certainly draw significantly less power than running the whole dive off of the solenoid. It's far from " high voltage, high current"

 

[silent running]

I respectfully disagree and would be surprised if high current or high voltage is required to operate the new "solid state" O2 sensors.

Other than some changes to the chemistry of the dye to accommodate really high pO2s, I don't believe the sensors are much of a departure from the current optical oxygen sensors. At the end of the day the sensor consists of an LED and a photodiode. Neither components requires high voltage or current.

To give us an idea we can check out other optical oxygen sensors we can a datasheet for (see for example https://www.sstsensing.com/wp-content/uploads/2018/01/DS0144rev2_LOX-02-S.pdf ). On that one it expects a 5V input and draws something like 7.5mA to 20mA, depending on how fast you want to sample the sensor. That means one could operate continuously on a single 9V for 45 to 120 hours.

Thanks for the data and your respectful disagreement. To be clear I define high current in this context as anything not measured in millivolts. The high current I refer to is battery voltages, and as we all know battery voltage does not mix well with any level of moisture and the more there is passing through cables and connections, the more failure points there are. I’m not saying electronics are shorting out on CCRs all the time, but as we have seen with even the biggest manufacturers, there have been plenty of failures of cables, connectors and switches which operate with battery voltages.

As far as the current draw, if we are talking about 2 sss, a computer and a solenoid, we are definitely talking about a bigger battery than a single 9v, probably need some type of high capacity rechargeable. Which should be potted and removable, which sounds expensive to manufacture and replace.

Sss or not, my main point in all of this is that the more complex we make our CCRs the more that can go wrong. I place a very high value on the fact that I can read my high output galvanic sensors directly on an analog needle gage, without any battery voltages or additional vulnerable battery voltage driven components. Galvanic sensors are simple devices whose failure modes are known and can be diagnosed during the dive. It’s a question of what you are comfortable with and what you put your faith in: something simple but which may require some brain power to troubleshoot, or something complex but “plug and play” which you cannot troubleshoot and on which you are totally dependent...

 

[rjack321]

As far as the current draw, if we are talking about 2 sss, a computer and a solenoid, we are definitely talking about a bigger battery than a single 9v, probably need some type of high capacity rechargeable. Which should be potted and removable, which sounds expensive to manufacture and replace.

Why would you pot it? The battery compartment on a shearwater basically never floods.

 

[silent running]

Why would you pot it? The battery compartment on a shearwater basically never floods.

Because the battery would be bigger than any battery that goes in a Shearwater, and I doubt you want a 1ATA compartment that big in the head.

 

[Jeremy Williams]

The XCCR has two 1 atm compartments for 18650 batteries. They are outside of the loop also. It’s plenty of power for electronics, a solenoid and electronic sensors.