AI Cells Again

dnscarboro

I have lost 4 cells from the same run 801324**-27-28-29 now 30...Working with AI to replace how ever this weekend lost #31 out of the package. Now # 25 is 156mv yes 156 anyone have one that high? Thanks

Dsix36

highest that I have had yet was 65mv

__________________
THE MORE THAT I LEARN, THE MORE THAT I STILL NEED TO LEARN!!!!!!

ScubaDadMiami

Sometimes cells will suddenly and dramatically increase the MV reading at the end of life. Not sure exactly how high it in fact goes when this happens, but, assuming all other aspects of the connections are working properly, it certainly sounds like something is not right with that cell you mention.

__________________
Howard Packer
IANTD CCR Instructor
Miami Beach, Florida
CCRDiveTraining.com

AD_ward9

Going high generally indicates the cell is losing electrolyte (leaking), or the internal load has gone. You can tell the latter easily by putting a 470 Ohm resistor across the cell and read the voltage. If the load is lost, then the voltage can go up and up (10V anyone?), if there is nothing to take the charge away.

The most common cause of loss of electrolyte is shock, second to that are sealing issues.

How many cells do you have from that batch?

Have AI identified the cause yet?

Incidentally, we recently found a new cell failure mode: oscillating cell. Very interesting, produces a nice waveform, the average of which is something near the PPO2 reading. We have some pictures. Still looking at what causes that one.

Alex

Drmike

aaaaannnnnd two more new AI cells I only put on my rig a few weeks ago have died!!!!

what a load of bloody crap!

Id send them back - but they will only send me more crap AI cells back again!!

__________________
Cave diving is a sport
Wreck diving is a sport
Diving in general is a sport

'Rebreather diving' is not a sport
its the delusional obsession with a highly dangerous and often inappropriate piece of equipment

anton610

Hello!

Are AI cells definitly crab or ...?

__________________
"Das Übel kommt nicht von der Technik, sondern von denen, die sie mißbrauchen."

Jacques-Yves Cousteau (1910-1997)

Lancer4545

With that sort of output you could do away with your batteries and run the unit off the cell........

Cripes and we are supposed to rely on these for "Life Support"......what a joke!!

Lance

__________________
You can run but you can't hide!

ISC Appointed Megalodon Dealer for East Australia
http://www.closedcircuitdivers.com.au
info@closedcircuitdivers.com.au

EBT

However this ends up playing out, both AI and Teledyne have had significant batch problems. The lessons to learn are;

- Manufacturers have good and bad times, validate your cells sensibly.
- Its not when things are going well that you get the true measure of a company, its during the bad times!

__________________
Eagles May Soar, but weasels don't get sucked into jet engines!

Rebreather World Terms of service
Real diving t-shirts for real divers

dnscarboro

Spoke with Mark @ AI he is sending complete replacements- will be returning the cells to see if they can advise what has happened. He will advise on the reason why these numbers are crazy. I expressed concern about the cells in a roll (lot)--- I can say he has been very good about working on this problem. We will see how this goes next week again...

AD_ward9

You are not supposed to rely on them for life support.

You are supposed to rely on an effective characterisation and masking algorithm.

I just tried uploading an Excel file with an example of a rebreather running nicely on one good cell out of 4, including where one of the cells is oscillating, but the Rebreather World uploader says "Invalid file", so instead below is the graph of the file.

• The purple curve marked Series 2 is the good cell, with a PPO2 set point of 0.6 ATM.
  
• The light blue curve marked Series 4 is an OK cell in a slightly noisy socket (water in it) that after half an hour is swapped with the cell in position Series 3 (monitoring suspended during swap out).
  
• The dark blue curve marked Series 1 is an oscillating cell that is swapped with the cell above after half an hour.
  
• The yellow curve marked Series 3 is a seriously ceiling limited cell, chosen for this test (as was the set point) because it limits just below the set point.

PPO2 is controlled very well, and user is warned there is only one good cell out of 4 in the first half of the run, and two out of four in the second half.

The cell in Series 2 position matches the PPO2 monitor readings on the top surface.

The little dips in the purple curve are due to the ADV firing. The samples are at one second intervals.

All eCCRs should have a very good means to characterise and mask cells automatically. Cells are far too unreliable to trust. There is enough data on cells published for enable anyone with the inclination to conceive of and implement a similar scheme.


Alex

Attached Images

cell masking example.png (27.2 KB, 309 views)