sensor load and life expectancy

York · 30th September 2007, 16:10

Quote: (Originally Posted by aknelson)

Good question........

The internal current load by the approximatly 100 Ohm internal cell circuit is much greater than the current load added by the 10k external load.
The cells internal load actually causes the voltage you are measuring at the pins. Since a current flows between the electrodes.

I have seen different industrial designs for analog input circuits. Dependig what you want to measure a Current or a Voltage; A load resistor is in place mainly to protect the analog input. A lot of times a capacitor and a zehner diode is added too to limit the input voltage. Resistors in series and capacitors in parallel result in a signal delay or a low pass filter. The 10k load could be part of the recommended analog input circuit. Unless you know the complete analog input circuit it is hard to say that the 10K makes a difference. This brings us to the question how rugged the circuit is for electromagnetic compability if the circuit was ever evaluated for EMC.

Regards,
Andreas

I was just thinking along similar lines. The 10K load resistor is usually placed close to the op-amps input. See my post and edit just as you were submitting yours ...

aknelson · 30th September 2007, 16:46

Quote: (Originally Posted by York)

I was just thinking along similar lines. The 10K load resistor is usually placed close to the op-amps input. See my post and edit just as you were submitting yours ...

Well it needs to be added that the OP-amplifiers are only in older design circuits required. You need a 10nf capacitor as a minimum parallel to the analog input.
Todays low power Analog circuits have the amplifiers build in and are programmable beside 24 bit accuracy.

Andreas

**nigelh** · 30th September 2007, 16:54

The cells are always current limited or the output voltage would be set by the chemistry not the quantity of fuel.
The simplest cct would just be a resistor across the cell, say 80R and the voltage read would be proportional to the current and thence to the fuel supply.
Sadly the cells are also quite good thermometers so some sort of compensation is needed and a bead thermistor stuffed down the back of the cell works for Teledyne. They keep playing with the circuit. The first ones I took to bits had the thermistor on the top of the PCB well away from the cell.

I don't think the 10K load is significant. They just want you to use >10K so there is no significant loading.

The cct came from

the write up is here.

***jradomski*** · 30th September 2007, 17:10

Quote: (Originally Posted by York)

The circuit is by concept slightly different from what Nigel quotes. There is no in-line resistor (probably to deliver the maximum potential to the connector).

When you say 3 plus thermistor, do you mean including the 10K load?

They could have included the 10K into the the internal circuit, so there would be no requirement for an additional resistor externally - historical reasons possibly? Earlier sensors needed it, so now they stick to it?

Edit: Just thought about one more possibility: without the external load, the electronics would be measuring the potential, but without current flow (nA or less). This would make measurement very susceptible to interference from magnetic fields if long sensor leads are used. By adding the requirement for an external load (and planning for it in their internal circuit) it would ensure at least some current flows, hence provide more stabil reading - maybe?

J

Typo should have been 2 (corrected)

York · 30th September 2007, 17:48

Quote: (Originally Posted by nigelh)

The cells are always current limited or the output voltage would be set by the chemistry not the quantity of fuel.
The simplest cct would just be a resistor across the cell, say 80R and the voltage read would be proportional to the current and thence to the fuel supply.
Sadly the cells are also quite good thermometers so some sort of compensation is needed and a bead thermistor stuffed down the back of the cell works for Teledyne. They keep playing with the circuit. The first ones I took to bits had the thermistor on the top of the PCB well away from the cell.

I don't think the 10K load is significant. They just want you to use >10K so there is no significant loading.

The cct came from

the write up is here.

I agree 100% on the current limitation, no idea where that expression came from to describe the aging effect. I believe "diffusion limited" may be more accurate (as electrolyte depletes, and product buildup occurs on the anode), but that would be another discussion.

So you are saying there are historic reasons for the 10K load, and it could be omitted without too much of a negative influence? I know the VR3 includes a 100K load and is used for the R22s - supposedly successfully. We were looking at EMFs and their influence, not sure we came to a conclusion.

Thanks for the link, a nice write-up.

This is a great discussion folks, I learned a lot! thanks!!!

Joerg

York · 30th September 2007, 17:52

Quote: (Originally Posted by aknelson)

Well it needs to be added that the OP-amplifiers are only in older design circuits required. You need a 10nf capacitor as a minimum parallel to the analog input.
Todays low power Analog circuits have the amplifiers build in and are programmable beside 24 bit accuracy.

Andreas

Not sure if 24 bits are needed, but would you have an example for an IC with 3 build-in opamps? Why are build-in op-amps different than external op-amps?

thanks
Joerg

WD8CDH · 28th January 2008, 10:10

The main reason for 10K load (or whatever is spec'd by the manufacturer) is to maintain temperature stability. A higher value will give you only a max. of about 1% longer life.

The only two things that I can think of to improve life is:

1. store sensors in dry nitrogen or Argon.

2. back bias the cells at a voltage very near the open circuit voltage of the internal sensor to reduce current generation.

Number 2 is difficult to do sense you don't easily know the open circuit voltage without opening the cell.

Mike_Eitel · 28th January 2008, 11:41

Hi WD8CDH

Feeding back is not a good idea

As we had too many defective cells, I had a while agoo a long discussion with a develloper of cells of AI. As he explained me, backfeeding current is a sure killer. Since I try to avoid that, I got longer lifetime ...

Regards Mike

http://www.rebreatherworld.com/cells...o2-sensor.html

WD8CDH · 28th January 2008, 13:38

Hi Mike,It depends on the Polarity. If you reverse the polarity, it will shorten the life of the cell or kill it compleatly. Also if you drive it in the proper polarity but too high, it will kill it too.I have only tried it with the Japan Storage Battery KE-25 and I wouldn't reccomend it for a rebreather but it has worked for me on a number of cells in O2 analysers.

**nigelh** · 28th January 2008, 14:06

Quote: (Originally Posted by WD8CDH)

but it has worked for me on a number of cells in O2 analysers.

It sounds a horrible thing to do to the chemistry of a cell.

Were the ones you tried it on accurate afterwards or did they just give a reading and that was enough for the user?

30th September 2007, 16:10	#11 (permalink)
York probubbly not Current Rebreather/s: Megalodon Other Rebreather/s: Join Date: Mar 2005 Location: Bristol, UK Posts: 118	Re: sensor load and life expectancy Quote: (Originally Posted by aknelson) Good question........ The internal current load by the approximatly 100 Ohm internal cell circuit is much greater than the current load added by the 10k external load. The cells internal load actually causes the voltage you are measuring at the pins. Since a current flows between the electrodes. I have seen different industrial designs for analog input circuits. Dependig what you want to measure a Current or a Voltage; A load resistor is in place mainly to protect the analog input. A lot of times a capacitor and a zehner diode is added too to limit the input voltage. Resistors in series and capacitors in parallel result in a signal delay or a low pass filter. The 10k load could be part of the recommended analog input circuit. Unless you know the complete analog input circuit it is hard to say that the 10K makes a difference. This brings us to the question how rugged the circuit is for electromagnetic compability if the circuit was ever evaluated for EMC. Regards, Andreas I was just thinking along similar lines. The 10K load resistor is usually placed close to the op-amps input. See my post and edit just as you were submitting yours ...
(Offline)

30th September 2007, 16:46	#12 (permalink)
aknelson New Member Current Rebreather/s: Ouroboros Dolphin Other Rebreather/s: Ouroboros Dolphin Join Date: Dec 2005 Location: Minnesota Posts: 69	Re: sensor load and life expectancy Quote: (Originally Posted by York) I was just thinking along similar lines. The 10K load resistor is usually placed close to the op-amps input. See my post and edit just as you were submitting yours ... Well it needs to be added that the OP-amplifiers are only in older design circuits required. You need a 10nf capacitor as a minimum parallel to the analog input. Todays low power Analog circuits have the amplifiers build in and are programmable beside 24 bit accuracy. Andreas
(Offline)

30th September 2007, 16:54	#13 (permalink)
nigelh New Member Current Rebreather/s: Inspiration Vision Other CCR Other Rebreather/s: Inspiration Classic Other CCR Join Date: Mar 2005 Location: Brighton, Sussex, UK Posts: 784	Re: sensor load and life expectancy The cells are always current limited or the output voltage would be set by the chemistry not the quantity of fuel. The simplest cct would just be a resistor across the cell, say 80R and the voltage read would be proportional to the current and thence to the fuel supply. Sadly the cells are also quite good thermometers so some sort of compensation is needed and a bead thermistor stuffed down the back of the cell works for Teledyne. They keep playing with the circuit. The first ones I took to bits had the thermistor on the top of the PCB well away from the cell. I don't think the 10K load is significant. They just want you to use >10K so there is no significant loading. The cct came from the write up is here. __________________ nigelh
(Online)

30th September 2007, 17:10	#14 (permalink)
*jradomski* Moderator Current Rebreather/s: Inspiration Classic Sport Kiss Optima rEvo Other CCR Home Build Other Rebreather/s: Inspiration Vision Evolution Megalodon Classic Kiss rEvo Other CCR Home Build Join Date: Mar 2005 Location: "Da" Bronx Posts: 2,796	Re: sensor load and life expectancy Quote: (Originally Posted by York) The circuit is by concept slightly different from what Nigel quotes. There is no in-line resistor (probably to deliver the maximum potential to the connector). When you say 3 plus thermistor, do you mean including the 10K load? They could have included the 10K into the the internal circuit, so there would be no requirement for an additional resistor externally - historical reasons possibly? Earlier sensors needed it, so now they stick to it? Edit: Just thought about one more possibility: without the external load, the electronics would be measuring the potential, but without current flow (nA or less). This would make measurement very susceptible to interference from magnetic fields if long sensor leads are used. By adding the requirement for an external load (and planning for it in their internal circuit) it would ensure at least some current flows, hence provide more stabil reading - maybe? J Typo should have been 2 (corrected) __________________ Joe Radomski CCR Trimix Instructor Trainer ANDI Instructor Trainer Director #10 All posts are personal opinions and DO NOT reflect any affiliated agency unless specifically stated.
(Online)

30th September 2007, 17:48	#15 (permalink)
York probubbly not Current Rebreather/s: Megalodon Other Rebreather/s: Join Date: Mar 2005 Location: Bristol, UK Posts: 118	Re: sensor load and life expectancy Quote: (Originally Posted by nigelh) The cells are always current limited or the output voltage would be set by the chemistry not the quantity of fuel. The simplest cct would just be a resistor across the cell, say 80R and the voltage read would be proportional to the current and thence to the fuel supply. Sadly the cells are also quite good thermometers so some sort of compensation is needed and a bead thermistor stuffed down the back of the cell works for Teledyne. They keep playing with the circuit. The first ones I took to bits had the thermistor on the top of the PCB well away from the cell. I don't think the 10K load is significant. They just want you to use >10K so there is no significant loading. The cct came from the write up is here. I agree 100% on the current limitation, no idea where that expression came from to describe the aging effect. I believe "diffusion limited" may be more accurate (as electrolyte depletes, and product buildup occurs on the anode), but that would be another discussion. So you are saying there are historic reasons for the 10K load, and it could be omitted without too much of a negative influence? I know the VR3 includes a 100K load and is used for the R22s - supposedly successfully. We were looking at EMFs and their influence, not sure we came to a conclusion. Thanks for the link, a nice write-up. This is a great discussion folks, I learned a lot! thanks!!! Joerg
(Offline)

30th September 2007, 17:52	#16 (permalink)
York probubbly not Current Rebreather/s: Megalodon Other Rebreather/s: Join Date: Mar 2005 Location: Bristol, UK Posts: 118	Re: sensor load and life expectancy Quote: (Originally Posted by aknelson) Well it needs to be added that the OP-amplifiers are only in older design circuits required. You need a 10nf capacitor as a minimum parallel to the analog input. Todays low power Analog circuits have the amplifiers build in and are programmable beside 24 bit accuracy. Andreas Not sure if 24 bits are needed, but would you have an example for an IC with 3 build-in opamps? Why are build-in op-amps different than external op-amps? thanks Joerg
(Offline)

28th January 2008, 10:10	#17 (permalink)
WD8CDH New Member Current Rebreather/s: MK 15.X Other CCR Other SCR Home Build Other Rebreather/s: Not Bought Yet Other CCR Other SCR Home Build Join Date: Dec 2007 Location: NY Posts: 51	Re: sensor load and life expectancy The main reason for 10K load (or whatever is spec'd by the manufacturer) is to maintain temperature stability. A higher value will give you only a max. of about 1% longer life. The only two things that I can think of to improve life is: 1. store sensors in dry nitrogen or Argon. 2. back bias the cells at a voltage very near the open circuit voltage of the internal sensor to reduce current generation. Number 2 is difficult to do sense you don't easily know the open circuit voltage without opening the cell.
(Offline)

28th January 2008, 11:41	#18 (permalink)
Mike_Eitel New Member Current Rebreather/s: Home Build Other Rebreather/s: Home Build Join Date: May 2006 Location: Switzerland Posts: 5	Re: sensor load and life expectancy Hi WD8CDH Feeding back is not a good idea As we had too many defective cells, I had a while agoo a long discussion with a develloper of cells of AI. As he explained me, backfeeding current is a sure killer. Since I try to avoid that, I got longer lifetime ... Regards Mike http://www.rebreatherworld.com/cells...o2-sensor.html
(Offline)

28th January 2008, 13:38	#19 (permalink)
WD8CDH New Member Current Rebreather/s: MK 15.X Other CCR Other SCR Home Build Other Rebreather/s: Not Bought Yet Other CCR Other SCR Home Build Join Date: Dec 2007 Location: NY Posts: 51	Re: sensor load and life expectancy Hi Mike,It depends on the Polarity. If you reverse the polarity, it will shorten the life of the cell or kill it compleatly. Also if you drive it in the proper polarity but too high, it will kill it too.I have only tried it with the Japan Storage Battery KE-25 and I wouldn't reccomend it for a rebreather but it has worked for me on a number of cells in O2 analysers.
(Offline)

28th January 2008, 14:06	#20 (permalink)
nigelh New Member Current Rebreather/s: Inspiration Vision Other CCR Other Rebreather/s: Inspiration Classic Other CCR Join Date: Mar 2005 Location: Brighton, Sussex, UK Posts: 784	Re: sensor load and life expectancy Quote: (Originally Posted by WD8CDH) but it has worked for me on a number of cells in O2 analysers. It sounds a horrible thing to do to the chemistry of a cell. Were the ones you tried it on accurate afterwards or did they just give a reading and that was enough for the user? __________________ nigelh
(Online)

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