I recently used this CO2 sensor:

Figaro TGS4161
www.figarosensor.com

to detect when someone breathes on it.  It worked well!  I was able to 
tweak things to make it quite sensitive.  I did: sensor -> sensor buffer 
opamp -> PIC A/D, and wrote some code to watch the flow of numbers and 
detect a fast-moving trend.

I think they are around $40, the client bought them.

Jesse


Russell McMahon wrote:

> I wish to build a Capnograph.
> Don't know what that is?
> I didn't either until yesterday - even though I wanted one.
> It's an instrument to measure exhaled CO2 content.
> Exhaled CO2 content is typically in the range 2% - 10.%
> Y.CO2.M.V.
> 
> Real ones cost real money.
> Most straight forwards method is measuring absorption at about 4.2 -
> 4.3 nm where CO2 has an absorption peak.
> This usually involves producing light which includes this wavelength
> and then filtering it with a suitable filter. These are commonly
> enough avail;able due to their use in Automotive gas analysers, and,
> not surprisingly, in Capnographs. Such filters are nastily
>  expensive - typically hundreds of dollars. This is a 'discretionary'
> project and while such a cost is bearable it would be much better if a
> cheaper solution was available.
> 
> One possible solution is the use of a diffraction grating or possibly
> a prism.
> 
> Other people may have alternative innovative suggestions for
> addressing the 4.2 nm line approach OR may have alternative methods. I
> want a potentially "breath by breath" response and ideally electronic
> logging so chemical means are unlikely to be useful.
> 
> I can imagine a "colour wheel" approach where CO2 is absorbed on a
> substrate causing eg colour change, read by a sensor and then
> regenerated by eg heating, all on a rotating wheel, might provide a
> viable real time solution. How one achieves such a colour or other
> change I don't yet know.
> 
> Pushing that a little further, exhaled air could be bubbled through a
> solution causing transparency change and light occlusion and the
> solution subsequently regenerated. Much messier than I would like, but
> such "idea starters" may lead to other things.
> 
> FWIW the End Tidal value (the CO2 level at the end of exhalation)  is
> an amazingly sensitive and valuable indicator for many medical
> situations. ET value gives far faster response than Pulse Oximiter
> blood oxygen readings to oxygenation changes (one breath versus a
> minute plus), is a sadly accurate indicator of resuscitation prospects
> and a good indicator of whether resuscitation is working (reading at
> 20 minutes resus gives near certain indication of death/life
> prospects). It's invaluable as an intubation indicator and monitor for
> patients during critical transfers.
> 
> 
> 
> 
>         Russell McMahon
> 
> 
> 
-- 
http://www.piclist.com PIC/SX FAQ & list archive
View/change your membership options at
http://mailman.mit.edu/mailman/listinfo/piclist