• Descent 305 - Pre-order is open!!

    Our August/September issue has a publication date of 2 August, and it is a special summer bumper issue full of stories of exploration and wonder from the underground world.

    Click here for details of this edition

Remote sensing for air movement

cap n chris

cap n chris

Well-known member
Is there in existence (or can someone make one!) a device which can be left underground for short periods (i.e. doesn't have to be super-cave-proof) which can determine, with absolute certainty, the presence of a non-caver-induced draught?

I was thinking of some kind of Rube Goldberg joss-stick burner with a film camera and light but clearly that's nuts. Seaweed on a string?

Ideas?
 
There are certainly such devices. There used to be an anemometer in the entrance passage to Reed's Cave in Devon as part of an experiment. You just need a very sensitive one. I should try a web search on the subject. I have a theory that bat's could be indicators of locations with good airflow. They seem to group at passage junctins in Reservoir Hole. I hypothesize that this enables them to detect minor temperature fluctuations outside the cave so that in mild weather they can emerge and forage (lesser horseshoes are certainly active in mild conditions as we are seeing this year).
 
Thanks Rob - I can't do anything, myself, with your suggestion though since wiring a plug is pretty much my limit nowadays.
 
I am tempted to find an image of a bat drinking a cup of tea but this is the best I can do  :coffee:
 
Rob's suggestion is certainly an interesting proposal and is (if you can wield a soldering iron) not too complex a project.  But I note it cannot record the data, only provide a signal.  Any suggestions for a cheap stand alone data recorder which can be directly interrogated by a computer at intervals of say 7 or more days?  I guess one is looking at capturing data at a frequency of say every second, so that would be under one million samples worth of data.  Which in turn would require some software to portray the data.  (Excel is limited to only displaying 32,000 samples / 9 hours in a graph.)
 
Well there is this http://interestingto.me.uk/Open%20Source/low-cost-low-power-datalogger which I am just ironing a few wrinkles out of but it is nearly there.

There does seem to be a bit of a need for something like this, however it is not the easiest thing to solder (esp. the SD card holder where the pins are underneath) so if a few people were interested it could be worth getting a batch made up. Everything except the SD card holder could be through-hole rather than SMD if people wanted that.

Would you really want to sample it every second? I can't see that a draft would vary that quickly...
 
estelle said:
something along those lines could be very useful in one of our digs...
Click to expand...
We too could benefit from something like this in some of our digs, especially if it records air temperature as well (not sure if that Wind Sensor does).
 
TheBitterEnd said:
Well there is this http://interestingto.me.uk/Open%20Source/low-cost-low-power-datalogger which I am just ironing a few wrinkles out of but it is nearly there.

There does seem to be a bit of a need for something like this, however it is not the easiest thing to solder (esp. the SD card holder where the pins are underneath) so if a few people were interested it could be worth getting a batch made up. Everything except the SD card holder could be through-hole rather than SMD if people wanted that.

Would you really want to sample it every second? I can't see that a draft would vary that quickly...
Click to expand...

Looks very good, but as I am no techie, so some clarification please.  As I understand your project, the data is pushed to an SD card, so there is no problem about holding a quantity of data.

The speed of the  ATMEGA168-20AU suggests it can cope with much faster data acquisition times than my suggested 1 sample per second.  Any ideas on top speed?  Whilst I have no idea how the air movements could vary, I guess the sensor will have a fair sized thermal lag with respect to 1s.  So perhaps one can sample at a slower rate.  But other potential uses might desire faster rates; our work on rope goes up to10us per sample for example. 

I understand it will take both analog signals and digital data.  I presume the analogue voltage supplies have to be limited.  8 bit operation means data precision is around 0.4% which should be more than enough for most applications. 

I guess there will have to be a fair amount of instructions if it is to be used by non programmers.

Bottom line is what cost?

In reply to Rob, one could use a second detector which is shielded to obtain temperature or use http://moderndevice.com/product/sht21-humidity-and-temp-sensor/
 
I have thought of using my Drager CO2 meter to try to detect if a cave is breathing.
The meter can be set to log the CO2 level at intervals (say every 10 or 15 minutes) over weeks or even months.
If the cave is breathing then there may be a regular increase and decrease of CO2, particularly in the summer when plant roots are producing the gas.
Not quite what Chris was asking for but it may be worth a try?

 
Aubrey said:
I have thought of using my Drager CO2 meter to try to detect if a cave is breathing.
The meter can be set to log the CO2 level at intervals (say every 10 or 15 minutes) over weeks or even months.
If the cave is breathing then there may be a regular increase and decrease of CO2, particularly in the summer when plant roots are producing the gas.
Click to expand...
CO2 meter could well give useful answers in the dig i'm thinking of as we get driven out from digging after a couple of hours or so down to running out of air in there - but it must be 'breathing' since it's back to ok by the start of the next session!
 
all very interesting. i have been trying to make a fluorescein data logger and the logger bit is getting there (much faster than the sensor) it is smiler to thebitterends suggestion but using off the shelf modules that are easy to work with.
eg.
https://www.sparkfun.com/products/9802  
http://www.ebay.co.uk/itm/I2C-Tiny-RTC-DS1307-Real-Time-Clock-Module-AT24C32-Board-for-Arduino-AVR-MCU-PIC-/121079759380?pt=UK_BOI_Electrical_Components_Supplies_ET&hash=item1c30ea8214
etc

though one problem with the ATmega328 chip is it only has a 12bit analog to digital but using this addition i have got my one to log at more like 15 bit. http://shop.pimoroni.com/products/adafruit-ads1115-16-bit-adc-4-channel-with-programmable-gain-amplifier?utm_source=googlepla&utm_medium=cpc&gclid=CLqL4qKglLwCFU_KtAodCF8AIg#.UuEE4NdFDGJ

i cant get the 16bit to sit in the usable range.

any way to cut to the chase the logger i have would cope fine with logging robs suggestion of wind meter and a temperature if it were wanted.

but at the moment i have not got it running on batterys so its not as usfull as it sounds.  but as the progect progresses i think there is a place for a cave logger with meany different ends that can be changed at a reasonable price.   
 
yes bob these things can log alot more than once a second if i take all the delays out i can log about 100 reading a second.  i worked out that taking a reading every second it would still take about 18 months to fill a 4gig card so given that you can get 32 gig in the simple sd format data is not the problem power is.

even if you shut down all the not used bits of the chip and power down between readings i am struggling to get the average currant below about 10-15mA at 5 volts so batterys are more of a problem.

if any one can come up with a very efficient 5volt regulate that would help.  it need to be very good at low currants but capable of higher draws some times.



 
 
We use LP2954 to regulate six AA batteries to 5V but that can only cope with up to 250mA at 1.2% tolerance.  We have also used LM78M05 which can give up to 1A but only at 2%.  HTH
 
I hope Nigel is reading this cos I think we have the means to knock one up with the information that's been dug up.
 
You must log in or register to reply here.
Back
Top