ionisiation?

From the input pin to +5V.

ok thanks will have to have a play with that when i get time

hi again
i have made bit of a rig up and ran your theory above and i couldnt seem to get it to work at all. i think there is a bit more to it than i originally thought ?

A flame consists of ionized gas, if two conductors are placed in it then current will flow.
Many years ago I put an audio signal through a flame simply by placing it in series with the speaker connection.

If you are not getting anything from it try and measure the resistance of the probe and sheath, it might me made of some resistive materiel so that it works on a higher voltage.

There will also be issues with the conductivity of the probe and flame, and your 1K resistor might be a bit much, if the probe's resistance is high enough. Measure the resistance, and use Ohm's law to find the power drawn. You want to make sure the power is below the max for the arduino, but high enough to register. If the voltage is dropping too much to register, or the current is too high and tripping protection or something, it wouldn't work.

Quick google suggests 100Mohms is a good starting point... A flame is not that good a conductor!

Also found this fascinating page: Flame Triode With Gain.

hi there
i have been playing with different resistances. im up in the 2000k to get it to work on my test rig on my finished product i might be able to get it lower than that but i was wandering is it ok to use these sorts of resistances?

Well it won't do any harm but it might be a bit suceptable to noise.
You can try a higher voltage with a 5.1V zenner to protect the input.

what switch the live instead of the 0v?

Pardon?

how would i wire it for high voltage?
diagram

Just put a high voltage on one terminal, with a 1K resistor in seriese with the input. Then wire a zenner between input and ground. The other end of the sensor goes to ground.

No like this.

probe.pdf (19.9 KB)

ok thanks
what do we call high voltage?

what do we call high voltage?

Dangerous.

Anything higher than 5V.
The higher the voltage the more current the flame will conduct.
Making the resistor higher will make any change in current a larger change in voltage and the easier it is to detect.

sorry to bring this back up again i had got this working but its extremely unstable as there is not much movement on the arduino, i was using 1500k resistor to even get it to work i cant seem to add any more than that or it seems to go the other way so i dont know how else we could make this work more reliably and give us more of a range on the arduino
i measured the resistance of the probe and flame was 1400k and my supply voltage i was running 17.11 and im getting back on the other end 1.65V so hell of a drop?

You need to use an op-amp to make the signal from the flame bigger.
Start of with a gain of about 100 and see how you get on.

ok thanks for that i have had a bit of a play and come up with this
http://s906.photobucket.com/albums/ac270/Joes-110/?action=view&current=Untitled.png
but now we have gone to the other way now it is to sensitive and can not seem to dye down any idea?

Three things wrong with that circuit.

  1. It only has a gain of 2, see how to calculate the gain of a non inverting op amp here:- Operational amplifier - Wikipedia
  2. You in effect have a floating input, that is the input is very high impedance and is picking up all sorts of interference.
  3. That sort of circuit will only amplify voltages, you are in effect putting a variable resistance on the input you are not inputting a voltage.

Try this. +ve input of amp connected to two 1M resistors, one goes to +5 and the other ground. Then feed this point and ground at your flame.
Change the resistor that goes from the -ve input to ground to a 10K. What sort of op amp are you using.