Photodiode and Op amp Question?

[3DSS]

I’m trying to detect a candle flame. Using a photodiode and op amp. But I’m not sure as to what kind of photodiode and op amp I should be using. If any one has any suggestions as to what brand of components I should use I’m open for your input. These need to be through hole so I can mount them on a breadboard.

Thanks

 

[IRstuff]

How far away?
Ambient lighting?

TTFN

 

[3DSS]

For now only 5 to 10 feet away, in the long run maybe 25ft.
I’m not sure what you mean by Ambient lighting.

 

[IRstuff]

It would be really hard to detect a candle in broad daylight as a example.

The more ambient light, the smaller the field of view for the detector and possibly more stringent color filtering. TTFN

 

[jimthompson]

Silicon Photodiodes respond to photons of light from the Blue region (450 nm) to infrared (around 1000 nm max). Each photon stricking the active area results in a very small current. Many photons result in a current that is easier to measure. Most applications use a low noise transimpedance amplifier rather than an op amp, because you are trying to amplify a current and then transform it into a voltage change. Texas Instruments aquired Burr Brown sometime ago and they have a product which combines a large area photodiode and a transimpedance amplifier. The datasheet is at

http://www-s.ti.com/sc/psheets/sbbs001/sbbs001.pdf

.

The problem you may have is that the photodiode will detect light from any wavelength in its range and will not be able to tell a candle from any other light so the candle will have to be brighter. You will probably have to add optics to increase the light from the candle on the photodiode active area to get sufficient current.

 

[IRstuff]

From 25 ft away, a 1/4" wide flame subtends 800 microradians. If your ambient light conditions are too high, you'll need to reduce the field of view of the photodiode to something around the size of the flame, which then poses some serious challenges with respect to alignment. If you relax the field of view, you'll need to significantly reduce the ambient background.

Have you considered using a CMOS camera? With some fast optics, you could image capture and then image process to find the flame. TTFN

 

[3DSS]

Thanks guys this information has been very helpful. Jimthompson I looked at the data sheet and this might work for my project. From what I can tell the sensor is reading around 700 to 800 nm in wavelength. So am I right in thinking that this should see a flame though smoke if there was any? Or would I need to find something in the 900 to 1000 nm wavelength? And if so what type of components would I need?

Irstuff as for the ambient lighting it would be in side a building so I can turn down the lights if needed. And since this is only an experiment the cost of a CMOS camera is out of the question.

 

[IRstuff]

how much smoke? Is it only coming from the candle?

If this is a prelude to a fire detection system, a visible photodiode will probably not work reliably. TTFN

 

[peebee]

It seems to me that some furnaces use optical detection to ensure that they have ignited. You might want to research/contact some furnace manufacturers to see what they use.