Choosing between photodiodes and phototransistor designs for a amplification circuit.

[FvM]

The glass filter design is only reasonable related to a differential measurement method suggest by Keith, for UV photodiodes that are sensitive to visible light. Of course they must have a sufficient UV sensitivity, too. UV photo cells, either gas filled (UVtron) or vacuum devices can be (and are usually) manufactured with solar blind photo cathodes, so they don't need optical filters. I assume there are other manufacturers besides Hamamatsu, but I haven't worked on this field for quite a long time.

 

[JPPoulin]

The glass filter design is only reasonable related to a differential measurement method suggest by Keith, for UV photodiodes that are sensitive to visible light. Of course they must have a sufficient UV sensitivity, too. UV photo cells, either gas filled (UVtron) or vacuum devices can be (and are usually) manufactured with solar blind photo cathodes, so they don't need optical filters. I assume there are other manufacturers besides Hamamatsu, but I haven't worked on this field for quite a long time.

Excellent FvM, thanks again for your valuable input. Shielding one of two identical UV photocell with glass (amplified with the same curve) is indeed an intersting trick that we'll evaluate... I wonder if some inexpensive part exists with low-cost UV shielding just for this differential purpose...

In particular you mentioned earlier UV photocells were the cheapest, but I cannot find them on Mouser / Digikey... do they exist by another keyword?

Thanks again!

 

[keith1200rs]

Something like the Vishay TEMD5080X01 might do, depending on the exact wavelength of interest. Bear in mind the differential method of "measuring" UV is crude - when I did it I was simply looking for a value between 1 and 10 so not great accuracy.

Keith.

 

[JPPoulin]

Something like the Vishay TEMD5080X01 might do, depending on the exact wavelength of interest. Bear in mind the differential method of "measuring" UV is crude - when I did it I was simply looking for a value between 1 and 10 so not great accuracy.Keith.

Thank you very much for this recommendaiton Keith... interesting part with a large sensitivity area... looking into it!

 

[dick_freebird]

I would think that what matters is the location of the IR source.
If the field of view is narrow and excludes the burner / pan but
includes the region where flame would be, any sensor that will
pick up red-orange wavelengths would do. You could presume
that the cold-burner case will not induce a flame so I do not
see why you care to distinguish a hot burner from hot burner
plus flame - only the flame matters, I think.

 

[JPPoulin]

I would think that what matters is the location of the IR source.
If the field of view is narrow and excludes the burner / pan but
includes the region where flame would be, any sensor that will
pick up red-orange wavelengths would do. You could presume
that the cold-burner case will not induce a flame so I do not
see why you care to distinguish a hot burner from hot burner
plus flame - only the flame matters, I think.

Hi Freebird, thanks for that input. In our case we're calibrating the system to word about 4-7 meters from the stove, and even with the narrowest beam we can get (which we will use) we'll certainly pickup both... Hence the need to detect the spectrum of flame over just heat.

I think thus far PIR + UV + visible light with some good coding can approach the task at hand.