All photodiode sensor amplifier circuits I have gathered use current-to-voltage or transconductance amplifiers, leading to my conclusion that current output from the photodiode must be proportional to the light input.
Since the UV sensor OS100 (Note that I do not call it UVB any more as I have just found out from its spec that it covers all UVA, B&C) outputs only tens of nA currents, I have been looking for Op Amps with input bias currents less than 1 nA. TL081 is one (so is AD795) though the required supply voltage level (7 volt) is higher than what I would like to have.
As for many other low supply voltage devices such as LM358 the input bias currents can be tens or even hundreds of nA, a level comparable to the current from the photodiode it self! I hesitate to choose any of them as I have no confidence in developing a reliable circuit around any of them.
May be I should still try? Any circuit technique to overcome the "large" input bias ?
I have not thought of this approach. A matched FET/MOSFET as differential inputs ? Would there be input offset problem then? Likely a balance adjustment pot is required ...... All need to be done at a supply voltage of 3 volt or below ......
I would try it out if I fail evetually to get an OP Amp working at this low supply voltage AND with input bias current less than 1 nA.
Whould be a nice touch to have 2 matched FET's in a single capsule. Those types of dev were used until mid 80's (I think) quite a lot. I haven't seen any new ofer. I used some pairs scavanged from old stuff and where stil OK (exept those i fried :lol).
I do not really mind inventing old stuff, when there is the need and when I have spare time.
Thanks all for offering the advice and help. As I said I cannot wait and I have used the TL081 to build a simple circuit for UV meter. Please refer to the circuit diagram.
Using a 9 volt battery the TL081 Op Amp is configured as a currrent to voltage converter and the output is displayed by a multimeter at 100 mVDC range.
It works just as expected. By comparing to real time display of local UV Index by our Obvervatory (local meterological office) I have been able to calibrate the meter, to an accuracy of no less than 20%, I expect.
With this meter it is interesting to find out that ordinary lamps give out little UV while window glass stops almost 90% of UV lights from the sun .......
Further improvements possible ?
Any comments ?
How about using 3 volt battery ?
How about lowering the supply current ? (now 2.5 mA)
How about ........?
Hong Kong, being a free port, does have a lot of trading companies accessing all parts of the world for "a lot of things". However that is only for trading, not for us hobbyists.
So when I say available locally, I mean available in some of the retail shops within just a few miles around the city.
Using a voltage converter to boot up from 3 volt is a good idea. However I prefer simple circuits --- so I am keen to find any op amps working at the low voltage level and yet have input bias current below 1 nA. If there is none in the market (but why none, I just wonder ?) I will need to build one my self ......
Well I've bought like 25pcs of some chips from nonretail companies....I ordered them with 3 friends (hobyst also).
About low voltage: https://www.electro-tech-online.com/custompdfs/2006/10/tlc25l4.pdf
Realy like those opamps (a bit exp for my taste thought).
U could use a Charge-Pump Voltage Converter like MAX619 (doesn't use a coil) and it's like 3.5 $ at retail. At 5V u'l found more suitable opamps.
Actually test it by using a reference mter, but your lizards get UV light from the sun, correct... so I say calibrate using the sun to be biologically accurate, is not Industrially correct but your lizzards did not evolve in a lab
redserpent1
Well I've bought like 25pcs of some chips from nonretail companies....I ordered them with 3 friends (hobyst also).
About low voltage: https://www.electro-tech-online.com/custompdfs/2006/10/tlc25l4-1.pdf
Realy like those opamps (a bit exp for my taste thought).
U could use a Charge-Pump Voltage Converter like MAX619 (doesn't use a coil) and it's like 3.5 $ at retail. At 5V u'l found more suitable opamps.
Thanks a lot Tarsil. This is exactly what I have been looking for in the last two weeks: TLC254 --- working at 1.4 volt up with input bias current less than 0.06 nA !
Thanks a lot Tarsil. This is exactly what I have been looking for in the last two weeks: TLC254 --- working at 1.4 volt up with input bias current less than 0.06 nA !