I need a simple circuit to give a 0-1 volt output for changing light levels from darkness to bright daylight. The ORP12 I'm using is very sensitive to light (too sensitive to be useful?). I have on-board regulated 2.5 and 5 volt supplies.
(I've converted a chart recorder to read light levels instead of temperature).
Thanks

 

It can't really be 'too sensitive', it's easier (and cheaper) to reduce sensitivity than increase it - a simple resistor or two is all that's required.

However, for your application I would suggest using an opamp to scale your output to the range you require.

Something like this should work, adjust P1 for zero volts on the output with the LDR in total darkness (it should have the same resistance as the LDR). Select the two resistors to give 1V output under bright light conditions - the values also need selecting to give a high enough input impedance, depending on the resistance of the LDR.

The opamp is used in inverting mode so you can set the gain as less then one, it's probably a good idea to replace R1 with a preset and fixed resistor in series, so you can vary the gain. It's probably also a good idea to put a fixed resistor in series with P1, to limit it's range.

Attached Images

ldr_opamp.gif (2.0 KB, 448 views)

__________________
PIC programmer software, and PIC Tutorials at:
http://www.winpicprog.co.uk

 

Thanks Nigel. Circuit looks exactly what I want. Is the output linear and is the opamp a 741?I don't have -ve rails to use, only ground and either +2.5V 0r +5V both regulated. Can I use a single rail supply in this circuit?
I think the CA3130 and CA3140 are useful in this case - am I right?

 

Any opamp should be fine, but you really need a split supply, you could use a 7660 to generate a -5V supply, as I do in my tutorial analogue board at http://www.winpicprog.co.uk/pic_tuto...ogue_board.htm.

The LDR is used in a simple potential divider circuit, I can't say I'm sure if that would be linear or not?, but an LDR isn't very linear itself either.

Here's a quick plot of the output voltage from a potential divider, does this help at all?.

__________________
PIC programmer software, and PIC Tutorials at:
http://www.winpicprog.co.uk

 

The output graph is fine - I guess I can't get more of a linear response than that. I'm not sure about the 7660 to generate a -5V rail though. I've never used one of these before, but do have one from a while back. Why is a -ve rail needed? What would happen if I connected op amp pins 3 and 4 together at the 0v rail? I checked out your tutorial page to see how it was being used. Any chance of a schematic with a 7660 integrated into the first circuit you sent? Thanks again.

 

I seem to have a vague recollection that the non-linear output of an ORP12 actually is the reverse of the potential divider graph - so they cancel out and end up more linear. I wouldn't swear to this, but I seem to remember it from somewhere?.

You need the negative supply because the input needs to be at 0V to give 0V output, a potential divider can't provide that from a single supply rail. The way I drew the circuit also means the input goes more negative as the light gets brighter (because the opamp is in inverting mode).

There's nothing much to draw - you just need IC1, C1 and C3 from the link I posted, that section has 0V and +5V in, and a -5V output - simply connect the -5V output to the -ve supply pin of the opamp (and the bottom of the LDR).

For initially testing the design you could feed the opamp from a couple of 9V batteries, this would prove that it works OK.

__________________
PIC programmer software, and PIC Tutorials at:
http://www.winpicprog.co.uk

 

Photodiodes should give you a linear response. There are tons of photodoide amplifier schematics on the web.

A trick to get dual supply op amps to work on a single supply is to use half of your supply voltage as the ground for the op amp. Dual supply opamps won't operate if the input voltages are to close to ground or supply voltage or in inverting mode(they can't output a negative voltage). Say you are using a 5V supply you create a 2.5V voltage using resistors and or Zeners and use that as the ground signal in the dual supply schematic (there are specialized ICs that do this called rail-splitters that also provide a buffer on the output voltage). The real ground becomes V- and the supply voltage becomes V+. The output will swing around 2.5V instead of 0V.

 

Have you considered photovoltaic (solar) cells, as opposed to photoresistive cells? They usually put out about 1/2 vdc per cell. Just put several in series and use a pot to calibrate the output in bright sunlight. (I used a battery symbol to represent the PV cells in the image). I don't have much experience with PV cells, but I believe they have a fairly linear output (Vout vs. Illumination). (Someone who has more experience with them, please correct me if necessary...) JB

Attached Images

pv_cell.gif (1.4 KB, 381 views)

 

Solar cells are just big photodiodes. Photodiodes are cheaper and faster. For them to be linear you need to use them in the photoconductive mode -as a current source rather than a voltage source.

 

Here's a single-supply way to do it with the ORP12. You will need an op amp whose output can reach the negative rail (GND).

Attached Images

orp12_amp1.gif (5.9 KB, 359 views)

 

Thanks for the schematic. I think U1 - U3 could be CA3130, which if I recall can go to 0v. I'll wire it up on breadboard.