Luxmeter, transimpedance photodiode amplifier?

[Coper]

Luxmeter, transimpedance photodiode amplifier?
I just need a sensor to automatically turn on the lights when the sunlight level is set.
Speaking of which, I want it to look a bit and be able to measure the intensity of light
I will describe the initial conditions .

1. The control electronics will be inside and so the light sensor itself will be connected by a cable up to 5m long.
2. Taking into account point 1, I will probably use an integrated sensor, but a classic photodiode, probably BPW21R.
3. The MCU has a 16-bit ADC with PGA ADS1115
4. Supply voltage only + 5V

Under the circumstances, it seems reasonable to me. Place an transimpedance photodiode amplifier on the BPW21,
Set so that the output voltage is 2.048V in full sunlight (10000 Lux).Maybe LT1880 nebo AD8622.
PGA is then used as a second amplifier at lower light levels.


What do you think about it?

 

[barry]

I’m pretty suspicious of putting a photodiode at the end of a 5M cable. It might work. Or not.

 

[betwixt]

I perform the same task in a much simpler way: I use a small PV panel to generate the voltage directly.

Brian.

 

[carpenter]

If you place only a photodiode on the end of the cable, it will not work
Why?
BWP21 has a current of about 100 nA at 10 lux.
In the long run, there will be more leakage current on the 5m cable.
I understand that, the author of the query wants to place the diode and transimpedance photodiode amplifier at the end of the cable.

 

[Warpspeed]

Agree with Brian, you can use a very small single solar cell.

If you want really cheap and small, as a basic day/night sensor, try measuring the voltage developed across an ordinary green LED in direct sunlight.
Expect to see about 1.7 volts in a clear blue sky.

Several LEDs in series hooked up to a mosfet gate, then run the source/drain down the cable. That should take care of any cable leakage problem.

A sixteen bit ADC ?
Do you really need that to turn a light on and off ?

 

[KlausST]

Hi,

You don't need accuracy, nir precision.
So a simple comparator circuit may work.

Klaus

 

[carpenter]

Agree with Brian, you can use a very small single solar cell.

Note that the PV cell does not necessarily capture the visible part of the spectrum
Whether switching the light, measuring the intensity of light in lux concerns the visible spectrum from a human point of view.
One picture from a1000 words. What the eye sees and what PV sees

Some remarks on the issue of measuring light intensity.
If we want to measure the intensity of sunlight for teaching comparable to, for example, meteorology, we should use real, relatively expensive, large-area photodiodes. For example, the BPW21 mentioned above or other larger and significantly longer, whether for measurements in Lux or energy in W / m2. To them is then proposed a low-noise transimpedance amplifier, usually with range switching, etc.
It is not a problem to design and implement such a thing, but it requires a different arrangement than the one Coper talks about, and it requires different design priorities.

The second option is to use an integrated sensor that has a photodiode, amplifier, PGA and ADC.
I can think of Vishay VEML7700 High Accuracy Ambient Light Sensor With I 2C Interface with 16-bit dynamic range for a mbient light detection from 0 lx to about 120 klx with resolution down to 0.0036 lx/ct. or BH1750FVI.
.The advantage is a small integrated design with a large dynamic range and resolution. The disadvantages are the small sensing diode areas, the difficult-to-define irradiation area and the not entirely clear longevity and accuracy. From Cooper's point of view it's about the I2C bus and the 5m cable problem, but it can be solved.
There is a driver for the long range I2C P82B715 I2C can then have a 30m and have a well-designed protection and ESD.

If I want it purely to switch the lighting can use a photo resistor.

 

[Warpspeed]

I just need a sensor to automatically turn on the lights when the sunlight level is set.

All the original poster wants to do is switch a bloody light on and off when sunlight is below/above a set level.

We don't need a 16 bit A/D converter.
We don't need a powerful microprocessor and complex software.
We don't need a full spectral analysis of the suns emission.

Almost anything that will detect light should be well up to the job.

 

[barry]

All the original poster wants to do is switch a bloody light on and off when sunlight is below/above a set level.

We don't need a 16 bit A/D converter.
We don't need a powerful microprocessor and complex software.
We don't need a full spectral analysis of the suns emission.

Almost anything that will detect light should be well up to the job.

Well said. Brevity is the soul of wit and simpicity is the soul of engineering. People often lose sight of that.