Photovoltage measurement with ADC

Thread Starter

MarcelP

Joined Aug 8, 2021
5
Hello !
I want to measure photovoltage of LED (as sensor) in open circuit condition with different light intensity.
This is easy with a voltmeter, even with low cost instrument, but with output connected to Arduino ADC input first of all an unstable offset is added .

I try to insert a buffer (JFET 2N3819 or with TLO71 OP) but the result is not so good.

I try also with an external ADC (Delta_Sigma) without succes.

May be the output resistance of LED in open circuit is high ( but it is surprising for my knowlege) and leakage current of input DAC are involved for this additional bias ?

Thank you if you have an idea .
MarcelP
 

Yaakov

Joined Jan 27, 2019
3,602
hi Marcel,
Welcome to AAC.
Do you have a model number/type for the opto sensor you are testing.?
E
He's using an LED, Eric. I suspect the input impedance of the DAC is mismatched but I have no idea what the output impedance of an LED is.
 

Yaakov

Joined Jan 27, 2019
3,602
I read an article where some one was getting 1V at 30uA from a LED. I could only get 0.3V at 1uA.
Please post a schematic.
Try a different type of LED. They were never designed to do this.
Just measured today using a green LED and UV light I got those same numbers (as you).
 

Thread Starter

MarcelP

Joined Aug 8, 2021
5
Hi , thank you all
I use an commercial UV LED , photovoltage is about 2.7V at daylight (it is an high band gap semiconductor)

I do not want to measure short circuit photocurrent for this application.

The variation of photovltage with light is logarithmic. In low level light the value mesured with an voltmeter is about 30 mV.

I perform some LTspice simulations with SiC model LED without no problem.

I attach simulation file (Voc variation versus photocurrent intensity)

photocurrent is proportionnal to the light....

Marcel
 

Attachments

KeithWalker

Joined Jul 10, 2017
1,937
Hello !
I want to measure photovoltage of LED (as sensor) in open circuit condition with different light intensity.
This is easy with a voltmeter, even with low cost instrument, but with output connected to Arduino ADC input first of all an unstable offset is added .

I try to insert a buffer (JFET 2N3819 or with TLO71 OP) but the result is not so good.

I try also with an external ADC (Delta_Sigma) without succes.

May be the output resistance of LED in open circuit is high ( but it is surprising for my knowlege) and leakage current of input DAC are involved for this additional bias ?

Thank you if you have an idea .
MarcelP
The input impedance of an Arduino ADC is 100 Megohms which is a virtual open circuit. If you want to get results similar to those obtained using a multimeter, Paralellel the ADC input with a resistor equal to the input impedance of the meter.
 

nsaspook

Joined Aug 27, 2009
8,781
I've done some experiments (Moon’s measured luminance during an eclipse) with solar panels in low light condition. A good external ADC (Delta_Sigma) with PGA was all I needed as the front-end but the signal did need processing to remove noise from several sources. It's really important for the P/N junction signal not to be ground referenced if it's remote from the actual measuring device so you need a good differential input to the detector to reduce common-mode signals.

https://forum.allaboutcircuits.com/threads/super-moon-shine.100322/post-896006
I'm planning to use a TI ADS1220 Low-Power, Low-Noise, 24-Bit, ADC (+- 2.5 for the analog power and 3.3 for the digital power from batteries) for the front-end with a PRi2 custom kernel SPI driver to process the signal data. It should be a fun project to see how much raw luminosity data I can capture on a good night.
 
Last edited:

Thread Starter

MarcelP

Joined Aug 8, 2021
5
hi M,
Have you considered using a Hi input Z with low leakage OPA as a buffer between the LED and the ADC module.?

E
Hi Eric,
I have only tried with an OP90 , it is a bipolar input OPA. Now our lab is close but in september I will try with your suggestion.
Thanks


The input impedance of an Arduino ADC is 100 Megohms which is a virtual open circuit. If you want to get results similar to those obtained using a multimeter, Paralellel the ADC input with a resistor equal to the input impedance of the meter.
hi M,
Have you considered using a Hi input Z with low leakage OPA as a buffer between the LED and the ADC module.?

E
 

Thread Starter

MarcelP

Joined Aug 8, 2021
5
The input impedance of an Arduino ADC is 100 Megohms which is a virtual open circuit. If you want to get results similar to those obtained using a multimeter, Paralellel the ADC input with a resistor equal to the input impedance of the meter.
Hi Keith,
I have tried with a 6.8 Megohms in parallel, results are not so good...
Marcel
 

Thread Starter

MarcelP

Joined Aug 8, 2021
5
I've done some experiments (Moon’s measured luminance during an eclipse) with solar panels in low light condition. A good external ADC (Delta_Sigma) with PGA was all I needed as the front-end but the signal did need processing to remove noise from several sources. It's really important for the P/N junction signal not to be ground referenced if it's remote from the actual measuring device so you need a good differential input to the detector to reduce common-mode signals.

https://forum.allaboutcircuits.com/threads/super-moon-shine.100322/post-896006
Hi
Thanks for your reply
I have seen your interesting project and he give me lot of directions to explore. I'll try this in september.
thanks a lot
Marcel
 
Top