ChAndrea
Newbie level 4
Hi everybody,
I am an engineering student working on a relative simple project: I need to measure the current generated in a photodiode (several of them actually but does not matter) stimulated by a LED.
The application requires a low bandwidth (no more than a few Hz) but high accuracy and resolution. In particular I am interested in monitoring the "DC" component of the signal over time.
I choose to work with a simple transimpedance amplifier (DC gain 1 Mohm, -3 dB bandwidth between 1 and 2 Hz), followed by an antialiasing filter (a Sallen Key filter). The ADC I am using is a delta sigma type (ADS1115, 16 bit resolution, conversion speed between 8 and 860 SPS).
To drive the LEDs I am using a PWM led driver (TLC59711) or a simple LM317 configured as a current regulator.
The signal I obtain is "clean" (no 50-60 Hz, really low noise and everything). The problem is that there is a drift over time in the measured signal ( a few mV) in response to a (ideally) constant LED brightness (constant current or duty cycle).
This phenomenon is evident when I change the LED current or duty cycle (usually when I increase the current or the duty cycle the signal drifts toward a lower value and viceversa).
The circuit is built on a breadboard and both the LED and the photodiode are secured inside an enclosure to remove the ambient light. The wires used to connect the LED and the photodiode are 30 cm long (no shield).
The nominal current value is 5 mA (the LED is specified for a maximum of 20 mA).
I tested the circuit without the photodiode sensor, just amplifying a fixed voltage: no drift detected. I assumed the problem is not related with the Op Amp used for the TIA, the Sallen Key and the ADC.
I think the problem is related to the LED (or the LED driver, or the system response to the LED driver) and the photodiode. I know some stuff about flicker noise, but I don't think this the case. LED brightness goes down with heating (and so grows up with cooling), but seems too much to me. Maybe some non idealities caused by the long wires. I tested the circuit with a LED previously used in a medical device (decent quality), same result. The current which flows inside the LED, once set, is constant (I used a sense resistor to check this).
At the moment I am stuck: I don't know what to try to remove this "drift" or atleast I'd like to know what is the cause of this behaviour.
Right now I can't upload a schematic, my internet is really bad. I will upload one asap. The op amp used is the LTC1053, a quad channel auto-zero amp with 2.5 MHz GBW.
Thank you for your precious help,
Andrea
I am an engineering student working on a relative simple project: I need to measure the current generated in a photodiode (several of them actually but does not matter) stimulated by a LED.
The application requires a low bandwidth (no more than a few Hz) but high accuracy and resolution. In particular I am interested in monitoring the "DC" component of the signal over time.
I choose to work with a simple transimpedance amplifier (DC gain 1 Mohm, -3 dB bandwidth between 1 and 2 Hz), followed by an antialiasing filter (a Sallen Key filter). The ADC I am using is a delta sigma type (ADS1115, 16 bit resolution, conversion speed between 8 and 860 SPS).
To drive the LEDs I am using a PWM led driver (TLC59711) or a simple LM317 configured as a current regulator.
The signal I obtain is "clean" (no 50-60 Hz, really low noise and everything). The problem is that there is a drift over time in the measured signal ( a few mV) in response to a (ideally) constant LED brightness (constant current or duty cycle).
This phenomenon is evident when I change the LED current or duty cycle (usually when I increase the current or the duty cycle the signal drifts toward a lower value and viceversa).
The circuit is built on a breadboard and both the LED and the photodiode are secured inside an enclosure to remove the ambient light. The wires used to connect the LED and the photodiode are 30 cm long (no shield).
The nominal current value is 5 mA (the LED is specified for a maximum of 20 mA).
I tested the circuit without the photodiode sensor, just amplifying a fixed voltage: no drift detected. I assumed the problem is not related with the Op Amp used for the TIA, the Sallen Key and the ADC.
I think the problem is related to the LED (or the LED driver, or the system response to the LED driver) and the photodiode. I know some stuff about flicker noise, but I don't think this the case. LED brightness goes down with heating (and so grows up with cooling), but seems too much to me. Maybe some non idealities caused by the long wires. I tested the circuit with a LED previously used in a medical device (decent quality), same result. The current which flows inside the LED, once set, is constant (I used a sense resistor to check this).
At the moment I am stuck: I don't know what to try to remove this "drift" or atleast I'd like to know what is the cause of this behaviour.
Right now I can't upload a schematic, my internet is really bad. I will upload one asap. The op amp used is the LTC1053, a quad channel auto-zero amp with 2.5 MHz GBW.
Thank you for your precious help,
Andrea