• State TI Thinks Resolved
  • Locked Locked
  • Replies 3 replies
  • Answers 2 answers
  • Subscribers 31 subscribers
  • Views 521 views
  • Users 0 members are here
Support feedback
Options
Options
Related

This thread has been locked.

If you have a related question, please click the "Ask a related question" button in the top right corner. The newly created question will be automatically linked to this question.

Using the DDC112EVM to detect a modulated photodiode current signal

Patrizia Huggenberger
Prodigy 20 points
Other Parts Discussed in Thread: DDC112

I would like to use the dual-channel DDC112EVM to measure two photodiode current signals.

With the photodiodes I measure the absorption of gas in the MIR region of light, where the light source is an LED.

The LED cannot be driven in a continuous mode, so it is modulated at 1 kHz with a 50% duty cycle (switching the LED basically on and off).

How can I get the DDC112 to integrate only over the ac component (signal amplitue during the "ons" of the LED)? Is there a way to trigger the DDC112 externally? I don't think it makes a lot of senese in terms of signal to noise to let the DDC112 randomly integrate over the modulated signal. Any application notes for a similar setup or general tipps how to design a suitable setup are very much appreciated, I do not have a lot of expertise in electronics (I am a chemist).

The photodiodes are very sensitive to biases, especially reverse bias. Are there any precautions I need to take when using the DDC112EVM, or can I just simply connect the photodiode to the input and measure?

  • 0
    TI__Expert 4680 points

    Hi Patrizia,

    If I understand right, you want to integrate only when LED is on. This is very standard operation. The CONV signal sets the integration period. Basically integration happens between two consecutive edges. There are two integrators on every channel. One integrates in half period and the other on the next. Please check the DS for details.

    On the reverse bias, I am not sure I follow. Do you plan to apply reverse bias or just 0V? And when you say sensitive, do you worry about damage or simply change on performance (like leakage, etc...). Anyhow, the input of the DDC112 is virtual ground. There is a small error to that (see spec on DC Bias Voltage (Input VOS)) 2mV worse case under DS conditions. Only if the current is very large it will saturate the input and then start drifting up to 0.5V or so. The traditional way of connecting to the DDC is directly (most of the users do not apply any bias across the PD, as figure 3 on DS) and I never heard any issues with that.

    Regards,
    Edu

  • 0 in reply to Eduardo Bartolome
    Prodigy 20 points

    Dear Eduardo

    Thank you for your reply :-). So when you mean that integration happens between two consecutive edges, does it automatically detect the "on" and "off" states so that one does not need to set the start of the integration time (with the help of an external trigger signal from the LED driver, so that integration starts right when the LED is "on")? Right, there are two integrators. If each of them integrates one half period, this means that one ideally integrates over the whole "on" state whereas the other over the whole "off" state. So in the end we get a smoothed out modulated digital signal, right? Is there a possibility to get the ac component from this, so only the result of the integrator that integrates during the "on" state so to say?

    About the reverse bias: It should be 0V. The photodiode is sensitive to damage. The maximal allowed reverse bias is 0.1 V. With the current optical system, the photodiode current is roughly around 100 nA. So I guess the error of the virtual ground is not a problem here?

  • 0 in reply to Patrizia Huggenberger
    TI__Expert 4680 points

    Hi Patrizia,

    You get continuous samples, one from one integrator, then the other, then the first again, and so on. So, it is up to you to discard the samples you don't want. And yes, there is no trigger. The CONV signal is basically the ON/OFF control for the switches that select the integrator, the reset, etc. Please check the DS for detailed info - DEVICE OPERATION - Basic integration cycle (Pg 9)

    On the reverse bias, based on the spec, I think you are fine. Also, I am pretty sure that if it saturates, the voltage actually would fwd bias the diode. Do you have a part number for the sensor?

    Regards,
    Edu