Please go to update pages for the latest modification, class-wise issues, and others if links for those pages exist.
update page a  update page b   update page c

Lab work modification: - Do Part 0 ( Introduction), Part 1 (Instrumentation), and Part A only (three parts total): do everything, including answering/discussion all questions. - Experimental work of Part B and C are not required: It means no circuit, no measurement needs be done. You are required to only observe and write a report of your understanding, thoughts, and summary of these Parts. This should be put at the end of your report, a synthesis of everything that you learn in Lab 4. Create a subsection, titled "Report on observation of Parts B and C demonstrations". Do not mix this with what you summarize for your work in Part A. However, if you do anywork along the line of Part B and Part C, you are welcome to include for extra credit.

back to page 2



Lab work (continued):

Circuit A

Before you start:

    - handle the photodiode with a pair of needle-nose pliers or tweezers, grabbing its legs or the side. Avoid using your bare fingers as much as possible, and do not touch the top surface (light sensitive region), especially the small wire (~250-um diameter) that connects the top layer to one of the electrodes.

    - identify the cathode, which has a marker on the leg, a notch, or a dark line on the light-sensitive surface. Look at the spec sheet for details, also this figure.

    - critical: apply POSITIVE bias voltage to the cathode; this is known as REVERSE bias, which is OPPOSITE to forward bias in the case of LED. Please do not ever forward bias your photodiode.

    - photodiode is a delicate device. Think of it like your eyes. Some of the maximum ratings from the spec sheet are shown below:

•  For forward bias (if accidental)
  

    At a mere 1.3 V forward bias, the device already draws 100 mA, which means 130 mW. As one can see in the below, the maximum power dissipation is 150 mW. If you accidentally connect the 15 V power supply to its anode, it would likely draw several hundred mA for a fraction of a second, heated up quickly, and will be subsequently degraded or damaged.

•   For reverse bias and maximum power dissipation
  

Since we use only +15 V max, we don't have to be concerned with the 32-V limit. The light level we apply will be in the range of uW or << 1 mW, hence, the photocurrent will be << 1 mA and we need not be concerned with the 150-mW power limit. However, as discussed above, forward bias will result in ~ or > 150 mW, which will cause degradation or damage to the device.
We do have a few spares and you can ask for another one, but it is best if you can avoid this mistake from the beginning.

    -if you are curious, read the below if you wish to know the most relevant reason why we must use in reverse bias voltage (look under the topic of dark-current shot noise).

•   introduction to photodetectors, part 1

•   introduction to photodetectors, part 2

•   photodetector dark current and shot noise concept

Step 7 -  Build Circuit A (see page 1)
 
Make sure you read through the above.


Do not turn on your power supply until you finish and double check on the wiring.

7.1  Choose the feedback resistor R0 = 10 kOhm to start. Later, if you see Circuit B output saturated at 14-dB gain (x27) with moderate light level, you can select a lower value, such as 2.7 kOhm.

7.2 Connect +15 V source (in off-state) to the photodiode cathode (very important - do not make mistake of its polarity, this is called reverse bias). The anode is connected to the op-amp inverting input. Note: in all photodiodes, the cathode side always has a marker, which can be a tag (a little ear), a spur or a notch, depending on the package. It also comes in with a shorter leg than the anode, however, just in case the legs of a photodiode might have been modified for soldering or whatever, identify the cathode by the marker.

Below is an illustration, your photodiode may or may not look identical to this, but the method of marking of the cathode is the same.

    
notice the notch on the right side? That's the cathode, where +15 V  goes to. (you can also tell by the wire contact at top. The majority of commercial silicon PIN PD are grown on n+ substrate, which means p-layer (or p-type or anode) is on top and hence you might see the ohmic contact wire on the anode side - be very careful not to touch it, the wire is delicate in an exposed package. In the future, we will purchase PD in a TO can with protective cover, but for now, please be careful).

7.3 Put a light-shield enclosure over the photodiode as instructed previously.

7.4 Connect the output of Circuit A to the input of Circuit B like the above (you can choose the Rfb of Circuit B to be 100 kOhm for maximum gain to start, later you can change to any other Rfb to have your signal amplitude in the ~0.1-few V range).

7.5 Turn on the power supply, look at output vout2 on the oscilloscope, you may see the high-frequency modulation of the room fluorescent light that leaks through the light shield. Move your hand back and forth over the PD enclosure aperture. You should see signal responses.

Step 8 - Test light detection with a modulated LED

8.1  Choose a LED, if you don't have one, ask the TA. Set up a square wave signal from the function generator as shown below. The high-level voltage for each color LED is only an approximate, there is no need to be more precise than +-0.1 V. Do not connect to any LED yet until you are sure of the signal on the oscilloscope (if you are concerned, have a 100-Ohm resistor in series with the LED first, then remove later if not needed). Set the frequency between 2-10 Hz.

8.2 Connect the square wave signal to the LED as shown, you should see the LED blinking. If you have a 100-Ohm resistor in series, you will need to raise the high-level as shown by one extra volt, which is the voltage across the 100-Ohm at 10 mA. In fact, you can clip a probe across the 100-Ohm resistor, plug into the scope and you should see ~ 1-V signal. You can adjust the high-level such that you have ~ 1 V across the resistor, and you can be sure that the LED current is ~ 10 mA. There is nothing magic about 10 mA. It is just a safe and more-than-sufficient level to generate plenty of light to test your sensor.  It can be very sensitive; and you will be surprised that it can see signal at 1 mA or less when our eyes cannot tell.
Once the LED is blinking, you can change the frequency to any where between 1-10 kHz.

8.3 Determine your eye bandwidth by increasing the frequency slowly until you can hardly tell if the LED is continuous or blinking. Do you think TV or display refreshing rate at 30 Hz or 60 Hz is good enough? (Some 3D VR gamers claim that they need > 60 Hz to avoid headache or whatever bothers their vision).

8.4 Move the modulated LED close to the aperture of the photodiode enclosure (no need to stick it directly onto the PD, you can easily saturate it). Observe the output from Circuit B on the scope. Capture the waveform for the following frequencies: 1 kHz, 10 kHz, 50 kHz, 100 kHz, save to USB, ppp in your lab notebook. (Or you can take pictures). Does the signal rise/fall edge look sharp? You will find that the LED and PD bandwidths are much larger than that of the op-amp, and the ability to test both the LED and PD is limited by the op-amp.

Step 9 - Test light detection with your remote control

This is not required, but you will get extra 20 points if you can point the remote control toward the PD enclosure aperture and capture the communication signals (set up positive edge trigger and save). Think of it like the remote control "texting" your TV, telling it what to do. The signal consists of a carrier, which is a square wave of ~ 40 kHz, superimposed by the digital "texting signal."

Below are examples.

This is the "texting" message: the codes consist of bit 0 and 1 superimposed on the carrier wave on the right. If you capture something like this, you will get extra credit. You see that the communication rate is ~ 4 bits/2.5 ms, or ~ 1.6 kb/sec	This is a zoom-in of the signal on the left, showing square wave at 37.04 kHz. This is the carrier wave.

Feel free to perform any other test that you can think of, beyond the remote control. For example, if you have a mechanical rotary device, like a small fan (such as a computer fan), or the rotor blade of a drone, you can measure its frequency by letting the blade chop a continuous LED beam, or even ambient light. Optical measurement is considered to be the best method in terms of sensitivity and accuracy to study most mechanical phenomena, and widely used in research and industry. Do experiments that interest you. If it is relevant and demonstrates your learning, you will get extra credit. 

(You can watch the video below to get your idea. Your optical sensor can be used to detect the light scattered from an object like the video).



Step 10 - Discussion and conclusion

Discuss the research questions. Cite your data as evidence to support your points. Below is some "food for thoughts" guidance, something that we can think about: Is this model of the circuit absolutely correct?



The answer, as we see, is "sort of", but "no":  vs[t] cannot be instantaneously proportional to vX[t]. Review your square wave data. If some entity, such as vX[t] depends on another entity vs[t] as cause-and-effect, then there must be mechanism for time delay, because nothing is transmitted instantaneously in the Universe. The time-delay effect can  be observed at the leading edge or trailing edge or both when the input changes so fast that it is "instantaneous." We discuss this in the introduction as well. For those who have learned phasor and Fourier analysis, it means that the response cannot be the same for all frequencies. In fact, the whole Step 6, especially 6.2 is about the frequency-dependence effects. Hence, it is more appropriate to express the op amp circuit in the frequency domain:

 
In this figure, vX[omega] is the input phasor. See the app for phasor concept. vs is the output phasor, which may have a different amplitude and phase than those of the input. Ratio Zfb and Z1 is a complex number that represents that amplitude and phase difference.


Here, the various resistances become complex impedances that implicitly contain the finite-time-response property of the op amp.

The above discussion is for your learning. You don't have to understand if it is beyond what you have learned. But it is an introduction to things that you will learn later in more advanced courses. Hence, you should feel free to express your thoughts, do not feel as if there are certain correct answers and try to guess how to answer them. This is your work, your research, and this last part is to synthesize your knowledge. If you do Circuit A as well, and learn something interesting from your results, such as certain properties of the harmonics of your drone blades, or some objects you like etc., feel free to report and write what interests you and why.

 

Appendix    Quick Lab flowchart








continue to appendix - page i