Internship Experience at Texas Instruments

I read a quote the other day over the internet which I felt would describe me best.

“Sometimes, my attitude is like a Square Wave Signal; either I’ll give my 100% to a thing or none at all! There’s no in-between.”

Electronics is something that I like to give my 100 percent to, and the Electrical field is something that I despise. While it has many fascinating aspects, I have never been as interested in the latter as I have been in Electronics. Now, an internship for me as an Analog Design Engineer at the world's leading semiconductor manufacturing company is like giving Forrest Gump Jenny; that’s all he ever wanted.

Initially, due to the COVID scare, it was speculated that the internship was going to be virtual. A little saddened, I came back to my home after the semester was done. Literally 2-3 days before the internship was about to start, the company gave us the choice of coming onsite as well. Lazy me thought of staying back because last-minute travel and accommodation arrangements would be difficult.

Little did I know that I would be kicking myself for this decision later; seeing my fellow mates witness their respective internships onsite made me envious.

Work-wise it wouldn’t have been different, because the project I had been assigned was a new one, that is, it wasn’t a continuation of someone else’s leftover work. This meant I would be working on the inception stage of the project, and it would be far-fetched to think that the project would reach the testing or verification stage which requires onsite presence.

I was assigned to the Systems Team of the High Speed Amps section. Any operational amplifier (op-amp), with a Gain Bandwidth Product (GBW) greater than 50MHz comes under High Speed Amps Category (HSAMPS), and as a Systems Engineer, my job was to develop useful products/systems using already available ICs/OP-Amps, based on customer requirements.

If you are wondering, there are tons of engineering disciplines which you might find, not just in Texas Instruments (TI), but in many other companies as well. Here we had, going in alphabetical order, an Application Engineer, Design Engineer, Facilities Engineer, Product Engineer, Product Marketing Engineer, Software Engineer, Systems Engineer (aka me), Test Engineer, and a Validation Engineer to name a few. Given the word limit in this blog, I would rather leave the job of explaining each discipline’s functionality to Google.

Finally, coming to my project, my job was to design an active probe for Texas Instruments. Now you must have come across oscilloscopes in your first year physics lab, and in case you missed it due to COVID, there’s one present in the Ideas² Room - go take a look if you haven’t. They are fun and very useful. Now if you want to display and analyze a signal or voltage on a device or circuit, you use oscilloscopes. But there should be a medium to carry this signal from the device under test towards the inputs of the oscilloscope, without altering/interfering much with the original signal. This is where probes come to the rescue. The probes that are present in the Physics lab or in the Ideas2 room are called passive probes, meaning they contain only passive components such as resistors, capacitors or inductors. You must have also come across the concept of capacitive compensation of probes in your Principles of Measurement course. These probes are cheap, rugged and a safe choice for general-purpose measurements, probing and troubleshooting. A general characteristic of a probe should be such that it should offer high input impedance, otherwise any loading (drawing currents) of the circuit will alter the voltage we wanted to measure, causing misinterpretations. So high input impedance would ensure less loading and the circuit under test will behave similar to how it would have behaved if a probe wasn’t connected to it.

But there are few limitations for passive probes. They only take floating point measurements, i.e., they measure voltage of a point with respect to ground, and in most cases we need differential measurements, wherein we are interested in voltage difference between two points on the circuit, neither of which is ground. The input impedance which I was talking about earlier is frequency dependent. That means, due to the presence of inevitable input capacitances, as we increase the frequency of the measured signal, the input impedance decreases. So there will be higher loading at higher frequencies and it would result in errors.

To measure differential voltages, i.e., voltage difference between two points, neither of which being ground, and to have a low input capacitance to reduce loading at higher frequencies, we employ the usage of active components like Op-Amps and ICs. So any probe that contains active components is called an Active Probe and the probe will need an external power supply to run the active components present inside it. Active probes cater for large bandwidths (typically few Gigahertz) and are relatively very expensive (few lakhs).

My first task was to understand various categories of Differential Active Probes that are there in the market. After collecting that data, my mentor and I sat down to identify leading players in the oscilloscope and probe market to know the kind of specifications they provide for their probes in that respective category. Keysight, Teledyne Lecroy, Tektronix, Rigol, etc. are some of the companies that are ahead in the oscilloscope game. So, by the end of the internship, my goal was to design a TINA-TI schematic for any two of these probes in such a way that the signal chain specifications of my design, should match/compete with the signal chain specifications of top probes in that specific category. When I say signal chain specifications, the primary things that a user looks at before buying a probe are its maximum input voltage limit, its 3dB and 0.1dB large signal bandwidth, its input impedance at DC level, its noise specs and cost.

Let's get back on track after having gone a little technical for the last few paragraphs. One of the interesting things about the High Speed Amps Systems team is that it is an inter-continental team. I would have midnight meetings with team members in the Dallas Office when I had to get clarity on certain aspects. I was also a part of a sales call with field engineers of RIGOL, where package and distribution aspects of one of the most efficient op-amps developed by the systems team was being discussed.

The journey of reaching the end results was a bit frustrating at times, a bit overwhelming initially, but whenever I felt that the direction I was headed toward was going to yield me the right results, ecstasy and enthusiasm used to chip in. One of the good things about the way mentors at TI deal with the interns is that they never spoon feed them, rather nudge them in the right direction. This way, it helps the intern in pushing his/her limits and the satisfaction one gets from finding the solution all by himself can hardly be compared. In spite of you being just an intern at TI, the employees at the company will always respect your ideas and treat you as an equal.

Overall, it’s a mixed bag of emotions, but at the end when you get it right, you’ll be over the moon, and the feeling cannot be expressed but only experienced. I urge all the aspirants to explore yourselves and find your interests, because nothing but self interest can drive you towards your goals. To survive in this field, you need not learn fancy problem solving methods or remember a lot of equations. A good conceptual clarity, little bit of logical reasoning and a good deal of self interest is sufficient. Before going into this internship, I never knew that Op-Amps have input capacitance, nor did I know that resistances have a small parasitic capacitance parallel to them. I had never heard of Current Feedback Amplifiers where error current is fed back to the input in place of error voltage as in case of Voltage Feedback Amplifiers which we are all familiar with. But now, Current Feedback Amplifiers are an integral part of my design. So as long as you have that interest, you’ll not find it difficult to inculcate new things.

Find your interests first, find what excites you the most. After that, as MS Dhoni says, ”Trust the process, forget about the end result.”