GD 2nd interview Flashcards

(31 cards)

1
Q

What fundamentally changes when designing PCBs at 10+ Gbps?

A

Traces behave as transmission lines; impedance control, loss, reflections, crosstalk, and discontinuities dominate performance. Layout is as critical as the schematic.

How well did you know this?
1
Not at all
2
3
4
5
Perfectly
2
Q

How do you control impedance on a PCB?

A

By defining stackup, trace width/spacing, dielectric properties, solid reference planes, and validating with field solvers (e.g., HyperLynx).

How well did you know this?
1
Not at all
2
3
4
5
Perfectly
3
Q

What causes eye diagram closure?

A

ISI from bandwidth limits, jitter (random and deterministic), noise, attenuation, reflections, and crosstalk.

How well did you know this?
1
Not at all
2
3
4
5
Perfectly
4
Q

Why are vias a problem at high speeds?

A

They add parasitic inductance/capacitance and can create stubs that cause reflections; mitigated with back-drilling or blind/buried vias.

How well did you know this?
1
Not at all
2
3
4
5
Perfectly
5
Q

What is insertion loss vs return loss?

A

Insertion loss is signal attenuation through the channel; return loss is energy reflected due to impedance mismatch.

How well did you know this?
1
Not at all
2
3
4
5
Perfectly
6
Q

What layout practices reduce crosstalk?

A

Adequate spacing, solid reference planes, orthogonal routing between layers, controlled impedance, and minimizing parallel runs.

How well did you know this?
1
Not at all
2
3
4
5
Perfectly
7
Q

How do you isolate sensitive analog from high-speed digital circuitry?

A

Functional partitioning, careful return-path control, continuous ground planes, minimizing split planes, and keeping high di/dt currents away from analog sections.

How well did you know this?
1
Not at all
2
3
4
5
Perfectly
8
Q

What common mistakes cause noise in analog measurements?

A

Poor grounding, long current loops, power-supply noise, improper probing, and coupling from digital signals.

How well did you know this?
1
Not at all
2
3
4
5
Perfectly
9
Q

How do you select an op-amp for high-speed or low-noise use?

A

Based on GBW, slew rate, noise density, input bias/offset, stability with capacitive loads, and supply constraints.

How well did you know this?
1
Not at all
2
3
4
5
Perfectly
10
Q

Why might you use multiple gain stages instead of one high-gain stage?

A

To maintain stability, bandwidth, noise performance, and avoid saturation or slew-rate limitations.

How well did you know this?
1
Not at all
2
3
4
5
Perfectly
11
Q

Why does low ESR output capacitance improve transient response?

A

It lowers output impedance, supplying immediate charge during load steps before the control loop can respond, reducing voltage droop.

How well did you know this?
1
Not at all
2
3
4
5
Perfectly
12
Q

How would you debug a DC-DC converter that droops under load?

A

Apply load steps, observe transient response, check output capacitance ESR/placement, verify compensation, and inspect switching-node behavior.

How well did you know this?
1
Not at all
2
3
4
5
Perfectly
13
Q

How do you reduce ripple and noise on a switching supply?

A

Proper LC filtering, good layout/grounding, optimized compensation, shielding, and optionally post-regulation with an LDO.

How well did you know this?
1
Not at all
2
3
4
5
Perfectly
14
Q

What belongs on a power entry module?

A

TVS diodes, EMI filtering, inrush current limiting, reverse-polarity protection, and hot-plug protection.

How well did you know this?
1
Not at all
2
3
4
5
Perfectly
15
Q

How does power noise affect high-speed data links?

A

It introduces jitter and threshold modulation, increasing bit error rate.

How well did you know this?
1
Not at all
2
3
4
5
Perfectly
16
Q

A new board draws too much current at power-up. What do you do first?

A

Current-limit the supply, check for shorts, inspect thermally, isolate rails, and bring up power incrementally.

17
Q

How do you probe a switching node correctly?

A

Use a short ground spring or differential probe to minimize loop inductance and measurement artifacts.

18
Q

What is a BERT and what does it measure?

A

A Bit Error Rate Tester measures link integrity by comparing transmitted and received data over time.

19
Q

Why use a sampling oscilloscope for high-speed links?

A

For extremely high bandwidth, low jitter, and accurate eye-diagram and timing analysis.

20
Q

What electrical issues most commonly impact optical systems?

A

Power-supply noise, ground noise, jitter, thermal drift, and poor signal integrity.

21
Q

What must be considered when interfacing analog signals to an FPGA?

A

Signal conditioning, ADC requirements, grounding, noise, timing, and protection of FPGA I/Os.

22
Q

How do you protect FPGA I/O pins?

A

Series resistors, clamps/ESD diodes, proper sequencing, and avoiding back-powering.

23
Q

What should be simulated vs measured?

A

Simulate early to guide design decisions; measure to validate assumptions and catch real-world effects.

24
Q

How do you approach vague requirements?

A

Clarify assumptions, propose an architecture, prototype quickly, validate, and iterate with stakeholders.

25
When do you ask for help?
After isolating the problem, forming hypotheses, and being able to ask a precise technical question.
26
How do you handle technical disagreements?
Data-driven discussion, simulations or measurements, and collaborative decision-making.
27
How do you explain complex technical issues to non-engineers?
Focus on impact, risks, and outcomes using simple language and visuals, not equations.
28
Why General Dynamics and this role?
This role aligns well with my background in mixed-signal PCB design, integration, and test, and it offers hands-on R&D work in a lab-focused environment. The Optical Center of Excellence combines high-speed electronics, systems integration, and mission-driven work, which is exactly the direction I want to grow in.
29
What does your ideal work environment look like?
A collaborative environment with strong technical ownership, access to lab hardware, and opportunities to learn from experienced engineers. I do my best work when I’m close to the hardware and working through problems as a team.
30
Describe a time you had to troubleshoot under time pressure.
During flat-sat integration testing, we were on a tight schedule to verify system-level communication. I created and executed a structured test script so we could isolate failures quickly instead of ad-hoc debugging. By focusing on the highest-risk interfaces first, we confirmed all components could communicate and stayed on schedule.
31