After completing the soldering and initial testing of the V0.5 version of my audio switch module, I encountered several issues worth documenting. This post summarizes the debugging process, insights gained, and potential improvements for future versions.
1. Initial Soldering: Failed to Toggle Between Active and Mute States
Upon the first power-up after soldering, the device only switched from the working state to mute once — and then became stuck. I initially suspected the LED indicator circuitry, but after a round of multimeter probing, I discovered the real issue: one lead of the capacitor’s current-limiting resistor was not soldered at all.
What puzzled me was that the resistor lead was still in physical contact with the pad. I had assumed this should allow at least minimal conductivity, but in practice, the device failed to function until the missing joint was soldered properly.
Lesson learned: Reliable operation requires proper soldering — even physical contact without solder isn’t sufficient for stable electrical connectivity.
2. Audio Signal Transmission Failure
Once relay switching started working correctly, I moved on to soldering the three 1×2 TRRS jack terminals. These are SMD components. Soldering a single one is manageable, but when placed closely together, the tight spacing made the process harder, especially for my soldering iron tip.
After soldering, I noticed that audio was only partially transmitted. Initially, I suspected TRRS/TRS pinout mismatch. However, a round of re-soldering revealed the problem was poor contact due to insufficient soldering. Only one channel passed through on the 1-to-B side. Another round of reflow fixed this.
To improve this for future assembly:
- I plan to develop an external continuity testing module to quickly verify proper contact on all lines.
- Also, I’ll revisit the pad design for the TRRS jacks: their legs are relatively thick and gold-plated, but the current pad size doesn’t promote good “trapezoidal” solder flow. A wider pad might allow a more robust joint.
3. “One-Click Mute” Incomplete Silence Issue
The mute function was designed to cut the R2 and S lines (right microphone and ground), which, in theory, disables the mic for both TRS and TRRS plugs. But in practice, even when mute was active, I could still hear faint audio—background music remained, and only the vocals disappeared.
This seems to be due to a “floating ground leak” phenomenon, where signal still flows through unintended paths. While I’m still verifying the root cause, an effective fix might be to redesign the mute feature to cut T and R1 (left and right audio) instead, which should ensure more complete silence.
Further testing is needed, and I plan to create a dedicated experimental setup to explore and document this behavior in detail.
4. PCB Trace Widths
Currently, all four TRRS signal traces are 0.25 mm wide. While this works, it’s suboptimal for the S (ground) line. In the next version, I’ll increase the width of the S trace to ensure better current handling and lower impedance. 0.4mm to 0.5mm for the S signal line may be better I think.
5. Missing Pull-Down Resistors on All Jacks
All three 1×2 TRRS jacks currently lack pull-down resistors on the T/R1 lines to S. While not a critical issue, this can cause floating noise in low-end audio equipment when one of the jacks is not in use. I’ll add these resistors in the next revision.
Additionally, the T2 line (mic input in TRRS) should have a pull-down as well to avoid potential grounding noise or erratic detection from connected devices like PCs or phones.
6. Flyback Diodes on Relay Coils: Still Missing
Although I included pads for flyback diodes across the relay coils, I haven’t soldered them in the current version. I’ll review designs from other relay circuits to confirm best practices here.
Candidate diodes: 1N4148, 1N4007, or SS14.
7. Grounding Isolation Diode Optimization
The current design includes two SS14 diodes to isolate the pulse-triggered ground path and prevent interference between relays. This was crucial — I previously tested without them and saw clear crosstalk causing false triggers.
However, these diodes drop some voltage (Vf ~ 0.6V), so the relays don’t receive the full 5V supply. A possible upgrade is to switch to BAT54 series Schottky diodes, which have a lower voltage drop (~0.3V), ensuring more voltage reaches the relay coils.
This becomes more important if future versions are powered by batteries or boost converters—lower Vf extends functional range.
Also considering using dual-package versions (e.g., BAT54C) to save PCB space and simplify routing.
8. Resistor Package Optimization
For the test version, I used 1206-sized resistors (especially the 510Ω ones) to facilitate manual probing. These will be changed to 0805 in the production version for improved layout compactness and aesthetics.
That said, routing space is already quite tight on this board, so careful re-evaluation of the PCB layout is necessary.
Final Thoughts
V0.5 brought me valuable insights—not just about circuitry and soldering, but also about real-world quirks in audio transmission, connector behavior, and signal grounding. It’s reaffirmed the importance of mechanical reliability in electrical design and the complexity of mixed analog-digital audio systems.
Next steps:
- Design the external jack continuity testing module
- Investigate “floating ground leak” with experiments
- Revise the mute logic and improve pad layout for better solderability
- Evaluate and install flyback protection diodes properly
- Optimize pull-down resistor placement and package size
The journey continues toward a fully robust and production-ready V1.0 version.