Charger PCB Assembly

[digitaldanny]

Describe the solution you'd like Assemble the Charger PCB and note any issues with the PCB layout here.

Extra Context:

• 41 Battery Charging Circuit

• 38 Waveform Generator Prototype

[digitaldanny]

PCBs delayed until next Monday (3/29). Remainder of parts came in from Digikey this week, and I will assemble the PCB after I complete #48.

[digitaldanny]

The 2.5x5 mm barrel jack does not fit the wall plug I have. Will need to re-order wall plug on amazon for correct dimensions.

[digitaldanny]

Messed up the connections between the glove connection detector, up/down controller, and frequency controller in the top level charger schematic. The GCON1 header was originally supposed to be shorted from pins 1-2 and pins 3-4, but the corrected connections are from pins 1-4 and pins 3-2 as pictured below.

[digitaldanny]

Completed the Audio Ctrl portion of the charger PCB. Looks like 4 distinct frequencies are being generated as expected. There is some portion right at the beginning and end that need to be eliminated.

[digitaldanny]

The issue I was seeing in #38 or #33 is showing up again where the direction bounces around at the beginning before choosing a direction. Need to look back into these issues to see how I solved the problem in the breadboard prototype.

[digitaldanny]

Completed

• Validated that the Audio Controller logic is working to produce a 4-note tune that plays forward when connected and backwards when disconnected. Still requires some tuning but functionality is all there.
• Validated that the AB amplifier drives the speaker.

Todo:

• Solder CC-CV charger components and tune CC, CV, Schmitt, and Termination circuits.
• Tune resistors for producing specific notes.
• Tune 555 circuits to play tune fully forwards or backwards per connection/disconnection.
• Get a 9-10 VDC wall plug that fits the 2.5 x 5 mm barrel adapter.
• Validate volume amplifier once I get the 9-10V DC input.

[digitaldanny]

Using the variable voltage wall plug, I was seeing a ~250 mV periodic noise at 60 Hz. I tried adding bypass capacitors to fix this, but I was still seeing the noise. My solution was to switch to my fixed 10V power supply, which is outputting much less noise.

[digitaldanny]

Todo:

• Tune 555 circuits to play tune fully forwards or backwards per connection/disconnection.
• Tune CCCV charger circuit.
• Create a CCCV charger circuit demo.
• Test the CV circuit after removing the 100 Ohm resistor on the output voltage.

[digitaldanny]

Because I added in the 10V power supply to the LM324, I do not need to divide the battery input anymore. The screenshot below shows the changes I am making on the PCB.

[digitaldanny]

The CV circuit is miswired. If the top pin of R3 is adjusted by the op-amp to be 4.2V no matter the current draw, the voltage drop across R3 will be variable. This means the voltage will not be a stable 4.2V in CV mode. Shorting the 100 Ohm resistor resulted in an extremely hot Q1. I need a solution that bypasses the R3 resistor while still allowing Q2 and Q3 to control whether the circuit is on/off.

I think rewiring the circuit this way will fix the issue I am seeing, but I will need to test on a breadboard before rewiring on the PCB.

UPDATE: This change did fix the voltage drop across the resistor. But for some reason, I am still seeing variation in the voltage output with varying load resistances. Going to switch to the LM317 as a last minute solution.

[digitaldanny]

Going to try this rewire tomorrow using the LM317 as a 4.2V constant voltage source.

[digitaldanny]

Looks like the Schmitt trigger's output is inverted for the current FET logic setup. Need to add an inverter to the end of the Schmitt detector circuit.

UPDATE:

• After implementing this, I am realizing that this would work much better with a comparator as an inverter. The hysteresis of the Schmitt trigger means that the op-amp output can't swing high and low quickly since the open loop gain is not used. The open loop gain of a 2nd stage op amp would solve this problem because it would swing between both rails instead of slowly switching between high and low Schmitt_n outputs bvetween the 3.7 - 4.1 V levels. Additionally, I wouldn't have the overheating issue of the mosfet inverter.

[digitaldanny]

Here are the issues I am currently seeing.

• The CC circuit cannot provide constant current because Q4 is not providing enough voltage to the LM317 to provide current at 55 mA.
• The threshold circuit still needs to be tuned to switch at around 5 mA.
• Record demo of the CCCV charger.