Repair case · VCC hiccup
KA3845 VCC Hiccup Case: High-ESR Bias Capacitor
This case separates a controller that never starts from one that starts correctly but loses bias energy during the first switching pulses. The repeated VCC ramp is the key evidence.
Case summary
This is a documented repair pattern. Exact capacitance, voltage rating and VCC waveform depend on the board.
Initial symptom
- Fan twitched or the machine produced a regular ticking sound.
- Display briefly flashed and went dark.
- The 24 V rail rose momentarily but could not remain stable.
- The input limiter did not indicate a hard primary short.
Power-off evidence
- No hard short was found on the 24 V, 15 V or 5 V outputs.
- Secondary rectifiers and the switching MOSFET passed basic short checks.
- The VCC electrolytic had no dramatic visual damage, but ESR was elevated relative to a suitable replacement.
- The startup resistor chain remained continuous and within the board's expected range.
Controlled live evidence
| Signal | Reference | Observed pattern | Interpretation |
|---|---|---|---|
| Controller VCC | Controller ground pin | Ramp → startup → rapid collapse → repeat | The controller starts but bias energy is not sustained. |
| VREF / RT/CT | Controller ground pin | Briefly present during each cycle | The controller is not permanently dead. |
| Output pulse | Controller ground pin | Short burst during each VCC peak | Switching begins, then stops as VCC falls. |
| Secondary rail | Secondary return | Brief rise synchronized with VCC | Energy transfer exists but cannot continue. |
Diagnosis
The controller reached startup and produced pulses, which ruled against an open startup resistor and a permanently disabled oscillator. With secondary shorts isolated and auxiliary-winding continuity confirmed, the weak VCC capacitor became the evidence-supported cause: it could charge slowly but could not supply the controller during the first switching interval.
Repair action
The VCC electrolytic was replaced with a suitable low-ESR, temperature-rated part matching the board's capacitance, voltage and polarity requirements. Solder joints on the auxiliary-winding diode and bias path were reworked only where inspection showed a defect. No controller replacement was required.
Verification sequence
- Confirm VCC rises once and stays stable after switching begins.
- Confirm continuous VREF, RT/CT and output pulse.
- Confirm the auxiliary winding takes over the bias supply.
- Measure 24 V and other secondary rails unloaded and loaded.
- Reconnect fan, relay and control loads individually.
- Repeat cold start and warm restart to confirm the hiccup has not returned.
Stop conditions before another power-up
- A secondary short has not been excluded.
- The replacement capacitor polarity or rating is uncertain.
- VCC rises uncontrollably or exceeds the board's expected design.
- The MOSFET gate waveform is abnormal or the current limiter remains bright.
- The auxiliary-winding diode or transformer shows overheating.
Technical sources
- Texas Instruments UC3845 product page — pin functions, UVLO, VREF, current limit and maximum duty-cycle reference.
- UCx84x current-mode PWM controller datasheet — controller-level limits; it does not replace the actual board schematic.
- Texas Instruments TL431 product page — secondary feedback reference when the board uses TL431/KA431 regulation.
- WelderData internal technical records — used for measurement sequence and repair-pattern organization; exact readings remain board-specific.
Related repair map
Follow the evidence path from symptom to measurement, circuit and repair case.