Rethinking this simple circuit - September 2025
Input Protection PCB v1.0 - August 2025
Protoboard Testing - August 2025
| Hot Plug | Inrush Current Max | Transient Voltage Max at Input | Transient Voltage Max at Load |
|---|---|---|---|
| No Input Protection into 4.7uF Cap (from previous test) | 29.4A | 70.4V | 75.2V |
| No Input Protection into LED Driver PCB (from previous test) | 29A | 57.6V | |
| With Input Protection into 4.7uF Cap (300W DC supply) 10 ohm ICL | 4.24A | 51.8V | ~136us turn on |
| With Input Protection into LED Driver PCB (300W DC supply) 10 ohm ICL | 4.36A | 50V | ~80us turn on |
| With Input Protection into LED Driver PCB (CUI) 10 ohm ICL | 4.08A | 52V | ~100us turn on |
| With Input Protection into 4.7uF Cap (300W DC supply) 5 ohm ICL | 6.4A | 50V | ~75us turn on |
| With Input Protection into LED Driver PCB (300W DC supply) 5 ohm ICL | 6.24A | 51V | ~50us turn on |
| With Input Protection into LED Driver PCB (CUI) 5 ohm ICL | 6.88A | 50V | ~50us turn on |
| Short Circuit | Short Circuit Current Max | Steady State Power Draw during Short Condition |
|---|---|---|
| No Input Protection 300W DC Supply (from previous test) | 49.6A | 17W |
| No Input Protection CUI SDI24-48-UD-P5 (from previous test) | 44A | - |
| With Input Protection (300W DC Supply) 10 ohm ICL | 14A | 48V 0.023A 1.104W |
| With Input Protection (CUI SDI24-48-UD-P5) 10 ohm ICL | 5.4A | - |
didnt get any sparking this time from short circuit tests
| Overcurrent (48V with 25 ohm load 1.92A) | Time to trip | PTC Temp at Overcurrent |
|---|---|---|
| With Input Protection (300W DC Supply) | 4.83S | 110C |
| Input Protection | Power Draw Max light | Temp (thermocouple) | Temp (FLIR) IGNORE | Power Draw idle | | --- | --- | --- | --- | --- | | No Input Protection | 48V 0.413A 19.82W | 25C | 25C | 48V 0.021 1.008W | | Input Protection ICL: 10ohm 1A 5mm dia 0.4ohm at 1A, 0.8ohm at 0.5A 150C max temp, 6.5mW/C SL0510001 | 48V 0.439A 21.07W (+1.25W) | 74C | ICL 88C (+63C) | 48V 0.021 1.008W (+0W) | | Input Protection ICL: 5 ohm 1A 750mOhm 6mm dia unspecified 150C max temp, 5mW/C SL055R001 | 48V 0.437A 20.93W (+1.11W) | 62C | ICL 72C (+47C) | 48V 0.021 1.008W (+0W) | | Input Protection ICL: 8 ohm 2.2A 279mOhm 8.5mm dia 0.279ohm at 2.2A, 0.9 at 1A, ~2ohm at 0.5A expected 75C at 2ohm B57153S0809M000 | 48V 0.432 20.73W (+0.91W) | 70C | 85C | 48V 0.021 1.008W (+0W) | | Input Protection ICL: 8 ohm 2A 8.5mm dia expected 3ohm at 66C. 170C max temp, 8mW/C ICL800D07MIB | | 66C | 91C | | | Input Protection ICL: 5 ohm 2A 7mm dia expected 1.5ohm at 71C 150C max temp, 7mW/C ICL500D05MIB | | 71C | 91C | |
I am realizing a flaw in the v1.0 design of this circuit. Can you guess what it might be? It has to do with overvoltage conditions. Not transient voltage spikes but persistent overvoltage. The SA48A TVS diode isn’t designed to experience persistent overvoltage. This isn’t something I expect because I am planning on the supply being Class II (<60V, <100W). But lets say you apply 60V at the input in the correct polarity. SA48A will breakdown at 53.30-58.90V. 60V will cause the diode to reverse conduct. The max temp of the diode is 175C. At 25C ambient the max allowable rise is 150C. And the Thermal Resistance Junction to ambient is 75C/W. So it will rise about 75C per watt. Meaning the diode can dissipate about 2W continuously. So at 60V that is about 33mA. TVS diodes typically fail in a short circuit which would then overload the supply if it is not limited.
In order to protect the TVS diode I put the PTC resettable fuse in front of it.
I also decided to add a thyristor in order to protect against constant overvoltage events. The thyristor SP1 acts as crowbar device, meaning when the voltage exceeds a threshold, the parts acts a near short and triggers the PTC fuse. And I added a small capacitor to smooth input voltage transients.
Input Protection PCB Schematic v1.1
