Hi everyone,
I’m designing a board that uses the AMS1117-3.3V fixed LDO regulator to power logic and peripheral components. I’m seeing more dropout voltage and heat than I expected. Hoping someone here can point out what I might be overlooking.
My Setup & Observations
Input voltage is ~5.5V, output meant to be 3.3V.
Load current is about 250-300mA (mix of MCU + sensors + some LED indicators).
I’m using the TO-SOT-223 package version, mounted on a small PCB, with only basic copper pour around the regulator.
I placed a 10 µF capacitor on the input, and a 22 µF electrolytic on the output (no special bypass on the adjust pin).
The voltage drop is higher than ideal: instead of seeing ~1.2V drop at 1A (per datasheet), even at ~300mA I see the input-output difference hovering around 2.0V. The regulator gets warm to the touch after a few minutes, and the thermal limiting seems to kick in (voltage sag under steady load). There’s a noticeable noise/ripple when load changes fast (e.g. when sensors turn on/off).
Is my observed ~2V drop under 300-mA load within “normal” variation for AMS1117, given small PCB copper area and modest heatsinking? Would improving thermal dissipation (e.g. larger copper pours, adding a heatsink) noticeably reduce dropout/voltage sag? If I needed better performance under these conditions, would switching to a switching regulator be preferable, or is there a better LDO alternative you’d recommend?
Thanks in advance.
I’m designing a board that uses the AMS1117-3.3V fixed LDO regulator to power logic and peripheral components. I’m seeing more dropout voltage and heat than I expected. Hoping someone here can point out what I might be overlooking.
My Setup & Observations
Input voltage is ~5.5V, output meant to be 3.3V.
Load current is about 250-300mA (mix of MCU + sensors + some LED indicators).
I’m using the TO-SOT-223 package version, mounted on a small PCB, with only basic copper pour around the regulator.
I placed a 10 µF capacitor on the input, and a 22 µF electrolytic on the output (no special bypass on the adjust pin).
The voltage drop is higher than ideal: instead of seeing ~1.2V drop at 1A (per datasheet), even at ~300mA I see the input-output difference hovering around 2.0V. The regulator gets warm to the touch after a few minutes, and the thermal limiting seems to kick in (voltage sag under steady load). There’s a noticeable noise/ripple when load changes fast (e.g. when sensors turn on/off).
Is my observed ~2V drop under 300-mA load within “normal” variation for AMS1117, given small PCB copper area and modest heatsinking? Would improving thermal dissipation (e.g. larger copper pours, adding a heatsink) noticeably reduce dropout/voltage sag? If I needed better performance under these conditions, would switching to a switching regulator be preferable, or is there a better LDO alternative you’d recommend?
Thanks in advance.