{"id":2470,"date":"2026-10-15T06:41:55","date_gmt":"2026-10-15T06:41:55","guid":{"rendered":"https:\/\/photo-detector.com\/?p=2470"},"modified":"2026-10-15T06:42:01","modified_gmt":"2026-10-15T06:42:01","slug":"capacitancia-de-union-del-fotodiodo-cj","status":"publish","type":"post","link":"https:\/\/photo-detector.com\/es\/photodiode-junction-capacitance-cj\/","title":{"rendered":"Domando la Capacitancia de Uni\u00f3n del Fotodiodo Cj: La Gu\u00eda del Ingeniero de Hardware para Amplificadores de Transimpedancia Estables y de Alta Ancho de Banda"},"content":{"rendered":"\n

If you have spent more than an hour at a hardware bench trying to coax clean 10 MHz or 50 MHz pulses out of an optical receiver, you know the drill. You calculate your feedback resistor, pick what looks like a screamer of an op-amp with hundreds of megahertz of gain-bandwidth product, power up the rail, hook up your light source, and hit auto-scale on the scope.<\/p>\n\n\n\n

Instead of a clean, sharp pulse response, you get a 65 MHz sine wave humming away at rail-to-rail levels. Or worse: a mushy, slow step response that looks like someone dragged your signal through cold molasses, followed by three cycles of violent ringing.<\/p>\n\n\n\n

You check your power supply rails. Clean. You inspect your bypass caps. Solder joints look solid. So why is the transimpedance amplifier (TIA) acting like a Colpitts oscillator?<\/p>\n\n\n\n

Nine times out of ten, the culprit sitting silently on your copper is the photodiode junction capacitance Cj<\/strong>.<\/p>\n\n\n\n

When engineers design analog front ends (AFE) for optical sensing, pulse oximetry, LIDAR, or industrial flame detection, they usually obsess over the photodiode’s responsivity and dark current. Those matter, sure. But if you ignore your photodiode junction capacitance Cj, the parasitic physics of your silicon detector will wreck your loop phase margin before the signal ever makes it past your first amplification stage.<\/p>\n\n\n\n

Let’s walk through how photodiode junction capacitance Cj actually interacts with your amplifier loop, how dropping your capacitance from 70pF down to 15pF (or lower) rescues circuit cutoff bandwidth, and what you can do on your PCB right now to keep your optical front-ends quiet, fast, and rock-solid stable.<\/p>\n\n\n\n

\n\n\n\n

The Root Problem: Why Your TIA Loop Hates Input Capacitance<\/h2>\n\n\n\n

A transimpedance amplifier converts an input current from a photodiode into an output voltage. In textbook schematics, it looks deceptively simple: one operational amplifier, one feedback resistor Rf, and a photodetector diode pointing toward the inverting terminal.<\/p>\n\n\n\n

\n

Reality check from the lab bench: ideal op-amps do not exist, and zero-capacitance diodes are a fantasy. The photodiode junction capacitance Cj forms a low-pass filter with your feedback resistor in the feedback path, which turns into a zero in your noise-gain transfer function. That zero eats your phase margin alive.<\/p>\n<\/blockquote>\n\n\n\n

To see what is happening under the hood, look at the total input capacitance Cin hanging off the inverting node:<\/p>\n\n\n\n

Cin = photodiode junction capacitance Cj + Ccm + Cdiff + Ctrace<\/p>\n\n\n\n

Where:<\/p>\n\n\n\n