When Surge Protection Fails: Lessons from a 600 W TVS That Couldn’t Save an Op-Amp

If you are an electronic designer engineer you may want to read this article....

Surge and transient immunity testing often reveals weaknesses that are hard to predict during design.

In one of my recent EMC test sessions, I encountered a case that perfectly illustrates why component-level surge ratings can be misleading — and why system-level surge design is what really matters.

The Setup

The test involved surge testing on an input port connected through cables longer than 3 meters.According to IEC 61000-4-5, any external cable exceeding 3 meters must be surge-tested, since it represents a realistic path for conducted surges entering the equipment.

To protect the input circuit, I had fitted a 600 W bidirectional TVS diode (transient voltage suppressor). On paper, it seemed more than sufficient. The diode’s clamping voltage was well below the op-amp’s absolute maximum rating, and it could easily handle the surge energy defined in the standard test waveform (1.2/50 µs voltage, 8/20 µs current).

Yet the circuit failed the surge test. The operational amplifier input pin was damaged, even though the TVS diode survived perfectly.

in the schematic above the oparational amplifier did get damaged, since we had a voltage applied at its positive input higher that 15V + 0.6V or lower thab -0.6V

The Surprise

After the test, I went in my lab and I simulated a surge test with a power supplly, a switch and a current limit of 6A. To my surprise, I was able to "damage" the op amp and repeat what I hjave seen in the EMC testing lab! the input voltage at the op-amp pin briefly exceeded the chip’s maximum voltage — even though the TVS diode was clamping as expected.The failure happened within microseconds, much faster than what we normally think of as a “surge event.” So I had to improve the input protection!

Understanding What Happened

At first, it was puzzling. A 600 W TVS should easily handle a surge of this size, so why did the op-amp see a higher voltage?

The explanation lies in the dynamic and parasitic effects of the protection network.

1. Cable and trace inductance:The test cable was over 3 m long. That adds roughly 3 µH of inductance per meter, meaning the surge current encountered nearly 10 µH in total.During a fast-rising surge (with rise times in the order of microseconds), that inductance causes a voltage spike V = L·di/dt that can easily exceed tens of volts.

2. TVS diode dynamic resistance:TVS diodes are not ideal clamps. When conducting surge current, their voltage increases with current (V = Vclamp + I × Rdyn).A 600 W TVS may have a dynamic resistance of several ohms, so even moderate surge currents can push the clamping voltage well above the datasheet value.

3. Lead and PCB inductance:The distance between the TVS diode and the op-amp input was small — but not negligible.A few centimeters of trace at high di/dt adds enough voltage overshoot to momentarily exceed the IC’s rating.

Together, these factors created a situation where:

• The TVS diode clamped the voltage correctly at the connector.

• But the op-amp input pin still experienced a short transient beyond its limit.That short spike was enough to damage the input ESD structure and destroy the amplifier.

  
  

How I fixed?

The schematic below shows the extra protection I had to add after the TVS. I had to add an extra protection "closed to the op amp pin, in a way that even if higher voltage would "pas the TVS" the resistior + the clamping diode, would protect the operational amplifier.

I added a resistor and 2 clamping diodes. the resistor was around 10k. I choose 10 k, since I wanted to make sure that the energy going at the input of the operational was so low that the clamping diodes would have worked.

Back to the EMC testig lab

Worked perfectly.... a big lesson learned from my side!

Practical Design Tips

If you design circuits that will face surge or EFT (fast transient) tests, here are some points worth remembering:

• Use two-stage protection: a robust clamp at the connector, and a fine clamp near the IC.

• Keep TVS leads and traces short — every millimeter adds inductance.

• Add series impedance (resistors, ferrite beads, or RC filters) to limit di/dt at sensitive nodes.

• Don’t assume the TVS clamping voltage is the voltage at your IC. Measure it!

• If possible, use surge simulations or oscilloscope captures to understand real voltage at the pin. In my lab I have simply used a power supply with a switch and worked!