The Nuances of Transient Protection

Protecting circuits against transients is easy in theory, but the details matter.

I've been spending a lot of time on the road as of late. Last week I was on a ferry to visit a customer in Connecticut. Today, as I write these words, I'm on a Jet Blue flight to Texas to visit a different customer. Since I'm spending this part of my life as a transient, I thought that transient protection would be an excellent topic to discuss.

Voltage and current transients are a risk in practically all electronic devices that operate from a distributed power system. LED lights are no exception. The power network in a typical home experiences many different potential sources of electrical disturbance on the power lines, including:

75 Volt surge on 28 VDC bus.

• Load-induced transients (switching of high-power loading devices, e.g., electric stove)


• Inductive loads (primarily motors in applications such as AC units, refrigerators, etc.)


• Lightning-induced transients. Lightning doesn't have to strike the lines, it just has to come close to power lines to induce spikes.


• Utility-induced transients (brown-outs, line repairs, power surges, etc.)

For these reasons, it's very important to protect the electronic device, and in our case the LED light. The big benefit to LED lighting is long life. There's no point to touting this advantage if the driver circuit fails at the first transient. Given the LED's long potential life, the odds are high that it will see a transient.

Most of the electronic components inside the LED driver are low-voltage devices consisting of driver controller chips, capacitors, resistors, FETs, LEDs (of course), and other similar devices. These devices are susceptible to catastrophic damage when they are exposed to high voltages or current surges.

The protectors

Enter transient protectors such as transorbs and metal oxide varistors (MOVs). They are the simplest and most common means to protect circuits and devices. They behave similarly although they have entirely different construction.

A transorb is a semiconductor device akin to the Zener diode. The MOV is more closely related to the thermistor. Generally speaking, the transorb is more precise than an MOV, whereas the MOV can usually absorb more energy. Unlike most other electrical components, these devices are designed to absorb and dissipate lots of energy very quickly.

Transient suppressors sit across the input to the electronic assembly and act as a watchdog. They normally do nothing to the circuit. Their entire mission in life is to spring into action if, and only if, the input voltage exceeds a threshold that could damage the downstream components.

Finesse

Designing them properly into a circuit requires a little finesse. I have often seen these devices installed directly across the power input. I prefer to add a small input resistance between the transient protector and the input, especially with lower power devices. Yes, I know, adding input resistance does reduce efficiency, but it also adds a known quantity to the input impedance, and a side benefit is that it usually helps the input filter design.

The transient protector clamps the input line to the the protector's rated voltage. The input impedance limits the current to the transient protector. If there is no impedance between the transient source and the transient suppressor, the resulting current could exceed the transient suppressor's maximum current rating. You may well point out the input lines have do have some impedance -- but it's an unknown quantity. Counting on line impedance is dicey at best. It depends on the distance from the transient source.

It's very important to make sure that the transient protector clamping voltage is less than the maximum rating of the downstream components. The main part of the design is then to ensure that the transient protector can absorb the energy delivered on the input lines. It's a matter of balancing the maximum input current (the current that flows through the input resistor with the voltage of the spike at one end and the maximum voltage of the transient protector at the other) and spike duration with the energy or power rating of the transient protector.