What makes test equipment fail

[arydberg]

TL;DR Summary

why is powering up test equipment more stressfull than running it 24 hours a day.

I worked for a relay company and the test equipment was left on 24 hours a day. One year they decided to shut down the company for 2 weeks so everyone got the same vacation. All the equipment was turned off. After the 2 weeks when everything was turned back on they had a tremendous number of failures. Any thoughts as to why.

 

[Paul Colby]

Thermal cycling can causes additional stress?

 

[Joshy]

I'm sure there are many answers and I wouldn't even come close to knowing half of it.

I did work on one role that involved transient protection circuitry. Something I never thought about often was capacitative loads responding to an input step function that looks awfully like something turning on. The thing I worked on had so many requirements like how much power it could consume during DC operation, maximum voltage and current it would need to be able to support, temperature ranges, and rise time; it really took a lot to make it happen especially while considering real parts with parasitics (mostly unlike the classroom) it was nothing like the simple RC I was expecting although this was ultimately how I wanted it to behave.

 

[berkeman]

Summary:: why is powering up test equipment more stressfull than running it 24 hours a day.

After the 2 weeks when everything was turned back on they had a tremendous number of failures. Any thoughts as to why.

Was this standard off-the-shelf test equipment, or test fixtures that your company had custom-built? If custom-built, what kind of power supplies were used in them, and how close to full load were they running?

As a power supply wears out, the turn-on transient can be one of the more stressful situations for it. That's when you can have trouble with overshoot transients that can damage devices powered by the supply, for example. If the power supply uses a relay to turn itself on and has a high-inrush load, the turn-on cycles put a lot of wear and tear on the relay contacts (which is why such relays often use an NTC thermistor in series).

Other types of equipment (especially those with displays) will wear out faster if left on all of the time. CRT phospurs wear out, and even LCD displays and their backlights wear out. Devices with displays are generally best left powered-off when not in use.

 

[Baluncore]

Switching supplies have a high value resistor (≈120k) in the power supply to start the converter. The wattage rating is correct but not the voltage. That resistor has the full rectified AC supply across it whenever turned on.

Resistors go open circuit if continuously subjected to high DC voltages. If equipment is left on continuously for a few years, then turned off and on again, it may not re-start. Loss of supply can be due to a power cut or a vacation.

Resistor ratings that were tested on 120 VAC, may then be run on 230 VAC and show that high voltage = high resistance = failure.

 

[russ_watters]

Did they turn off the air conditioning too?

 

[DaveE]

It's primarily thermal cycle stress on the semiconductors. They really don't wear out in "in spec" steady state operation. This is particularly true with the ones that dissipate a lot of heat. A common failure mechanism is thermal stress (sort of like work hardening) of the interface materials between the silicon and the heat sinks, which is typically some sort of solder like alloy.

The big exception is large electrolytic capacitors which do slowly "cook to death" (i.e. the electrolyte leaks out).

Electrical transients at turn-on are only significant in bad designs, which aren't uncommon. This is opposed to electrical transients that occur at random (like lightning hitting your building). One is bad luck that can be minimized by turning the equipment off so it's not as exposed, the other is lack of knowledge or minimizing costs (often development time).

You will seldom see it in any specification, but good equipment designers are concerned with the number of power cycles that there equipment is expected to see. This often requires accelerated life testing of key components, like power semiconductors, since the manufacturers will not give you that data. Some device manufacturers are, in fact, much better than others in this regard. This is why large companies, that can afford it, will do "acceptance testing" for new suppliers. The data sheets are a great starting point, but if you want a very reliable product, you must do your own evaluation and testing.

 

[Rive]

There was also one spectacular failure mode with electrolytic caps: some caps tends to 'dry up' with time and lose capacitance - but as long as they are 'warm' they remain partially operational. Once they cool down the remaining capacitance goes zero => you switch off the device and it just won't start up again (not the normal way, at least).

Spoiler

We kept a hair dryer at the office so we could provide 'warm startup'

 

[jsg2021]

Summary:: why is powering up test equipment more stressfull than running it 24 hours a day.

I worked for a relay company and the test equipment was left on 24 hours a day. One year they decided to shut down the company for 2 weeks so everyone got the same vacation. All the equipment was turned off. After the 2 weeks when everything was turned back on they had a tremendous number of failures. Any thoughts as to why.

There's no reason why it should have equipment problems either from being left on 24x7 or having power cycled. Test equipment is designed for this. However there are contributors specific to relay testing and test equipment that themselves use internal relays:

• Relay coils can create large voltage transients when switched off (test equipment relays and front-end semiconductor electronics)
• Relays contacts have finite life cycle times - far shorter than might be imagined given modern electronics life spans (test equipment relays)
• Relays contacts are aged more quickly with hot switching (either DUT or test equipment relays)

These could cause a problem when a test cell is turned off if the "fixturing" to the DUTs are not designed right or if the test control puts out-of-spec voltages on test equipment inputs. At least without knowing more about the specifics of the equipment, the DUT fixturing and tests performed, that would be my wild guess.