When should a hipot test be performed?

Hi Pot is a destructive test (it degrades insulation), it should be performed last if possible.

Well, that's kind of my point. If it's a "destructive" test, shouldn't it be performed earlier, so that the damage that it may have caused is more likely to be detected and corrected before time is wasted in calibration?

All Hi pot testing is destructive. DC stresses insulation different then AC. It also depends on what you are doing.

Long story short. It doesn't matter when you do it.

"Long story short. It doesn't matter when you do it."

From a testing and operations standpoint thats well and true. From a QA and costing analysis it may not be.

Best practise would be to find the rates of failure of the various tests performed and order them from most to least. This way the loss in materials and labour from rejecting a unit is minimized.

So why do most places do the hi-pot test first? Because it is usually the most common failure mode for the product and also the one with the highest risk and liability associated with it.

There are AC and DC hipot tests. An AC hipot is definitely a destructive test and is generally only done by the manufacturer before declaring the product acceptable for shipment provided it does not fail the test.

DC hipot tests are performed in the field as a maintenence tool to attempt to predict failure without actually causing a failure. This is done by slowly increasing the voltage in steps and observing the charging and leakage currents. A plot of the leakage current can be used to determine the health of a cable or transformer by observing if the leakage current begins to increase disproportionally to the voltage increase. If it does, the test is terminated and a decision made based on the voltage value reached as to whether or not to place the device back in service.

There are various IEEE standards which provide guidance on performing hipot tests.

At our company in Florida, hipot was one of the last tests to be made.....

When I worked in the medical electronic industry, we did it the last step before packaging. When done anytime earlier, there is a possibility of a connector being disturbed and not being detected.

Consider this, if you were having a brain scan monitor connected to you, when would you want the hipot test to be done.

Perfectly put!!

I work for a medical instrumentation company. We perform Hipot after assembly - prior to test to: 1. protect our own people 2. on recommendation of our CSA officer. As a final step we perform the Metron or leakage current again. So Hipot once, Leakage current twice. This insures there are no pinched wires, shorts, etc. after covers are put back on.

How many times are you allowed to repeat a HIPOT test when a unit fails? I hope you could shed some light on this.

Thanks