Is Temperature Causing Electret Formation in High Voltage Testing?

[DCarr10760]

Hi folks,

This is my first post. It is a problem in a manufacturing process and I am hoping to find some information and a solution.

I manufacture high voltage pinhole detection systems use in the manufacture of plastic products. At the moment we have a system in place testing thermoformed LDPE vessels. They are simailar to baby bottle liners in size.

After the parts are thermoformed they are advanced to a test station where they are placed in an electrode fixture (a family mold arrangement, sometimes referred to as a probe and cavity fixture) the probe (Male) is charged to -9kVDC and the cavity (Female) is grounded. The test time is approximately 1 second. Our detection circuitry trips when a 600 uA current passes through a circuit in series with the HV generator.

The problem is that we are imparting a static charge to the surface of the parts under test. The customer has measured voltages as high as -4000 volts. They have tried to neutralize the charge with ionizing bars and a surface voltmeter checks at zero volts. However the next day the parts are again charged sometimes to 1000 volts or more.

Samples have remain charged for several months now.

Untested samples (not HV leak tested) often have a surface charge of up to -1000 vDC or more and after ionizing treatment maintain a voltage of only up to -400 vDC. This level is satisfactory. Higher voltages are not.

At the time of the test, the product is between 150F and 200F.

I am wondering is it possible we are forming an electret?

If so is there a critical temperature the products should be brought down to before bombarding them with voltage to avoid a permanent charge?

Where can I learn more about electret formation?

Many thanks to all!

David C

 

[berkeman]

Do you really need a DC test? Can you use an AC test instead, where the AC voltage magnitude is ramped up over about a second to the peak, and then back down over the next second? This would be similar to the de-Gaussing process used to demagnitize metal structures.

 

[NoTime]

Don't know about LDPE, but your process is generally how one makes an electret.
You could try cooling some to lower temps until it stops happening.
Berkeman's idea sounds promising.

 

[DCarr10760]

Thanks,

We do use AC on some products and as you suggest there are none of these problems. But in this case there are 32 separate cavities all with discrete detection circuits fed by a common supply. The current requirements for an AC test, given the significant capacitance of the cups in the common fixture would require an output currents well above let-go (safe) values. Our business model doesn't allow us to kill our customers :) (it discourages repeat business).

We do have similar installations, testing products that are cooler, which do not gain a permanent charge. So I think temperature may be a factor. I'll go and do some online testing when they will allow me time on the machine.

Thanks all!

David