ESD testing method, for Class II input medical devices
At the higher test level, any mistakes in the testing method become more pronounced, and may cause damage to the DUT (device under test). The more common error may made during testing at the high 15KV level, is failure to discharge accumulated charge voltage when testing Class II input medical devices, between subsequent ESD test events.
When a Class I input device is tested, the output is always attached to earth, which is the return path for ESD, therefore no high voltage can accumulate across the xfmr bridging Y1 series capacitor network.
However, due to the floating nature, of the output when the input classification is Class II (or Class II with functional earth), the voltage across the xfmr bridging Y1 series capacitor network (xbyscn), may take on a large DC value following a 15KV air discharge ESD event.
In this scenario, the power supply under test, will ring at thousands of voltages amplitude between typically four and ten megahertz precisely after the high ESD energy dump into the DC output of the power supply. The fundamental impedances during the ringing event are as follows:
- Input Common mode choke inductance.
- AC line common mode impedance, which may be either capacitive or inductive at the ring frequencies, but is typically measured to be inductive, when using a network analyser between four and ten megahertz.
- ESD gun series resistance, and series capacitance.
When the ringing stops the polarity of the DC across the xbyscn, will typically be the same polarity of the ESD gun. If the CMC has a spark gap in parallel with it (or 2), the spark gap will help limit the ringing energy, by dissipating said energy in the break-over spark gap. Should the spark gap be wide (over 1mm), or missing altogether, the residual DC voltage after the ring-down event may be in the 5000V area.
Applying a second ESD pulse of the same magnitude immediately after the first, may result in the residual DC voltage after the ring-down event increasing to perhaps 7,000V. Eventually the voltage will build up after 3rd or 4th ESD pulse test event, so the first ring ½ cycle peak voltage will exceed 12KV. With 12KV or greater directly across the xbyscn, virtually all power supplies will experience failure of (and/or logic) optocoupler, xfmr, output voltage error-amp, input SMPS controller.
However, if the residual DC voltage after the ring-down, is discharged according to the requirement in the ESD standard (standard referenced by 60601-1-2, 4th), each subsequent ESD pulse test event, will start from zero volts across the xbyscn, thus preventing ratcheting up of the voltage stress across the xfmr isolation barrier.
This discharge requirement is very well explained in the ESD standard, in the 2001 version of the standard, the clause number is 7.1.3, titled “Test method for Ungrounded Equipment”. Although not using the lexicon Class II to describe the input character of the system, certainly the Class II medical device system employing an external power supply fits the category of ungrounded equipment.
In the Rationale, of paragraph 2 of this section in the standard:
‘Rationale: Ungrounded equipment, or ungrounded part(s) of equipment, cannot discharge itself similarly to class I mains-supplied equipment. If the charge is not removed before the next ESD pulse is applied, it is possible that the EUT or part(s) of the EUT be stressed up to twice the intended test voltage. Therefore, double-insulated equipment could be charged at an unrealistically high charge, by accumulating several ESD discharges on the capacitance of the class II insulation, and then discharge at the breakdown voltage of the insulation with a much higher energy.’
The standard further instructs how to prevent the negative outcome of EUT failure, by discharging the residual charge in further paragraphs within the section:
‘To simulate a single ESD event (either by air or by contact discharge), the charge on the EUT shall be removed prior to each applied ESD pulse,’ and ‘a cable with 470 k-Ohm bleeder resistors, similar to the one used with the horizontal and vertical coupling planes, shall be used; see 7.1.’
As an alternative the following options can be used:
The time interval between successive -discharges shall be extended to the time necessary to allow natural decay of the charge from the EU.
A carbon fibre brush with bleeder resistors (for example, 2x470k-ohm) in the grounding cable.
An air-ioniser to speed-up the ‘natural’ discharging process of the EUT to its environment.'
At GlobTek, in our EMC lab, we typically use a discharging brush with 2x 470K-ohm resistors, but several EMC labs we work with have elected to use the long time interval between successive discharges method. A long time, would be greater than three minutes dwell time between discharges.
It should also be noted, (perhaps obviously), that if a ESD discharge event is near a metal part which can not be reached with the discharging brush, since the metal part has a narrow creepage or clearance path to the metal discharge between two insulating plastic pieces, attempting to discharge accumulated ESD charge from the plastic housing is a totally useless method. It is essential, if the ESD jumps to the local metal component, that ESD is discharged from that component in some fashion.
Since the failed component when the discharge procedure is not done properly (or typically, was not done), is in the power adapter, a correct discharge point will be on the output connector of the power adapter.
This may require removing the DC power plug, and discharging this point, and then reinserting the connector, before the next ESD test event. This method is typically faster than waiting 3 minutes between subsequent ESD test events.
When 15KV ESD testing levels are applied, to medical equipment systems using a Class II input configuration, it is necessary to remove accumulated ESD charge between successive ESD pulses. This is generally not well known information by many EMC test engineers, since most EMC immunity tested devices, are tested to a 8KV ESD test level, or if they are medical devices are Class I input.
Therefore if a Class II input medical system is reported as failing the ESD test regime, it is typically necessary to first determine if the EMC test lab performed the required residual ESD discharge process between subsequent ESD test events, as detailed in 61000-4-2.