RL –
My point was to question your statement that Enclosures as defined wouldn’t normally be earthed at all – for Class 1 devices they normally would be. Your later point that Enclosure Leakage from earthed accessible parts on the Enclosure will be negligible under Normal Conditions was the point I was trying to make – apparently not very well. I agree that under single-fault ‘open-earth’ conditions the measured Enclosure Leakage would rise – for earth-bonded Accessible Parts you’d normally then expect it to approximate the Earth Leakage current you’d measured under no-fault conditions.
The Enclosure is regarded as the exterior surface of the equipment – as simple as that, (there’s a definition in the Standard if you wish). If I understand you correctly you appear to have used the term ‘Enclosure’ in a more restricted sense (which I believe, is the source of our confusion) - you appear to be restricting its usage to only those parts of the ‘box’ from which leakage currents are to be measured in the Section 19 tests you have referred to. But Section 19 isn’t intended to define Enclosure (that’s the job of Section 2), it merely describes how you should measure leakage from different parts of the Enclosure. If it doesn’t require you to measure from earthed parts of the Enclosure that doesn’t mean these parts aren’t part of the Enclosure.
“I think you are confusing mechanical definitions of Applied Parts with electrical ones ….”
I think this distinction is yours, not the Standard’s.
Examining your statement -
“For Class1B I would test earth-bond, see what’s conductive and earthed and then any other accessible conductive parts are considered as enclosure. Then test for enclosure leakage of these non-earthed parts ….”
I think you’re misusing the term ’Enclosure’ (see above).
That aside, using this method, in the event that you identify a non-earthed conductive Accessible Part – how do you immediately know whether it should be earthed or not? It may well pass your subsequent leakage current tests, but in the event of a single internal fault could it become hazardous live? – could a stray internal 230V conductor make contact with it, for example? (ie: Is it intentionally un-earthed because it is internally double-insulated from hazardous voltages? – or is it un-earthed because the bonding strap has become disconnected or, in error, was never connected on the production-line?). You appear to have assumed that the unearthed conductive parts you have found weren’t intended to be earthed; on the other hand they could be an indication of one of the very things you are testing for – a failed/missing bond. Further investigation is therefore required to confirm which is the case – it isn’t always obvious - before proceeding with the leakage current measurements you describe.
“.. procedure for assessing which are the applied parts in section A1.10”.
I don’t necessarily agree with your application of A1.10 either. (Firstly, if you classify the bed as 1B then I think the ‘B’ indicates that you have already acknowledged it has an Applied Part - see below). Applying A1.10 – if we say patient contact with bedframe is incidental to the functioning of the bed then the frame may be relegated to the status of Enclosure if contact can only be made by deliberate action on the part of the patient. Personally I would say contact with a bed frame is likely without deliberate action - in fact the condition of many patients is such that they are not capable of doing anything deliberately. You may well debate my reasoning here, but it appears consistent with the ‘table-top’ example quoted above (which comes from A2 of the Standard).
“Just because you can measure patient leakage on a tester …………”.
Couldn’t agree more – all your tester does is measure current; we put the label (‘Earth Leakage’, ‘Patient Leakage’, Auxiliary Patient Current’ etc) upon it, which requires a good understanding of the definitions of Enclosure, Applied Part, etc. – hence this discussion.
Mains-on-Applied Part is something different again. Personally I wouldn’t do this test on an in-service medical device - it is a ‘Type-test’ which potentially stresses the device much the same as dropping it on the floor to test its robustness might do – to be done by manufacturers on sample devices but not by maintenance departments. The test not only has no meaning for a Type-B Applied Part – which is why there are no limits specified for Type-B Parts in the Standard – but there then really is potential for damaging the device under test.
To go back to where we started – I queried your explanation of Type B, BF, CF allocations. I think your understanding is based upon old versions of the Standard. In the current version of 60601-1 the Type classifications have been redefined to be applicable only to Applied Parts. A device with no Applied Parts therefore can have no Type category, (conversely a device with more than one Applied Part could carry more than one Type categorisation if its different Applied Parts differ in their level/means of protection). My understanding is that under the old definitions it would have been possible to have classified Equipment without any Applied Parts as 1B – now the same device would presumably simply be Class 1, because it has no Applied Part to be categorised. I think much of the discussion we have had arises from this distinction.
I think that most of this debate arises from misunderstandings resulting from the fact that we aren’t using terms such as Class, Type, Applied Part and Enclosure in the same way. It shows the importance of consistent terminology. If you still think I’ve misinterpreted aspects of the Standard please let me know.
