Interpretation of "Any or All"'?

L.S.

We had an in-house discussion on how to interpret the phrase "Any or All" that is used in various clauses. My native English team-leader says that the common interpretation of the phrase is apply it on one, all or any combination.

I am native Dutch and I try to interpret it the way it best suits me: as literally either all or a single element of (worst case) choice. My rationale being that in USA law texts the phrase "the combination of Any or All" is frequently used. And that my interpretation is more in the spirit of fault conditions. In this I presume that the wording of all IEC-60601 text is well thought and deliberately chosen.

The proof of this pudding is in the eating by our test house and NB, but before we turn to them both my team-leader and me would like some more ammo to shoot at each other.

What is the common opinion of this forum on this matter?

Definition of 'Any' is:

used to refer to one or some of a thing or number of things, no matter how much or many.

Click to expand...

This issue is English is a horrible language. It changes the rules constantly. In my opinion, 'any' is used verbally because it sounds better and is more authoritarian than saying something else.

For example, 'Any person can learn to walk' sounds better and then 'most people can learn to walk'. The use of 'any' is obviously incorrect, not everyone can learn to walk, but when used verbally, I have noticed that most people will use the 'any person' instead of 'most people'. I think it also has to do with the change from 'person' to 'people'.

I don't know if this helps with 60601 medical electrical equipment standards as it is mostly opinion.

Unless the context suggested otherwise or the matter was one of sufficient importance that rejection of the imprecise wording would be necessary, I would just go with "one, all or any combination".

"the combination of Any or All" is frequently used. "
Actually, usually written as "any and all". The use of "and", "or", etc. should be viewed as 'logic' meanings (mathematical logic), that is exclusionary or combination. As other point out, basic meaning would be 1 condition, up to all conditions, inclusive. This only excludes the null condition.

I too view it as, "one, all or any combination." However, Hogheavenfarm makes a good point that it is more familiarly used with the word "and" not "or." I would take a look at the context just to make sure your version makes sense.

Loekje said:

L.S.

We had an in-house discussion on how to interpret the phrase "Any or All" that is used in various clauses. My native English team-leader says that the common interpretation of the phrase is apply it on one, all or any combination.

I am native Dutch and I try to interpret it the way it best suits me: as literally either all or a single element of (worst case) choice. My rationale being that in USA law texts the phrase "the combination of Any or All" is frequently used. And that my interpretation is more in the spirit of fault conditions. In this I presume that the wording of all IEC-60601 text is well thought and deliberately chosen.

The proof of this pudding is in the eating by our test house and NB, but before we turn to them both my team-leader and me would like some more ammo to shoot at each other.

What is the common opinion of this forum on this matter?

Click to expand...

Dag Loekje,

Leuk om een Nederlander to ontmoeten hier.

Here is a link that may help you.

Let me know.

Hartelijke Groeten.

Jan.

In IEC 60601 "short circuit of any or all insulation" or something like that means that you can short-circuit each one separately "and" short-circuit all of them together.

For example:

Any insulation, CREEPAGE DISTANCE, AIR CLEARANCES, component or earth connection that does not comply with the requirements of 8.5.1.2 and 8.5.1.3 shall not be considered as a MEANS OF PROTECTION. Failure of any or all such parts shall be regarded as NORMAL CONDITION.

Click to expand...

Means that the failure of any of the separate insulations, and also the failure of all together, are normal condition, and this needs to be taken into account during testing.

And yes, this is related to worst case, but you have to try them all to conclude which is the worst case (and depending on the test, the worst case can vary).

OK,

Thanks so far.

It seems like Marcelo is in my corner, the rest is with my team-leader.

I'm happy to add some more to the confusion.
Although my French is even worse than my English, and this being an English forum, I take the liberty to take the French CEI-60601 along (I now deeply regret the fights I had in high school with my French teacher).

For example take 8.1:
French:

– le court-circuit de toute ou partie de l'isolation qui n'est pas conforme aux exigences de
8.8;
– le court-circuit d'une ou de toutes les LIGNES DE FUITE ou DISTANCES DANS L ' AIR qui ne sont
pas conformes aux exigences de 8.9;

Click to expand...

English:

– short circuit of any or all insulation that does not comply with the requirements of 8.8;
– short circuit of any or all CREEPAGE DISTANCES or AIR CLEARANCES that do not comply with
the requirements of 8.9;

Click to expand...

Both English sentences have the "any or all" phrase, but the French version distincts between: "toute ou partie" (if I translate correctly "All or a part of") and "d'une ou toutes" (which translates literally "of one or all"). Any native French speakers on this forum?

Do you think I can prove a point to my team leader and further on the test house/NB?

I'm specially keen on the Normal Condition of short circuiting "any or all " creepage distances that do not meet the requirements of 8.9. If I limit the current with 2MOPP in a reference electrode to 10uA (the return current of the signals) when short circuiting either the +5V or the -5V secundary power of a signal electrode input opamp the patient is always safe. But if I must short circuit one signal electrode to +5V and another to -5V the patient will have 10V over the two signal electrodes in NC.

The spirit of the norm is that a smal piece of metal, or a droplet of solder may bridge a too small creepage distance or air clearance in Normal Condition. Having a combination of any or all creapage distance bridged is like sprinkling metal all over your electronics. Not a Normal condition IMHO.
Bridging all small creepages distances and air clearances is almost never a problem, the power supply will likely fail.

This question might inadvertently highlight the problems with the "single fault condition". Fault conditions are good tools for testing and validating safety, but when used to derive requirements the SFC philosophy has a lot of problems.

In principle, the design should incorporate protection against hazards. The key point is to identify the protection, and design and test it to make sure it mitigates the hazard effectively. This testing may include fault conditions.

In the case of electrical shock, protection is normally provided by barriers consisting of solid insulation and spacings (MOP). The designer should know precisely where those barriers exist, and then verify them through tests and examination.

It should not be an accident that a device passes a test. A device may, for example, pass 4kV dielectric strength due to decorative plastic on the outside, or the accumulation of some spacings that were never intended to provide safety insulation, which prevent the voltage from getting to some sub-standard solid insulation. This situation should not be allowed.

On the other hand, the manufacturer that has 2MOPP@240Vrms creepage distance deliberately structured from three sections of 3.0, 3.0 and 3.0mm with no bridging components is fine (if not a little complicated).

Because the standard is based on the SFC, it is necessary to include the "any or all" statements in 8.1a, but really this is just a catch to cover cases like the accidental pass above. It should be rarely used in practice.

It would be better if the standard simply focused on identifying and qualifying the MOP. The so called "fundamental rule" should be changed from "no leakage above 8.7 in NC and SFC" to "for each source which can provide current above the limits of 8.7 the equipment shall provide identifiable MOP". (side note: to avoid trouble, the exclusions for voltages below 60Vdc/30Vac should be strengthened, or preferably excluded in general and only included for special cases).

The type test should also confirm that the identified MOP is of a type that can be feasibly controlled in regular production (again emphasising the risks associated with accidental compliance in a type test). Particular attention should be given to mains related MOPs.

A back up requirement similar to the existing 8.1a would be something like:

"An identified MOP shall not rely any additional components or spacings for compliance. If necessary, spacings and components not used as a MOP can be shorted physically or theoretically in order to evaluate the identified MOP".

This is easier to understand than the existing 8.1a approach.

Hello all and thank you Perter Selvey,

I needed some time to digest the comments and a visit of a consultant.
The problem with the way your "backup requirement" is phrased is when we can identify a MOP against a certain voltage (say 15V), but if we must identify the responsible 15V DC-DC converter outputting 18V as a SFC then the MOP must be against 18V. So the MOP does rely on other components that make the secondary circuit.

To rant a little more about "ANY": a consultant advised us to pay more attention to the informative Annexes when in doubt (he says all test houses interpret ANY as a "combination of any or all" being the worst case scenario).
Well that does not make it ANY better! (pun intended)
See Annex A4 on clause 8:

The formulation proposed for the third edition of this standard is to specify:
(...)
5) that NORMAL CONDITION includes short circuit of any insulation, AIR CLEARANCE or CREEPAGE
DISTANCE or impedance which does not comply with specified requirements for the relevant
WORKING VOLTAGE , and open circuit of any earth connection which does not comply with the
requirements for PROTECTIVE EARTH CONNECTIONS ; and
6) that SINGLE FAULT CONDITIONS include short circuit of any insulation, AIR CLEARANCE or
CREEPAGE DISTANCE which does comply with specified requirements for the relevant WORKING
VOLTAGE , short circuit of any relevant component, and open circuit of any earth connection
which does comply with the requirements for PROTECTIVE EARTH CONNECTIONS

Click to expand...

So what the Annex essentialy tries to explain is that whatever you implement as a MOP (or not); you have to short circuit it in SFC considering everyone regards ANY as a combination of any or all. Which is a bit silly.
This time the French text does not make me ANY happier beause here explicitly is referred to "all"("toute"):

4) que les CONDITIONS DE PREMIER DEFAUT incluent la mise en court-circuit de toute isolation,
DISTANCE DANS L ' AIR , LIGNE DE FUITE qui est conforme aux exigences sp?cifi?es relatives ?
la TENSION DE SERVICE correspondante, la mise en court-circuit de tout composant
appropri? ainsi que le circuit ouvert de toute mise ? la terre qui est conforme aux
exigences relatives aux LIAISONS DE TERRE DE PROTECTION

Click to expand...

Is this norm a mash-up of writings by different authors? Or is it a single author who writes different interpretations on different pages?
When we are done I need a vacation, sigh..

Loek Gijben