Hi All, I really hope you can help.

I work for a transformer manufacturer, and we are currently designing a Medical Isolation Transformer to BS EN 60601. I have been given the dubious task of ensuring compliance with the standard, and I have a question concerning the output connections.

I have been told (and have read on the web), that the output voltage MUST be floating from earth (i.e the output neutral cannot be tied to earth). Is this correct? And where can I find this (rather crucial) information in the 60601 standard?

Any help would be greatly appreciated - thank you.

Of course the output “neutral” (sic) cannot be tied to earth, or it wouldn’t be isolated, now would it? But don’t just take my word for it, Mate - check out the competition, they’re even answering all your queries on-line and f.o.c. See http://reomed.com/mits.htm

There’s some nice background information available also at http://www.bender-de.com/news/downloads/Hospital-Brochure-03-2002-e.pdf

Thanks for the reply Geoff. The Bender info looks useful.

My problem is proving to our design engineers that the output must be floating - ideally by refering them to a particular clause in EN 60601.

I'm ploughing my way through the standard, but for a Mechanical Engineer, deciphering a document written by a committee of Electrical boffins is no easy task!

Hi

As the man said it wouldn't be isolated would it; to put that another way the L pin would be at 230V wrt to E and thus indistinguishable from any other live. If your "design engineers" don't understand that then ask them what they think the transformer is for.

Once upon a time our Estates dept made up some transformers for us and they found it necessary to join N to E - cos "if you didn't you couldn't tell if there was a L-E fault i.e. the cb they had fitted wouldn't trip"! I had to spend a while telling them (starting with there is no "live" or "neutral" just L1 and L2) and I am still not sure they understood even then.

Yes Grahame,

The term galvanic isolation may not mean much to them but a practical "hands-on" demonstration might, eh?

Hi morgan240469,

You haven't mentioned whether the application you intend to use your transformers in is for line insulation monitoring systems, such as those referred to by Geoff in his postings above, or for mains seperation in medical electrical systems, such as those referred to in 60601-1 and 60601-1-1. There are some subtle and important differences between these two applications that I'm not going into here. The point is that there are different standards applicable in each case.

Seeing as you refer to it and I assume it's standalone seperating transformers for use in medical sytems that you refer to; in my opinion, as far as I am aware, the BS EN60601-1 general safety standard for medical electrical equipment is applicable to medical systems, including seperating transformers incorporated into systems, that supply "isolated" mains to medical electrical equipment.

The collateral BS EN60601-1-1 standard for medical electrical equipment gives information on the requirements for distribution of the mains, etc. These two standards should also give normative references to the other applicable standards that should be adhered to in electrical systems and your designers should be fully conversant with, or at least aware of, such as those concerning industrial & consumer goods, mains distribution, transformer manufacture, electrical and mechanical installation, etc.

The mistake that a number of experienced Medical Engineers make regarding seperating (or isolation/isolating) transformers, in my experience, is that they think that mains isolation means that the isolated system no longer has a protective earth connection. Those new to medical electrical systems seem to find it difficult to come to terms with the necessity for medical equipment to operate safely under single fault conditions or the subtle differences between industrial appliance testing and those required for testing medical electrical systems.

Yes, Richard is right (as always). What is the application? Could it be, for example, as a power supply box sitting on the bottom shelf of a “video stack” cart for use in operating theatres? If so, there are already plenty of nice examples about, Mr. Morgan. Has your company really done its marketing, I wonder. What new feature are you hoping to bring to this (already crowded, in medical equipment terms) market. Not low price, I presume? Better tell your boffins that what we really want is a parallel port to USB interface. See this thread .

Morgan, permit me to be a bit basic here to start with. Apart from auto-transformers (where Primary and Secondary coils are physically connected together) power-handling transformers are “isolating” i.e. the primary is completely separate from the secondary. Besides the advantage of separating, to some extent, electrically what happens on either side of the transformer, isolation does ensure, for application in particular parts of the medical environment, that the electrical connection on the primary side between the neutral and earth wires found in non-isolated systems is not present on the secondary side. This isolation, theoretically, separates from earth the two secondary wires. This means that neither of the wires on the output of your transformer will have a “neutral” potential with respect to earth. They will now both be “live” (with respect to earth), call them line 1 and 2 if you like. Remember, a neutral in a non-isolated system is only neutral - ie not “hot” - because it is tied physically to earth. This you can confirm by measuring the voltage (potential-difference) between earth and neutral in a non-isolated system. There generally should be no more than a few hundred milli-volts between them.

So you now might be asking …if most transformers are isolating what makes the difference to us and why not just use any normal (isolating) transformer? Well the answer lies in the "degree of isolation' from the hospital earth. Extra isolation has to be designed into a medical-grade isolation transformer so that we can use it in particular clinical areas, which is the important advantage to us in the medical field where we need to keep earth-currents extra-low on patient-oriented equipment. Although I said that “theoretically” in an isolated system there is no electrical connection between the two "line' wires (on the secondary side of the isolation transformer) and the earth wire (also on the secondary side – yes we still use and need an earth connection here) that goes back from the secondary side power-socket outlet system to the main earth on the non-isolated system (on the primary side of your isolating transformer), there is still a certain amount of coupling between them. This is provided by “stray-capacitances” all over the place. So when you design your transformer for the medical environment, in order to keep earth-currents low you need to make sure that the stray capacitances between the wires and the core of the transformer are kept to the barest minimum. The reason for this is that the customer's environment will inevitably have a considerable amount of additional stray-capacitance that will be added to any inherent capacitance found in the design of your transformer. All this will reduce the effectiveness of the overall isolation and hence electrical safety for patients.

If you measure the potential difference between each line and earth (when the transformer is installed at the customer's site), you might see that each line to earth may well be around half the normal live-to-neutral reading (e.g. 230 volts) you have on the primary (non-isolated) side, i.e say a reading of around 110 volts. But equally, this reading may be totally different as it is only a “floating” voltage now since there is no hard connection to earth on the secondary side and no such thing as ideal isolation. The resulting voltages will completely depend upon the relative capacitances between each line conductor and the system earth.

Lastly, because of uncontrollable build-up of capacitance in critical areas of the hospital as more equipment is added to power sockets on the output of your transformer in the medical working environment, an isolating transformer can not provide electrical safety for us from “Micro-(amp) shock” hazard (only Macro-shock – live wire touches earth for example) but it does keep the overall leakage currents that we are concerned about lower than if we did not use the transformer.

Medical-grade isolation transformers were originally put into areas such as theatres in order to minimize the risk of an explosion of anaesthetic gases due to say an electrical spark created in the event of an equipment malfunction.

I hope this back-ground information clarifies things for you.

Read some of my earlier postings for extra information.

Hi All, thanks for the replies. A few things are now a lot clearer.

I was concerned with some the comments stating that connection of the secondary 'low' output to earth (to create a grounded neutral), meant that the transformer was no longer an isolating transformer.

My understanding (and interpretation of EN 61558) is that isolation (or separation) transformers are so called because of the galvanic isolation between the primary and secondary windings, and that this is irrespective of output earth bonding.

The company I work for supply standard Isolation Transformers (for the UK) with the output tied to earth (unless the customer specifically requests a 'floating' output). This is because many IT (computer IT - not earthing IT) applications require a neutral line which is at zero potential (wrt earth).

I know that for an IT Isolated Power System (IPS), the isolation transformer must be ungrounded as clearly stated in BS7671 and referenced in IEC 60364-7-710 Annex to MEIGaN.

And (thanks to Nick et al), I now appreciate that grounding the secondary 'low' output, means that an isolation transformer is no longer suitable for Medical applications. And I believe (my understanding is getting stretched now!) that this is because the Earth Leakage Current must be kept to a minimum for patient protection.

But I still can't find it written in the 60601 standard that the output from a isolating supply transformer must be floating (that would make my life so much easier!!).

BTW - has anybody experienced any problems attaching PCs etc to an ungrounded Medical Isolating Transformer?

Thanks again for all your comments.

Morgan,

Actually galvanic isolation (output to input) is essential to enable the seperation of the "floating" output winding on the secondary from earth-referenced input-windings on the primary of medical isolating transformers. Thus the output is floating wrt earth.

The idea is to provide a mains output that is no longer directly referenced to earth, i.e. neutral is no longer connected to earth and line is therefore no longer referenced to it. This gives no direct and much reduced capacitive return path to leakage currents via earth reducing the maginitude of earth leakage and enclosure leakage currents under normal (N) and Single fault condition (SFC) for example.

When I commented that the earth connection still exists I meant that the chassis, etc, still has a protective earth tied to the non-isolated (earth-referenced/connected) input side of the seperating transformer for electrical safety purposes.

The transformer primary side still needs to be protectively earthed and is earth-referenced if the device is class I but the mains output is "floating". As I said earlier even experienced medical engineers have problems with the concept that a device such as an isolating transformer can still have an enclosure that's earthed.

Perhaps someone at your company needs to buy-in the services of an expert in the design of medical systems.

... I must admit, Richard, that I was thinking along the same lines – I just didn’t want to be the one to say it!

If you're not convinced: Try performing the safety tests laid down in 60601-1 by testing an isolation transformer with the output winding earth-referenced (N connected to earth) and then perform the tests again for one that has no-earth reference, while both are supplying their rated power into a class I or II medical device. Under normal and SFC conditions with mains polarity forward and reversed you'll see the effect that floating the output has on leakage currents.

The design of seperating transformers is not covered in 60601-1; this standard refers to the requirements for medical electrical equipment and how seperating transformers can be connected to medical systems (60601-1-1). Incidentally, as far as I'm aware, any seperating device becomes medical equipment when it's connected to a medical device manufactured to 60601-1 as explained in 60601-1-1. Hence the combined system must pass medical electrical safety tests.

If the seperating device is intended for the medical market it's probably wise to manufacture it to comply with 60601-1 and 60601-1-1 as well as the other applicable standards for isolating transformer manufacture, e.g. 61558 and other normative standards referenced in 60601-1, 60601-1-1, etc. The requirements for things such as markings, methods of mains distribution, fusing and elecrical safety requirements for medical devices will soon become apparent if you read these standards carefully.

Up to now I have agreed with all that has been said about keeping earth and secondary windings of an isolation transformer seperated. However on a closer reading of EN 60601-1-1 I find a confusing statement in Annex EEE.
'The transformer construction with protectiveley earthed centre tapped secondary winding is allowed, but not required.'
Does not this say that connection between earth and secondary output is allowed?

I don't think anyone has stated that an earth connection to the secondary is not allowed. My concern is with the effectiveness of seperation in limiting leakage current from and between medical electrical devices that are interconnected by functional or protective earth connections and may be referenced by N being connected to earth, otherwise.

The issue of erth-referencing referred to earlier concerns the connection of N to earth. Earth referencing of a centre-tapped secondary may be allowed but it is only probably of use in mains isolation monitoring applicaations (although I'm not certain of this and certainly not an expert).

I think that's a good point and well spotted RICK. Worth having a think about.

Morgan –

Unfortunately, as is often the case, different branches of engineering use the same word but with different meanings. I think the computer cabling applications you refer to relate to signal isolation for purposes such as interference control*. Your original query was, we have all assumed, with regard to power isolation for supply to medical electrical equipment, in which case the purpose and implication of ‘isolation’ is different.

In the latter case, as others have pointed out above, ‘isolated’ implies non-earth-referenced; so don’t tie either side of the secondary to earth through a low impedance (though you would normally still provide the protective earth conductor connection at the output for bonding of accessible parts of the equipment it supplies).

Isolation transformers are principally used for 2 purposes in medical device applications. In an IT-type installation (ie: IPS) supplying a medical area (as in Wiring Regs BS7671 and Guidance Note 7), purpose is to make the supply to critical equipment more reliable (since a first fault-to-earth will not trip circuit protection devices - it merely earths the supply). The other application, (presumably the one you’re wrestling with), is the introduction of a transportable transformer between the wall supply and a particular medical device (or combination of devices) - in which case purpose is to improve electrical safety (specificaly to meet 60601-1 requirements), particularly with respect to earth leakage through the patient. For this application a separation transformer may suffice (see 60601-1-1) – I’m not familiar with the transformer standards but my understanding is that the requirements on primary-secondary insulation are less for ‘separating’ than for ‘isolating’ transformers. (Note that, in this application, both types are commonlly referred to as ‘isolating’ transformers by hospital staff).

These are safety critical applications so your engineers need to be sure they understand what they’re about before proceeding further.

(* signal isolation, or 'separation’ may also be required for safety reasons in medical systems – in which case considerations will differ to those in your computing example)

It seems (and please don't be offended), that there are many "in my opinion.." or "I think.." interpretations, and precious few references to actual applicable clauses in the 60601 standard.

As Rick has found, the standard is not 100% clear on the subject of output earthing, which leaves it open to (mis)interpretation.

I get the impression that seperation of the transformer output from earth is not an actual requirement of 60601 (hence no specific clause), but is an infered requirement necessary to comply with the low earth leakage requirement of the standard.

I will pass all of your much appreciated comments and opinions onto our transformer design engineers, who I'm sure have the competentence and ability to ensure that our unit meets the requirements of 60601.

BTW Geoff - for your USB to parallel adaptor try: http://www.tigerdirect.com/applications/SearchTools/item-details.asp?EdpNo=1219598&CatId=471
http://www.picotech.com/usb_parallel_port.html
http://www.maplin.co.uk/Free_UK_Delivery...aptor_34448.htm

Hi Ged Dean

Thanks for your response, I hadn't read your post before my last comment.

The primary/secondary isolation requirements are indeed greater for a 60601 compliant unit than they are for standard isolation transformers.

It was just the output grounding issue that was worrying me.

Thanks again to all that have commented.

Thanks Mate, but they’re all USB to parallel. What we are looking for is the other way around (ie, the elusive parallel to USB).

Meanwhile, all the hedging you refer to (“in my humble opinion” etc.) arises as (most, at least) of your correspondents do not represent any learned authority, as it were (but are just trying to help, as always). But I doubt that any medical isolation transformer with either output line tied to earth would pass muster (as they say)!

I will leave you with a parting comment, however. The manufacturer’s plate on a medical isolating transformer that I noticed yesterday stated 1000 VA, with a maximum leakage current of 58 uA.

Hi Geoff, I thought I found them a bit too easily!!

I just wanted to give something back for all the help from you guys. How about this tranceiver for bi-directional parallel ports to USB?

http://www.shopping.com/xPF-Bencole_08305

Morgan,

May I ask how you came to this conclusion - where is the degree of "isolation" specified (in the medical standards) and what other standard (RE:isolation transformers) did you make the comparison with?

I think the IEC60601 regulation is changing next year. Check what edition you will be designing, building and testing too, as this might all change.

A.M

Hi Mr Ling

For the dielectric strength tests, BS EN60601 requires a higher flash test voltage than standard double wound isolation transformers.

60601 clause 57.9
61558 clause 18

Also the creepage and clearance requirements are greater in 60601.

60601 clause 57.9
61558 clause 26

I hope this helps.

Thanks.

IEC60989:1991 Seperating transformers, autotransformers, variable transformers and reactors is mentioned as a normative reference in 60601-1-1. The "3rd" edition standard of 60601-1 is currently available to the public as a draft proposed standard for comment: 00/565498DC BS IEC 60601-1. Medical electrical equipment. General requirements for safety and essential performance .

"...But I still can't find it written in the 60601 standard that the output from a isolating supply transformer must be floating (that would make my life so much easier!!)."...

Morgan

You won´t find what you want in the body of the standard because it´s a standard for "medical electrical equipment", not a standard for "medical electrical equipment with isolating transformer". I means that the isolating transformer is an option, and it means therefore that the secondary could be earthed in principle. However, as stated before, depending on the construction of the equipment (or system, or wherever) the tests will fail if the secondary is earthed. The best example, i think, is a floating (BF or CF) applied part. This floating means floating from earth, because if it´s connected to a earthed part and you perform the "mains on applied part" leakage current test (which is the test that determines if the applied part is really floating) the leakage current will be more than the limit and therefore the test will fail. This rationale can be found in a standard, but not in IEC 60601-1. You can found it in an annex (Annex A i think) of IEC 60513 Ed 2 - Fundamental Aspects of Safety Standards for Medical Electrical Equipment.

Hi Mantunes,

Actually these comments appear very misleading to me.

I just want to say that seperating transformers are not used to correct earth faults nor limit the leakage between applied parts to all of the BF/CF requirements of 60601-1.

Hello Mr Lig

The example i cited is the only one i remenber can be traced to a normative document. Surely, there´s more examples cited in IEC closed documents (i remember some from the comments on drafts of the third edition of IEC 60601-1) but they´re not normative docs. The

Also, the statement "just want to say that seperating transformers are not used to correct earth faults nor limit the leakage between applied parts to all of the BF/CF requirements of 60601-1" is in gereral right, but separating transformers are differente from isolating transformers. Separating doens´t necessarily means isolated from earth, but isolating means separated from earth.

That is why I made the statement.

Hum, so i think i didn´t understand what you said about the comment being misleading, as my comment was on isolating transformers and not on separating ones.

If you connect a floating applied part (the bit that connects to the patient) to earth the applied part is still isolated via the instrument - the test fails because you're applying earth-referenced mains direct to to earth, not testing the isolation of the applied part.

In practice leakage would not flow through the patient via the instrument connected to that applied part. The fault-current or "leakage" is actually bypassing the applied part connected to the patient and flowing to earth - you've made an external circuit.

Otherwise you must be talking about earthing of applied parts inside the DUT - which in that case they wouldn't meet the BF or CF requirements in the first place. Where do you mean connected to earth - internally or externally?

If a device using an external isolation or seperating transformer, for that matter, is used and the applied part is earthed then tested with earth-referenced mains on the applied part test, you still end up with the same result - a fail and excessive current flowing through the connection to earth.

I don't think it was a very good example to use, that's all, irrespective of the degree of electrical isolation being discussed.

You are totally right in what you say, but that´s not what i was traying to show. I surely wasn´t clear in my statement anyway. What i was trying to say is that earthing th secondary depends on the application, in this case, in which type o applied part the equipment will have. In this case, if the manufacturer says it´s BF or CF but it´s connected to an earthed secondary the tests will fail (as you kindly pointed out). I agree that´s not the best example, but it happens a lot in my laboratory. And, besides, it´s the only that i remember that is written on a standard, and it seemed to me that´s what the original poster wanted when he talked about the "i think...' or "in my opinion.." posts.