IET logo
 
IET
Decrease font size
Increase font size
Topic Title: PAT testing class II equipment - Time for a change?
Topic Summary:
Created On: 24 July 2009 11:34 AM
Status: Post and Reply
Linear : Threading : Single : Branch
1 2 Next Last unread
Search Topic Search Topic
Topic Tools Topic Tools
View similar topics View similar topics
View topic in raw text format. Print this topic.
 24 July 2009 11:34 AM
User is offline View Users Profile Print this message



MikeX

Posts: 21
Joined: 25 July 2008

Currently the guidance on PAT testing does not require any testing to be carried out only a visual inspection. With the increased use of switch mode power supplies in items such as laptops and phone chargers etc. there is an increased risk from the effect of leakage currents in the case of a fault.

In fact there is strong evidence that many poor quality or cheap copies of products, that do not conform to the required safety requirements, are entering the market that have increased the possibility of electric shock due to insulation failure.

One problem is also that many do not know how to test a class II device for leakage (or test insulation) correctly. Also the use of a DC insulation test may not provide a true indication of the units leakage in the case of a switch mode power supply. The best test, which is also the one less likely to cause the unit damage, is a direct leakage test as long as it is correctly performed.

If you want to give your self a surprise measure the voltage, with a DVM set to AC volts, between earth (use the mains socket earth) and the metal output connector of a class II laptop power supply (the pin that plugs into the laptop). The voltage you measure is correct and normal for class II products that use a switch mode power supplies! Ok if you now switch your meter to mA you should measure less than 250uA.

So how many think we need a change to the guidance for PAT testing?
 26 July 2009 12:32 PM
User is offline View Users Profile Print this message



MikeX

Posts: 21
Joined: 25 July 2008

No one have an opinion on this topic?

Anyone bothered to have a go at the test I mentioned on a laptop (or similar) switch mode power supply? It you did I am sure someone would have something to say!
 26 July 2009 12:38 PM
User is offline View Users Profile Print this message



AJJewsbury

Posts: 11461
Joined: 13 August 2003

I've not tried it, but using a high impedance meter, I'd expect it to be about half mains voltage, just due to capacitive coupling and the fact it's not referenced to any external potential until you connect the meter to it (would that be about right?).

We see very similar things with switch lines to light with the switch turned off - sometimes the capacitive coupling is enough to make compact fluorescent lamps flicker!

How would you propose that this 'direct leakage test' work? Would you have to deliberately earth the load site to provide a path for the leakage current?

- Andy.
 26 July 2009 12:57 PM
User is offline View Users Profile Print this message



MikeX

Posts: 21
Joined: 25 July 2008

Of course you are right about measuring about half the mains due to the capacitive effects from both the EMC filters required by switch mode power supplies and the small transformers used due to the high switching frequency. Both lead to a lot more leakage than a power supply using a conventional mains transformer. A DC insulation test will not reflect the true leakage at AC and so may not detect a device with excessive (dangerous) leakage. Do you feel this requires a change in the guidance and test procedures as I do? Remember if you have a class II laptop connected to a class II printer connected to a class II router connected to a class II monitor the sum of all the leakage will add up and travel through anyone touching a metal part (and earth).

The test is easily performed using most current PAT testers as you just measure touch leakage using the probe to touch any exposed metal part on the class II device (metal plug of output connecter etc.). After all if you can touch it and the current is safe for you then the tester can't cause any damage!
 26 July 2009 06:08 PM
User is offline View Users Profile Print this message


Avatar for gkenyon.
gkenyon

Posts: 4478
Joined: 06 May 2002

Originally posted by: MikeX

Currently the guidance on PAT testing does not require any testing to be carried out only a visual inspection. With the increased use of switch mode power supplies in items such as laptops and phone chargers etc. there is an increased risk from the effect of leakage currents in the case of a fault.



In fact there is strong evidence that many poor quality or cheap copies of products, that do not conform to the required safety requirements, are entering the market that have increased the possibility of electric shock due to insulation failure.
Why is this a problem for regular "PAT-Test" - surely, there are two areas other than "PAT Test":

(a) Protection of the public for domestic equipment - a question for Trading Standards.

(b) A procurement and bringing into service issue for a commercial/industrial customer, under PUWER and other Regulations, for equipment for use at work.

In either case, under Consumer Protection and H&S @ W legislation, it's the manufacturer who's MAINLY culpable. Are you suggesting there's a gap in the legislation ?



One problem is also that many do not know how to test a class II device for leakage (or test insulation) correctly. Also the use of a DC insulation test may not provide a true indication of the units leakage in the case of a switch mode power supply. The best test, which is also the one less likely to cause the unit damage, is a direct leakage test as long as it is correctly performed.
This is a "competence and training" issue under Electricity @ Work Regs?

If you want to give your self a surprise measure the voltage, with a DVM set to AC volts, between earth (use the mains socket earth) and the metal output connector of a class II laptop power supply (the pin that plugs into the laptop). The voltage you measure is correct and normal for class II products that use a switch mode power supplies! Ok if you now switch your meter to mA you should measure less than 250uA.
What I think is being overlooked here, is the situation when a number of these devices are connected via unearthed signalling/cabling systems. But again, it's a problem for either the manufacturer (need to provide instructions to ensure safe installation and use), OR the designer of the fixed cabling system (design under CDM Regs for safe use . . .).

So how many think we need a change to the guidance for PAT testing?


So overall, I voted "No", as there's insiffucient evidence in the original post, that regular "in-service" tests are the issue here.

Seems to be inappropriate procurement/supply chain, or inappropriate design issue.

-------------------------
Eur Ing Graham Kenyon CEng MIET TechIOSH
 27 July 2009 05:54 PM
User is offline View Users Profile Print this message



MikeX

Posts: 21
Joined: 25 July 2008

If we took the line that nothing dangerous could ever enter the workplace and nothing ever went wrong then the need for any testing, other than a simple visual inspection, is all that would be required for all PAT testing!

The problem with the cheap imitations is they are sometimes impossible to identify as they are visually identical to the original manufactures device. Devices enter the workplace via many paths these days so it would be impossible to stop a 'cloned' device just by controlling the procurement path.

The connection of multiple class II devices where there is no earth is a common situation and normally the combined leakage will be below acceptable limits so as not to cause concern, unless one device has excessive leakage when it would be nice to know before anyone gets hurt. After all we conduct a regular insulation (or leakage) test on earthed class I devices to ensure, in the event of an earth fault, the leakage does not exceed a safe limit. Why do this test if there is no possibility of the leakage changing from when it was manufactured?

In the past class II devices were generally safe as they used a basic 50Hz mains transformer for their isolation, which are fairly trusted and safe. The introduction of the small and compact switch mode power supplies has created the need for a change to the current guidance on testing as there danger is increased with their increased complexity and smaller size. A simple comparison of the leakage of a transformer based PSU against a switch mode PSU is all the indication you should need for increased safety testing. (Generally about 100 times grater.)

A change in the testing requirements would also mean reviewing competency of those conducting any PAT testing and a review of the IET code of practice to aid in the training and understanding in these matters.
 27 July 2009 09:17 PM
User is offline View Users Profile Print this message



Patnik

Posts: 171
Joined: 25 June 2008

I mostly agree with MikeX. There seems to be a steady stream of substandard phone chargers and the like appearing on the market and yet there is little guidance on how to recognise them. I notice a charger sold by ASDA with one of the increasingly popular interchangeable plug adaptors is currently being recalled as the adaptor can be plugged in on it's own risking shock by direct contact. It's too late for trading standards to deal with this sort of thing so I have to try and keep myself informed and look out for such appliances. Then there's the rash of foreign stuff sold with unsuitable travel adaptors. And lots of CE marked appliances that have no makers ID and have not so obvious problems like insufficient plastic between the pins and the edge of the case leading to pins snapping off in the socket etc.

Neither the CoP or many PAT training courses (including C&G ones) even touch on these matters and some PAT Techs don't seem to have a clue what to look for.

Whilst I usually do a leakage test with my clip on the SELV connector I don't think it is particularly useful. You do indeed get some pretty high volt readings between the SELV and supply earth on many of these appliances but this doesn't often correspond to getting any discernible shock so not sure how useful a new test would be.

I vote yes though - we need more up to date guidance. Trading Standards need to be a lot more more proactive as well and start policing ebay and the like.

Nik
 27 July 2009 09:24 PM
User is offline View Users Profile Print this message


Avatar for gkenyon.
gkenyon

Posts: 4478
Joined: 06 May 2002

Originally posted by: MikeX

If we took the line that nothing dangerous could ever enter the workplace and nothing ever went wrong then the need for any testing, other than a simple visual inspection, is all that would be required for all PAT testing!



The problem with the cheap imitations is they are sometimes impossible to identify as they are visually identical to the original manufactures device. Devices enter the workplace via many paths these days so it would be impossible to stop a 'cloned' device just by controlling the procurement path.
This is still a supply-chain issue. If it's cheaper, then why? If your supply-chain manager is only interesting in saving, then they aren't doing their job.



The connection of multiple class II devices where there is no earth is a common situation and normally the combined leakage will be below acceptable limits so as not to cause concern, unless one device has excessive leakage when it would be nice to know before anyone gets hurt. After all we conduct a regular insulation (or leakage) test on earthed class I devices to ensure, in the event of an earth fault, the leakage does not exceed a safe limit. Why do this test if there is no possibility of the leakage changing from when it was manufactured?
I disagree with the logic here. First, there's a difference between "leakage" and "protective conductor current". Second, with Class I, the earth continuity test is the one to conduct - because the protective devices can operate if the insulation fails. Not so with Class II.

With Class II, inspection is most important IF the device is constructed correctly.



In the past class II devices were generally safe as they used a basic 50Hz mains transformer for their isolation, which are fairly trusted and safe. The introduction of the small and compact switch mode power supplies has created the need for a change to the current guidance on testing as there danger is increased with their increased complexity and smaller size. A simple comparison of the leakage of a transformer based PSU against a switch mode PSU is all the indication you should need for increased safety testing. (Generally about 100 times grater.)
This is a "red herring" if the PSU is constructed to meet the requirements of accepted standards



A change in the testing requirements would also mean reviewing competency of those conducting any PAT testing and a review of the IET code of practice to aid in the training and understanding in these matters.
It is the Employer's responsibility to assess the competence of people carrying out PAT Test, and indeed Supply Chain/Procurement, roles.

I don't think it's the resposibility of the IET to address failures by manufacturers and employers, by providing "guidance on testing", when what's needed is "guidance on procurement" and "guidance on assessing competence": indeed, if we neglect the latter issues in preference to the former, we're putting ourselves in a world, where suppliers and employers can do as they please, and look to Institutions to bail them out when they are found out.

-------------------------
Eur Ing Graham Kenyon CEng MIET TechIOSH
 27 July 2009 09:29 PM
User is offline View Users Profile Print this message


Avatar for gkenyon.
gkenyon

Posts: 4478
Joined: 06 May 2002

Originally posted by: Patnik

I mostly agree with MikeX. There seems to be a steady stream of substandard phone chargers and the like appearing on the market and yet there is little guidance on how to recognise them.
I agree wholeheartedly. And I've brought these issues to the attention of TS myself on a number of occasions - don't even get a response.

But let's fact it, the IET CoP is for "In-Service Inspection and Test", NOT guidance on procurement of safe product. All product sold in the EU SHOULD be "safe" (fit for purpose at least), and manufacturers/suppliers SHOULD be held to account.

There is NOT a need for a "supplementary test industry" to make up for failings of manufacturers. People need to understand an old Lancashire saying "Tha gets what tha pays fer". If something is "cheap", ask "why?" !

-------------------------
Eur Ing Graham Kenyon CEng MIET TechIOSH
 27 July 2009 10:44 PM
User is offline View Users Profile Print this message


Avatar for ebee.
ebee

Posts: 5698
Joined: 02 December 2004

As A "Norf" I concour and as a Lancashire one 100%.

We used to be accused of over engineering things, nowadays phone chargers make me cringe, the wires are like limp cotton.


Lots of stuff is a lot cheaper nowadays in real terms - hand tools and power tools for a start and that is good so far as it goes but some of the cheap stuff we see ! well it worries me

-------------------------
Regards,
Ebee (M I S P N)

Knotted cables cause Lumpy Lektrik
 28 July 2009 06:52 AM
User is offline View Users Profile Print this message



Patnik

Posts: 171
Joined: 25 June 2008

Originally posted by: gkenyon

Originally posted by: Patnik



I mostly agree with MikeX. There seems to be a steady stream of substandard phone chargers and the like appearing on the market and yet there is little guidance on how to recognise them.
I agree wholeheartedly. And I've brought these issues to the attention of TS myself on a number of occasions - don't even get a response.



But let's fact it, the IET CoP is for "In-Service Inspection and Test", NOT guidance on procurement of safe product. All product sold in the EU SHOULD be "safe" (fit for purpose at least), and manufacturers/suppliers SHOULD be held to account.



There is NOT a need for a "supplementary test industry" to make up for failings of manufacturers. People need to understand an old Lancashire saying "Tha gets what tha pays fer". If something is "cheap", ask "why?" !


While this is undoubtedly true in reality the PAT tester is still left in the position of dealing with the substandard appliance often annoying the owner in the process.

So, as when an electrician gets the unenviable job of telling a customer that the previous chap who worked on their wiring left it unsafe before disappearing with the money and it'll have to be redone, PAT has to say 'sorry you can't plug that in here even if you did pay £4.99 for it'. But at least the electrician can point to the regs to illustrate the point, we don't have any such rule book for support or guidance.

Nik
 28 July 2009 08:11 AM
User is offline View Users Profile Print this message


Avatar for gkenyon.
gkenyon

Posts: 4478
Joined: 06 May 2002

Originally posted by: Patnik

So, as when an electrician gets the unenviable job of telling a customer that the previous chap who worked on their wiring left it unsafe before disappearing with the money and it'll have to be redone, PAT has to say 'sorry you can't plug that in here even if you did pay £4.99 for it'. But at least the electrician can point to the regs to illustrate the point, we don't have any such rule book for support or guidance.
We do - the Harmonised Product Standards, e.g. BS EN60335-series for Domestic & similar appliances, BS EN60950-series for IT equipment between then cover most of the appliances.

But think on this. As a "PAT-Tester":

- You can't see everything inside an appliance during an in-service inspection & test.

- You can't carry out the extensive type-tests that the manufacturer SHOULD have addressed.

- You can't prescribe "safe installation and use instructions" based on the design of the product.


In other words, you can't cover off the problems I highlighted in my previous post, which are the real problem. Let's not try and cover up the failings in the system by introducing an "after the fact" set of tests and guidance that won't identify all the potential problems. Still leaves a big hole, you see?

-------------------------
Eur Ing Graham Kenyon CEng MIET TechIOSH
 28 July 2009 10:40 AM
User is offline View Users Profile Print this message



Patnik

Posts: 171
Joined: 25 June 2008

Originally posted by: gkenyon

... Still leaves a big hole, you see?


Yes indeed - it leaves a great big hole that shouldn't be there. I have to try and fill it the best I can though as educating our engineers and my boss in even the basic requirements of safe electrical construction is proving to be a long job

I am constantly trying to get over the same point you are trying to make. Before I can PAT test it I need to know it is built to a safe standard. It's slow progress but surely better than just ignoring sub-standard appliances. (Not PAT testing an item doesn't stop them using it - I have to FAIL it to get the message over.)

The real world of the workplace is often a long way from where it ought to be

Nik
 28 July 2009 12:00 PM
User is offline View Users Profile Print this message



MikeX

Posts: 21
Joined: 25 July 2008

Graham, we all know electrical equipment is designed and manufactured to an appropriate type test standard and is therefore 'safe' to use.

The real question you seem to raise is the need for any electrical safety testing on a periodic basis at all, apart from a visual inspection and checking the earth bonding of a class I product. After all if a class I product has been correctly designed then why conduct an insulation or leakage test as indicated in the IET guidance on PAT testing?

To me it would only seem reasonable to apply the same guidance for all products unless past experience had shown a need for extended testing, thus I assume the need to check insulation on class I devices. If this were true then without testing how will any data be collected on the state of the newer class II devices that contain switch mode power supplies?

Maybe a class II device that sits in an office all it's life will remain safe but can the same be said of a laptop power supply being transported around constantly being exposed to vibrations and drop forces? Maybe a visual inspection would indicate a cracked case but would it show an internal component that had worked lose or a fractured ferrite core on the isolation transformer?

Medical equipment is type tested to the very stringent IEC 60601 standard, yet these devices must be tested annually with a full electrical test in accordance with IEC 62353 (BS EN 62353). If devices were so reliable then these tests would only include a visual inspection and earth test if appropriate. But in fact the tests include both the mains leakage (equipment) and any applied parts leakage tests (applied parts are those parts directly in contact with the patient).

Nice to see this topic has raised some discussion points, it started off so quietly!
 28 July 2009 12:37 PM
User is offline View Users Profile Print this message


Avatar for gkenyon.
gkenyon

Posts: 4478
Joined: 06 May 2002

Originally posted by: MikeX

Graham, we all know electrical equipment is designed and manufactured to an appropriate type test standard and is therefore 'safe' to use.



The real question you seem to raise is the need for any electrical safety testing on a periodic basis at all, apart from a visual inspection and checking the earth bonding of a class I product.
Potentially - you see, the tests you talk about don't cover all failures of insulation on Class II product. You'd need to attach connections from all data ports to a common ground before conducting the leakage test, for example - is this "practicable"???

(The "leakage" of a Class II laptop or printer may well be different when a network cable is attached.)

After all if a class I product has been correctly designed then why conduct an insulation or leakage test as indicated in the IET guidance on PAT testing?
Prevention of fire.



To me it would only seem reasonable to apply the same guidance for all products unless past experience had shown a need for extended testing, thus I assume the need to check insulation on class I devices. If this were true then without testing how will any data be collected on the state of the newer class II devices that contain switch mode power supplies?
Medical equipment is type tested to the very stringent IEC 60601 standard, yet these devices must be tested annually with a full electrical test in accordance with IEC 62353 (BS EN 62353). If devices were so reliable then these tests would only include a visual inspection and earth test if appropriate. But in fact the tests include both the mains leakage (equipment) and any applied parts leakage tests (applied parts are those parts directly in contact with the patient).
Yes - but the risk is greater, as the patient is in contact with the medical equipment all the time.



Nice to see this topic has raised some discussion points, it started off so quietly!
Well, I do agree with your sentiment, if not the actual consequences.

-------------------------
Eur Ing Graham Kenyon CEng MIET TechIOSH
 28 July 2009 12:47 PM
User is offline View Users Profile Print this message



MikeX

Posts: 21
Joined: 25 July 2008

Why should an insulation test of a class I device be needed for the prevention of fire (and not be applicable for a class II device)?

In the case of a L-N overload the fuse will blow, in the case on a L-E fault the fuse will blow (assuming the earth is in tact and the mains earth fault loop impedance is low enough). An insulation or leakage test will not indicate anything is about to combust, only that you might get an electric shock in the case of an earth failure!
 28 July 2009 05:52 PM
User is offline View Users Profile Print this message


Avatar for gkenyon.
gkenyon

Posts: 4478
Joined: 06 May 2002

Originally posted by: MikeX

Why should an insulation test of a class I device be needed for the prevention of fire (and not be applicable for a class II device)?
No-one said the Insulation Test was not applicable to Class II devices - but the effectiveness of what you're actually testing is an issue - as I illustrated. With Class I, you can get some "realistic" results with "leakage measurement" and Insulation test.

I Reiterate, to make it simple (although, to be honest, I'm paraphrasing the IEE CoP really): Inspection is most important for both Class I and Class II - more important than testing. Earth test is essential for Class I. Insulation "leakage" tests are (to be honest) subjective for Class II, particularly IT equipment with lots of data ports: if you've lots of leakage on a stand-alone test, you know you've got a problem, but if you've not a lot, are you testing correctly, and how do you know?



In the case of a L-N overload the fuse will blow, in the case on a L-E fault the fuse will blow (assuming the earth is in tact and the mains earth fault loop impedance is low enough). An insulation or leakage test will not indicate anything is about to combust, only that you might get an electric shock in the case of an earth failure!
If it leaks more now that it did last time, then you know it's "on the way out" - that was the theory in any case !

But I agree, in real life, you often tend to get a rapid break-down of insulation/isolation, for one reason or another (particularly in response to Overvoltage on the supply).

-------------------------
Eur Ing Graham Kenyon CEng MIET TechIOSH
 29 July 2009 03:40 PM
User is offline View Users Profile Print this message



MikeX

Posts: 21
Joined: 25 July 2008

Lets take the typical example of a device that frequently will fail a leakage or insulation test, a domestic steam iron. Here, due to the combination of water and metal, over time the electric elements insulation may be degraded so increasing the leakage current. If your lucky the RCD in your consumer unit will be the first to inform you that you have a problem! If not you have to hope the fuse blows before the earth in the mains lead breaks! Here a PAT test that included a leakage/insulation test may detect the problem before it can become dangerous.

Now lets take a class II device such as a laptop power supply. Lets say it gets a cup of tea spilt over it (while off) and the user just wipes it off and assumes all is OK. But now the liquid may have ingressed and coated the circuit board, drying to a resistive layer. This may not result in the failure of the power supply but may well have increased the leakage! No visual inspection will tell you this only a leakage (or insulation) test.

So far there is a 50/50 split on the votes but either way a change is needed to ensure we have a consistent and ideally safe test procedure.

In Germany (VDE 0701 & VDE 0702) all class I & class II equipment is tested for insulation and leakage during routine safety testing. Therefore there is no reason we can't do this testing.

Some seem to have problems obtaining 'realistic' results when conducting leakage measurements but in most cases this is down to not having enough (and correct) training in how to perform the tests correctly. Normally it is just a matter of plugging the item to be tested into the PAT tester and use the fly-lead probe to contact any exposed metal parts the user could come into contact with. The metal barrel of the laptop power supply being a good example, or the metal contacts of a battery charger. If using the differential method only the leakage of the device under test will be measured and not other items that may be connected via data cables etc.
 29 July 2009 06:07 PM
User is offline View Users Profile Print this message


Avatar for gkenyon.
gkenyon

Posts: 4478
Joined: 06 May 2002

Originally posted by: MikeX

Lets take the typical example of a device that frequently will fail a leakage or insulation test, a domestic steam iron. Here, due to the combination of water and metal, over time the electric elements insulation may be degraded so increasing the leakage current. If your lucky the RCD in your consumer unit will be the first to inform you that you have a problem! If not you have to hope the fuse blows before the earth in the mains lead breaks! Here a PAT test that included a leakage/insulation test may detect the problem before it can become dangerous.
Except there's no requirement to PAT-Test a DOMESTIC steam-iron in its normal ("DOMESTIC") use . . . but I agree.

NOTE: CLass I insulation test goes SOME way to checking FUNCTIONAL INSULATION as well as BASIC INSULATION. So, it's a better indication of degradation. Class II & Reinforced appliances you can't always do a meaningful test of all functional and basic insulation in the appliance - so you can't check for the fire risk.



Now lets take a class II device such as a laptop power supply. Lets say it gets a cup of tea spilt over it (while off) and the user just wipes it off and assumes all is OK. But now the liquid may have ingressed and coated the circuit board, drying to a resistive layer. This may not result in the failure of the power supply but may well have increased the leakage! No visual inspection will tell you this only a leakage (or insulation) test.
But this is something other than your orignal statement:

QUOTE: "In fact there is strong evidence that many poor quality or cheap copies of products, that do not conform to the required safety requirements, are entering the market that have increased the possibility of electric shock due to insulation failure."

Cup of Tea spilled on the device does not amount to "Insulation Failure" - it's "mis-use".



So far there is a 50/50 split on the votes but either way a change is needed to ensure we have a consistent and ideally safe test procedure.



In Germany (VDE 0701 & VDE 0702) all class I & class II equipment is tested for insulation and leakage during routine safety testing. Therefore there is no reason we can't do this testing.
There's no reason not to do the test, you're right.

But:

1. It still won't always pick up a potential problem on Class II device.

2. Don't forget "interface ports" change your potential for "leakage". Test method very important (competence of tester and knowledge of how products are made very important here - need Product Standards and Domain Knowledge, not additional IET guidance).



Some seem to have problems obtaining 'realistic' results when conducting leakage measurements but in most cases this is down to not having enough (and correct) training in how to perform the tests correctly.
So change the training.

Normally it is just a matter of plugging the item to be tested into the PAT tester and use the fly-lead probe to contact any exposed metal parts the user could come into contact with. The metal barrel of the laptop power supply being a good example, or the metal contacts of a battery charger. If using the differential method only the leakage of the device under test will be measured and not other items that may be connected via data cables etc.


I still don't see where you're going with this.

To properly test a Class II appliance's insulation using "leakage method", you'd have to wrap it all in foil connected to earth.

Take a phone charger, or other CLass II SELV plugg-in supply "transformer". There may well be no metal parts for the user to contact, except the outut power plug.

So, we do insulation and leakage tests to the power plug - everything appears OK, so we "Pass" it, because the SMPSU transformer and isolating ICs etc. are all OK.

But what we're not testing for, is performance of the insulating parts of the PSU that the user will touch more frequently, OR performance of the FUNCTIONAL and BASIC Insulation.

The appliance could therefore still burst into flames the moment you leave it !

-------------------------
Eur Ing Graham Kenyon CEng MIET TechIOSH
 29 July 2009 07:13 PM
User is offline View Users Profile Print this message



MikeX

Posts: 21
Joined: 25 July 2008

Originally posted by: gkenyon

To properly test a Class II appliance's insulation using "leakage method", you'd have to wrap it all in foil connected to earth.

Take a phone charger, or other CLass II SELV plugg-in supply "transformer". There may well be no metal parts for the user to contact, except the outut power plug.


You can test the leakage of almost any class II device as long as you can touch a conductive part that forms part of the secondary circuit.

In the case of a plug top phone charger you only have to touch the test probe onto any of the output connector pins to be able to measure the leakage.

It goes back to my initial test were I asked members to try connecting a meter, set to mA, between a good earth and the output (secondary) of any switch mode power supply (just touch the +ve or -ve output pins) and you will be measuring the true leakage of the device (expect between 60uA and 120uA typically). The wrapping in foil is only really needed when doing a type test!

As I said, it is not just a question of should we be doing the test but also the need to ensure those conducting tests to have the competence to do it correctly.

I still maintain if it is valid to conduct insulation (leakage) measurements on class I devices it is equally valid to conduct them on class II as well. Given the proliferation of complex and miniaturised switch mode power supplies it seems only prudent to update the testing recommendations?
Statistics

See Also:



FuseTalk Standard Edition v3.2 - © 1999-2014 FuseTalk Inc. All rights reserved.