PAT testing equipment with Inverters / Variable frequency drives.

Steve

So. I have a client who builds small machines, about 1.5 times the size of a domestic washing machine.
They consist of a control box, PLC/HMI, a couple of inverter controlled 1kW motors and the mechanical gubbins to so what it does.

Issue is this:

In order to comply with EMC regs, it has to include a RFI filter. The client chose to use an inverter with a built in filter to comply.

Problem is, no equipment (in my experience) can ever pass a PAT test if it has a RFI filter as these are, firstly, designed to leak to earth via capacitors and inductors ans secondly are required to have a discharge resistor over the capacitor which makes them fail for boat earth leakage and IR every time.

So, catch 22. Either fail RFI or fail PAT?

Anyone any experience with this?

For years I just did the test with the filters disconnected, but the recent law changes are not so clear on whether this is allowed.

brightspark

Are you sure it needs to pass PAT?

From recollection certain tests don't need to be applied as long as safety isn't compromised. (you need to test in a non-standard manner)

Is it even considered portable?

How much leakage is involved, if anywhere close to 30mA it probably can't be plugged in without tripping a RCD.

It probably needs CE certification anyway.

2011

Brightspark is correct. PAT stands for "Portable Appliance Testing", see link.

What you have described does not require a PAT testing.

By the way in terms of EMC compliance there is more to look at than filters.

Steve

It has a plug so needs to pass PAT under the new laws.

The leakage test is not compulsory, earth bonding and IR are. Leakage is about 18mA, again not a legal fail but is a PAT fail as it's over 5mA. The filter is DESIGNED to leak, this is the problem.
Also, and more of a problem is the discharge resistors in the filter. They are required (under CE) to be there to drain any charge, however they muck up the IR test as well as it sees them as an insulation breakdown..

The RFI filter is required to pass CE (machinery directive)

Steve

2011 wrote: »

What you have described does not require a PAT testing.

Can you expand on that please, not seeing it from the link.

brightspark

What defines the PAT fail of 5mA?

My understanding of PAT is to ensure the equipment is maintained as designed (The CE marking should take of any "design" issues)

If equipment cannot be tested in the normal manner, then surely it is appropriate to have alternative methods of testing that are suitable?

Steve

brightspark wrote: »

What defines the PAT fail of 5mA?

Actually that's a good point.
I need to re-look at that aspect. Thank you.

PAT test instruments seem to have this as a default fail, but I never looked into why.

My understanding of PAT is to ensure the equipment is maintained as designed (The CE marking should take of any "design" issues)

If equipment cannot be tested in the normal manner, then surely it is appropriate to have alternative methods of testing that are suitable?

Yes, and this is the preferred outcome, as long as it's legal.

brightspark

From SI 299 of 2007.

“portable equipment” means equipment, including hand-held portable equipment, which—

(a) because of the manner in which it is to be used, requires to be moved while it is working,

(b) is designed so that it can be moved while it is working, or

(c) is moved from time to time between the periods during which it is working;

Does the equipment fall under that definition?

(2) An employer shall ensure, where appropriate, that a competent person—

(a) tests any portable equipment described in paragraph (1)(c)(i) and (ii), and

(b) certifies whether or not the portable equipment (including any cables and plugs was, on the day of test, as far as could reasonably be ascertained, safe and without risk to persons coming into direct or indirect contact with any live part of the equipment.

If it does then is it safe?



However from rereading your posts it appears that you are certifying the equipment as being safe when leaving the factory? The testing would need to be more in depth with documented risk assessment of what could fail etc.


You will need to show compliance with the various I.S. and EN standards, those are not available for free online so I can't help there.

2011

brightspark wrote: »

What defines the PAT fail of 5mA?

According to this link:

Insulation Resistance – Class 1 and Class II appliances are put through this test to ascertain the effectiveness of insulation that surrounds live electrical components. An Ohm meter or a PAT meter is used to carry out the test.

The test is conducted by applying 500V d.c to the neutral and active conductors of the appliance for 5 seconds.

A 0 volt reference is located on earthed parts of class 1 equipment or on the external metal parts, in case of class 2 equipment.

Appliances with surge protectors or suppressors are likely to fail this test and need to undergo leakage current test, requiring rated voltage values to remain within 5mA and and 1mA for class I and class II equipment, respectively.

OldMrBrennan83

Can they not stick a blue 16A plug on it? That's the way I treat machinery I build. Leaves the premises with a blue 16A and if they customer so wishes they can swap it to a household 13A, then any testing that doing so requires is their baby.

2011

Have you actually PAT tested it and found the leakage current to be >5 mA??

brightspark

If the limit is 5mA then you have a fundamental design issue, if indeed the equipment is "portable".

Some information in the linked article, it mentions about >6mA but also refers to another standard for >10mA

https://electrical.theiet.org/media/1697/pat-testing-of-equipment-having-high-protective-conductive-leakage-current.pdf

brightspark

Patww79 wrote: »

Can they not stick a blue 16A plug on it? That's the way I treat machinery I build. Leaves the premises with a blue 16A and if they customer so wishes they can swap it to a household 13A, then any testing that doing so requires is their baby.

Changing the type of plug supplied used doesn't change the need for PAT or other forms of periodic inspection and as the machine supplier you would still be responsible for providing the correct installation and maintenance information. (which may include prohibiting the use of 13A plugs)

Steve

brightspark wrote: »

From SI 299 of 2007.
Does the equipment fall under that definition?

No it doesent, however, my understanding of the rules is it includes equipment that is not portable - think dishwashers, washing machines, ovens etc, that are used in either a domestic or commercial environment.

If it does then is it safe?

I have no doubts about whether it is safe, it is. It is fully machinery directive and EMC directive compliant, I'm staking my business reputation on that. The point is that by complying with these directives it cannot, by design, pass a PAT test.
I could easily change the spec to not include a RFI filter and probably nobody would ever have cause to test and fail it for that but that's cheating...

However from rereading your posts it appears that you are certifying the equipment as being safe when leaving the factory? The testing would need to be more in depth with documented risk assessment of what could fail etc.


You will need to show compliance with the various I.S. and EN standards, those are not available for free online so I can't help there.

Yes of course, this is one small part of the overall testing.

2011 wrote: »

Have you actually PAT tested it and found the leakage current to be >5 mA??

Yes, 17mA.

IR test also fails as it is below 1MOhm. With the filter out, leakage is <1 mA and IR is >20MOhm.

Steve

One of the documents I have found is here - though it refers to UK rules, I believe they are equivalent.

https://www.qub.ac.uk/safety-reps/sr_webpages/safety_downloads/PATesting.pdf

The definitions of appliances go beyond 'portable' and include:

Hand-held Class I and II
Portable Class I and II
Moveable Class I and II
Stationary Class I and II
IT Class I and II

I have further definitions for these groups somewhere, will try to dig them out. Specifies weight, portability etc.

brightspark

Steve wrote: »

I have no doubts about whether it is safe, it is.

When I said is it safe I was referring to the part or the law describing the aims of the periodic testing. It wasn't meant as a reference to the safety of the equipment ex-factory.

Steve wrote: »

It is fully machinery directive and EMC directive compliant, I'm staking my business reputation on that. The point is that by complying with these directives it cannot, by design, pass a PAT test.
I could easily change the spec to not include a RFI filter and probably nobody would ever have cause to test and fail it for that but that's cheating...

Are there alternative RFI filters etc. that would be more suitable for PAT?

At 17mA leakage current it will be very prone to nuisance RCD tripping

If the equipment will have to be subject to PAT, then is it appropriately designed?

Alternatively should the equipment be sold as non-portable requiring fixed wiring etc., that may change the testing parameters?

The link I referenced appeared to prohibit 13A plugs at just >6mA current leakage levels, different installation and testing requirements probably apply to your equipment.

Steve

brightspark wrote: »

When I said is it safe I was referring to the part or the law describing the aims of the periodic testing. It wasn't meant as a reference to the safety of the equipment ex-factory.

Yes, agreed. Not what I was asking either.

Are there alternative RFI filters etc. that would be more suitable for PAT?

This is something I would be grateful for anyone with experience on the matter to come forward with, i.e. the point of me posting this thread.

At 17mA leakage current it will be very prone to nuisance RCD tripping

In general this is not a prevalent issue (i.e customer complaint)

If the equipment will have to be subject to PAT, then is it appropriately designed?

An open question... no answer I'm afraid right now.

Alternatively should the equipment be sold as non-portable requiring fixed wiring etc., that may change the testing parameters?

I'm not sure that simply declaring it as non portable covers it.

See the attached, which is the signoff that I've use for the last few years.

brightspark

Are there alternative RFI filters etc. that would be more suitable for PAT?

This is something I would be grateful for anyone with experience on the matter to come forward with, i.e. the point of me posting this thread.

I think you would need to give the specs of the drive and filter for anyone to be able to suggest alternatives.

At 17mA leakage current it will be very prone to nuisance RCD tripping

In general this is not a prevalent issue (i.e customer complaint)

It does mean that you can only install one machine on each RCD circuit! Also from experience sometimes it means another separate supply usually has to be run to a new socket adjacent to the machine.

Alternatively should the equipment be sold as non-portable requiring fixed wiring etc., that may change the testing parameters?

I'm not sure that simply declaring it as non portable covers it.

It obviously wouldn't be that simple, at a minimum I expect it would involve labeling and a manual stating correct installation and periodic testing requirements.

Steve

brightspark wrote: »

I think you would need to give the specs of the drive and filter for anyone to be able to suggest alternatives.

It does mean that you can only install one machine on each RCD circuit! Also from experience sometimes it means another separate supply usually has to be run to a new socket adjacent to the machine.

It obviously wouldn't be that simple, at a minimum I expect it would involve labeling and a manual stating correct installation and periodic testing requirements.

Don't want to go to far from the original question which, in essence, now is how do I either get the equipment through a PAT test or find out if there is an alternative.

I'll PM you the inverter spec rather than posting here, would be grateful if you had an opinion.

Thanks for all everyone's input, much appreciated .

brightspark

Steve wrote: »

Don't want to go to far from the original question which, in essence, now is how do I either get the equipment through a PAT test or find out if there is an alternative.

From what has been said here to get it to pass the PAT you will have to reduce the leakage below 5mA.

The alternative is being able to prove a testing and installation method that is proven to be safe. If it was a larger piece of equipment with multiple inverters etc., then the cummulative leakage makes a standard PAT impractical, but nevertheless all equipment needs periodic testing.

I feel this is must come down to equipment standards (IS or EN) demonstrated compliance with them would then determine if you should be aiming to reduce your leakage levels to a level where standard testing applies or else to have a documented test procedure that gives expected ranges of test results etc.

As to the spec of inverter you are currently using, perhaps it might be worthwhile discussing it with one of the larger drive suppliers (e.g. Lenze) who would have a more comprehensive knowledge of the options available.(it's likely they have had to advise machine builders about this before)

Steve

Update:

Only thing I can come up with to solve this is to remove the RFI filtering from the appliance. This will ensure PAT test passes.

EMC compliance will have to be dealt with separately, however, the likes of this (or similar) should ensure both requirements are met.

2011

I think the solution is simple. Let’s take it as a given that the VSD and RFI filter are safe, fit for purpose and have the required certification to back this up. This is the case regardless of whether these units are installed in a portable skid or part of the fixed wiring of an installation.

Having accepted this PAT test the skid with the RFI filter disconnected. Next retest with it connected. Insert both test results in the handover pack with documentation to support the fact that the leakage current of the RFI filter is within safe limits and that this is simply due to the nature of the device. Don’t dwell on it anymore and move on.

Steve

2011 wrote: »

I think the solution is simple. Let’s take it as a given that the VSD and RFI filter are safe, fit for purpose and have the required certification to back this up. This is the case regardless of whether these units are installed in a portable skid or part of the fixed wiring of an installation.

Having accepted this PAT test the skid with the RFI filter disconnected. Next retest with it connected. Insert both test results in the handover pack with documentation to support the fact that the leakage current of the RFI filter is within safe limits and that this is simply due to the nature of the device. Don’t dwell on it anymore and move on.

I had thought about doing that.

The problem remains that the end user is required to periodically test the equipment. The tester won't necessarily have access to that information and will (correctly) fail the equipment.

I don't see any other approach other than to have the RFI filtering upstream of the socket.

From my, and my clients point of view, this admittedly makes the RFI issue 'somebody else's problem' but I don't see a workable alternative that can satisfy all of the requirements.

2011

Steve wrote: »

The problem remains that the end user is required to periodically test the equipment. The tester won't necessarily have access to that information and will (correctly) fail the equipment.

Which is precisely why its should be supplied with a handover pack like most skids. This pack would normally contain schematics, spare parts list, maintenance details and information such as this. Besides they can always contact you in the event of an issue.

From my, and my clients point of view, this admittedly makes the RFI issue 'somebody else's problem' but I don't see a workable alternative that can satisfy all of the requirements.

Not an issue, see above. You are not the first person to put a VSD with an RFI filter in a skid and you won't be the last.

Tuco88

Would it be worth putting trefolyte label or something in or on the item, explaining such in brief?

Might save the hassel of getting 100s of calls. Unfortunately not everbody enjoys reading the manual/Hand over pack.

2011

Why not? It couldn't hurt.