Washing Machines Failing PAT tests on High leakage

[sparkticus]

These are NEW machines! And some secondhand ones too.

This is the entire stock of a washing machine hire company - so far none have passed.

Different brand of washing machine.

Three different testers have been used (all of which are "in calibration")

Martin, if you perform a standard insulation test at 500V what insulation resistance reading so you get?

 

[bernardgreen]

Is the PAT leakage test performed with AC or DC applied ?

If it is AC then the "leakage" is almost certain to be current through the capacitors between live and earth in the filters on the electrical input to the machine. This is normal. Prove it by disconnecting the filter during the test.

If it is DC and there is still significant / excessive leakage then look around the motor brush holders for carbon dust build up creating tracks for leakage from brush to frame.

 

[JohnW2]

Is the PAT leakage test performed with AC or DC applied ? If it is AC then the "leakage" is almost certain to be current through the capacitors between live and earth in the filters on the electrical input to the machine. This is normal. Prove it by disconnecting the filter during the test.

That is clearly the crucial question - but it raises other ones. As you say, to get a measurable 'leakage current' with AC testing would be normal/expected if there were a capacitor between L & E. However (and this is something I know nothing about), the OP seems to imply that the PAT testing rules impose a limitationon acceptable 'leakage current'. If the testing is undertaken with AC, does this mean that it is intended that appliances with such capacitors (above a certain value of capacitance) automatically fail PAT tests?

On the other hand, if the test is undertaken with DC, then it is no different (except possibly in terms of voltage) from a 'standard' IR test - does PAT testing actually require both and, if so, why? The fact that this would represent senseless duplication suggests to me that maybe the leakage test is performed with AC and the IR test with DC.

Kind Regards, John.

 

[ericmark]

There are a number of issues here.

The washing machine normally incorporates a timer and the individual motors and heaters operate as time passes in the cycle so to test the machine requires one to do some dismantling in order to test the individual parts and they can't be tested using a standard PAT testing machine.

In the main the method of showing these types of machine are satisfactory is to take out a maintenance contract as of course no electrician is going to want to dismantle an appliance he is not familiar with.

The guy who works with these types of machines will have the maintenance manual and one would hope should know what is acceptable for each part within the machine. Likely there will be some power factor correction and filters included and this is likely to cause some leakage.

However it still must comply in general with BS7671:2008 which states:-

543.7.1.1 Equipment having a protective conductor current exceeding 3.5 mA but not exceeding 10 mA, shall be either permanently connected to the fixed wiring of the installation without the use of a plug and socket-outlet or connected by means of a plug and socket-outlet complying with BS EN 60309-2.

So unless you change the plug and socket to a commando type like used with the supply to a caravan or use a FCU then the machine must fail. That is simple. However after saying that the filters used can take some time to charge up and I have seen with a manual PAT tester the needle falling as the button is held and once steady it has been well below the 3.5 mA figure but with an automatic tester there is not a option to allow the first surge to clear.

I had this problem with some new scales which had been bought and it was discussed in the workshop with the other electricians and the resident engineer and we concluded that although likely not a problem it was also likely to be a recurring problem every time they were tested so we returned whole batch to the supplier as being non complaint. Within a week they were returned to us and passed the tests so I would guess the filter had been removed. This of course is why one should PAT test new items before any guarantee expires.

You have two issues. One should you have tested these in the first place without the knowledge given in the maintenance manual as how to test the individually parts? And secondly should you be changing the plugs for a FCU or would this need passing to another person.

I do not like the practice of cutting off the plugs although I know many testers do where items fail. But the moving of a washing machine into a quarantine area is not really easy, and if as you suggest there are a number of failures then this may be the best option, as to use it, will need cutting off anyway to wire into a FCU. However before getting out the side cutters I would talk to the client as cutting off the plug could invalidate the guarantee.

The whole idea of employing people to PAT test is a problem with so many testing only truly portable machines and leaving the freezer, dishwasher, washing machine, which all have timers and also immersion heaters, and hand driers which are connected with FCU's for the mythical someone else to test. The portable (Yes it had wheels and was moved with 9 articulated units) batching plant I worked on took three electricians two weekends to fully test. It was split into parts with each section being tested every month so over each three month period it was all checked. Clearly this was not a £2 per item job and all the electricians had full access to the workshop manual showing how each section should be tested.

Even then we had issues with the built in leakage device which did not trip as with British batching plants but only put on a lamp to say there was a fault. (It was German) After a major accident things I am told changed but since I was the guy injured I was not there to see what was done. It cost the firm a lot of money so I would ensure your insurance is valid for the work you are doing. Likely you need professional indemnity insurance not just public liability as you are giving a professional opinion as to the condition of the equipment. Some public liability insurance does cover for PAT testing but I would check to be sure.

 

[ericmark]

Is the PAT leakage test performed with AC or DC applied ?

The standard PAT tester measures the leakage with the appliance running so clearly tested with AC applied. The DC test is measured in kohms or Mohms not mA.

 

[JohnW2]

If it is AC then the "leakage" is almost certain to be current through the capacitors between live and earth in the filters on the electrical input to the machine. This is normal.

Eric has confirmed that 'leakage testing' is undertake with the appliance running, hence with AC. He has also mentioned the figures of 3.5mA and 10mA which apperar in BS7671. To get currents of 3.5mA and 10mA with 230V would require capacitors of at around 0.014 uF and 0.049 uF respectively - values which are in the ballpark of those found as filtering components.

Kind Regards, John.

 

[martinxxxxxx]

These are NEW machines! And some secondhand ones too.

This is the entire stock of a washing machine hire company - so far none have passed.

Different brand of washing machine.

Three different testers have been used (all of which are "in calibration")

Martin, if you perform a standard insulation test at 500V what insulation resistance reading so you get?

It says greater than 20Mohms on this tester. I have not tried it with my insulation tester. I might try it.

 

[martinxxxxxx]

These are NEW machines! And some secondhand ones too.

This is the entire stock of a washing machine hire company - so far none have passed.

Different brand of washing machine.

Three different testers have been used (all of which are "in calibration")

Martin, if you perform a standard insulation test at 500V what insulation resistance reading so you get?

It says greater than 20Mohms on this tester. I have not tried it with my insulation tester. I might try it.

 

[JohnW2]

Martin, if you perform a standard insulation test at 500V what insulation resistance reading so you get?

It says greater than 20Mohms on this tester. I have not tried it with my insulation tester. I might try it.

That's what I would have expected. It sounds as if the machines are fine (as witness the insulation resistance) but, as discussed earlier today, with 'high' readings for leakage current very probably due to filtering components within the machines. The machines are therefore very probably 'as they should be' - but I haven't got a clue as to what that means in terms of passing/failing a PAT test!

Kind Regards, John.

 

[securespark]

If these are stored in an unheated premises, the IR reading may fall due to dampness.

The guys I know do a basic L/E & N/E IR test at the plugtop, with the door closed.

They pass 1 Meg and over.

 

[JohnW2]

If these are stored in an unheated premises, the IR reading may fall due to dampness.

True, but I feel sure that the >20MΩ would be OK, particularly since the '>' implies that it is the highest answer the PAT tester can give (so the true IR might well be much higher than 40M&#937.

The problem appears to be the 'slightly high' so-called 'leakage current', measured with AC and probably due to a filter capacitor - and the question is what that means in terms of the outcome of the PAT test. I certainly don't know.

I also thought that the 'P' of 'PAT' stood for 'portable'. In what sense is a washing machine 'portable'? Is it because it is not fixed wired'?

Kind Regards, John.

 

[ericmark]

I also thought that the 'P' of 'PAT' stood for 'portable'. In what sense is a washing machine 'portable'? Is it because it is not fixed wired'?

Kind Regards, John.

You are of course correct but really it should be called "In-service Inspection and Testing of Electrical Equipment" however this is shortened to PAT testing.

The problem with any equipment where a timer, contactor or relay energises parts within the machine is that measuring at the plug with a 500 volt insulation tester is meaningless. So this items of that type and also any items where 500 volt may blow some component then leakage test is used instead.

If a 500 volt tester is used on a filter network which has an earth connection then it could well blow one of the electrolyte capacitors and cause too much current to flow with the leakage test.

Using the Robin PAT tester one could write ones own sequence and pass limits and I know I had some written for extension leads with RCD plugs and some IT equipment.

I wonder if there is a problem where the testing is damaging the electrolyte capacitors? Also if it was ran for some time before testing if it would re-form the electrolyte capacitors and would then pass.

It is so easy to forget all the electronics within a washing machine and to test with the insulation test. Likely I would have made same mistake but clearly from what is said one should not do an insulation test on any equipment which has a filter network.

Will be interesting to hear if the results improve when insulation testing is omitted from the test.

As a result I will watch this post.

 

[JohnW2]

I wonder if there is a problem where the testing is damaging the electrolyte capacitors? Also if it was ran for some time before testing if it would re-form the electrolyte capacitors and would then pass.

I doubt very much that electrolytic capacitors would be used for this purpose.Apart from anything else, they would have to be non-polarised or a 'back-to-back' pair to be used with AC. Perhaps more to the point, one doesn't usually find electrolytics with values less than abut 1uF - and that would have a reactance of only 3183Ω at 50Hz, hence a 'leakage current' of around 72.3mA at 230V AC.

Mains filter capacitors are much more commonly in the range 0.01 - 0.05 uF and, per my post yesterday, that woulld result in 'leakage current' in the 3.5-10mA range.

However, having said all that, a capacitor does not have to be electrolytic for its dielectric to be damaged by overvoltage - so what you suggest is possible. However, if that happened, the damage would usually be permanent - so a subsequent IR test would give a low result, which does not seem to be what's happened with the OP's machines.

I still suspect that the tester is simply seeing the reactance of a filter capacitor - but I don't know what that means in terms of the test outcome.

Kind Regards, John.

 

[ericmark]

I would agree with JohnW2 however this still leaves the question as what to do in this situation.

The regulations clearly limit the leakage with a 13A plug to 3.5 mA so using a 13A plug one should fail them.

So why would a manufacturer make machines that will clearly fail UK rules.
1) Not made in UK and designed to another standard.
2) Commercial machines designed to use Commando type plugs or direct connection.

As to way forward of course writing to manufacturer is first line and maybe there is a replacement network to allow the machines to be used in the UK. Since some are new clearly there will be some guarantee and this would be first line of approach.

If they are commercial machines fitting commando type plugs would be an option. As to if one could offer adaptors 13A to 16A would still allow their use which does seem to completely defeat the whole idea?

Maybe this could be an interim solution until the manufacturers have replied and said how it should be redressed.

One could of course ignore the 17th Edition and set ones own limits but I would consider this would invalidate and insurance and I would not do this.

Or of course simply fail all of the machines.

It would be interesting to know are they commercial machines and are they British? What we of course have to remember the 13A plug is British so these machines could well comply with European rules but still fall foul of British requirements.

 

[JohnW2]

I would agree with JohnW2 however this still leaves the question as what to do in this situation.
The regulations clearly limit the leakage with a 13A plug to 3.5 mA so using a 13A plug one should fail them.
So why would a manufacturer make machines that will clearly fail UK rules.
1) Not made in UK and designed to another standard.
2) Commercial machines designed to use Commando type plugs or direct connection.
As to way forward of course writing to manufacturer is first line and maybe there is a replacement network to allow the machines to be used in the UK. ....

Yes, something odd is going on here. I find it hard to believe that anyone is manufacturing machines for the UK market whose design is such that they are almost bound to fail the tests. Indeed, I wonder if they would even be allowed to sell such machines in the UK?

Kind Regards, John.