IET logo
 
IET
Decrease font size
Increase font size
Topic Title: SECTION 705 OF THE BGB
Topic Summary: A CONUNDRUM
Created On: 05 August 2014 11:26 PM
Status: Post and Reply
Linear : Threading : Single : Branch
1 2 3 4 5 6 7 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.
 05 August 2014 11:26 PM
User is offline View Users Profile Print this message



John Peckham

Posts: 7457
Joined: 23 April 2005

As I am sure you all know 705.411.1 and 705.422.7 requires all circuits on a farm to be 300mA RCD protected for ADS and fire protection regardless of the earthing system (it's TNS).

On a farm I EICRed the electrical contractor is doing all the required remedial works to bring the farm up to current standards.

The farm is a dairy farm with some 1500 cows in sheds. The output from the cows is taken by scrapers and an in floor augur to a huge swimming pool size slurry tank. The slurry is pumped up in a large diameter pump to the top of a tower where rollers extract the solids and the liquid is pumped in to a further large tank. The very large pumps have VSD speed controllers. When running these VSDs trip the newly fitted 300mA RCD.

If this kit were to fail in the night because the RCD had tripped taking out all the power to the pumps would be a disaster.

I have told the contractor to take a separate non-RCD supply to the pumps for the moment whilst I have a think.


Given these pumps and tower are outside and have normal circuit protection and there is no hay , other than the wet contents of the tanks, I am thinking it would be sensible to omit the RCD protection for fire although this would be contrary to 705.

I would value your views especially from Stu L?

-------------------------
John Peckham

http://www.astutetechnicalservices.co.uk/
 05 August 2014 11:33 PM
User is offline View Users Profile Print this message


Avatar for sparkingchip.
sparkingchip

Posts: 6030
Joined: 18 January 2003

I suspect if asked you know exactly what the earth leakage is on this equipment, so I'll ask what it is.

Andy
 05 August 2014 11:40 PM
User is offline View Users Profile Print this message



John Peckham

Posts: 7457
Joined: 23 April 2005

I don't know the earth leakage on the particular piece of kit but it was high on the board supplying the whole installation due to a neutral to earth fault. I would imagine that the VSD is chucking out loads of harmonics and a high level of earth leakage. I know even small VSD will trip RCDs.

-------------------------
John Peckham

http://www.astutetechnicalservices.co.uk/
 05 August 2014 11:41 PM
User is offline View Users Profile Print this message


Avatar for mapj1.
mapj1

Posts: 2772
Joined: 22 July 2004

Is it practical to measure how bad the earth leakage really is on each pump/control set?

If it is really anything like1/3 amp at 50Hz, that's a serious filter capacitance, I make it would have to be about 700j ohms, or 4uF L-E if it was single phase, and as I assume really 3 phase, then it must be more, but that 4uF is just the un-cancelled part !!

Could it be there is a wiring error, and there is a filtering star point that should be going to neutral, not to earth, or even an NE link fitted in the controllers ?

If its HF stuff causing the problem, then pre-filter down to neutral, not to earth, if required with the filter on spacers in an insulating enclosure to get the fast stuff off the lines and back up onto the L-L-L-N route rather than L-E.

some inverters have instructions for lifting the Y capacitors in this way for use on RCD protected supplies. The price is that stronger external filtering is then needed.

Is there something a bit like this anywhere

PWM reconstruction filters - see page 6 for waveforms

Or is there scope to comply in spirit and not in letter, and have more than one RCD (say one per pump) and get well below threshold that way.

If you have makers part nos for the controllers I may be able to get some info, before I have to vanish again.

edit fixed links - missed that it was https.
Also, if its nearly there then there is normally the option to change the PWM chopping frequency - if it is higher than need be, then lowering it reduces the number of pulses per second and therefore the general EMI background level, almost for free

-------------------------
regards Mike

Edited: 06 August 2014 at 01:37 AM by mapj1
 05 August 2014 11:45 PM
User is offline View Users Profile Print this message



Zs

Posts: 2899
Joined: 20 July 2006

I'm thinking page two of the big green bumper fun book JP:

' Whilst the publisher and contributors believe that the information and guidance given in this work is correct, all parties must rely upon their own skill and judgement when making use of it...'

Then, given that, a sensible decision about the safety and protection of the beasts from electric shock and from other issues such as bacteria.

What do other establishments such as this do?

Zs
 06 August 2014 12:11 AM
User is offline View Users Profile Print this message


Avatar for sparkingchip.
sparkingchip

Posts: 6030
Joined: 18 January 2003

I was going to keep it simple and ask the result of the ramp test on the RCD then compare it to the leakage to see the RCD had any chance of hanging on in there without all the fuzzy stuff in the system.
 06 August 2014 12:52 AM
User is offline View Users Profile Print this message



alancapon

Posts: 5765
Joined: 27 December 2005

I would expect the leakage to even out with a three phase supply. This leaves either Mike's suggestion of a wiring error, or the rcd is being affected by harmonics.

Regards,

Alan.
 06 August 2014 01:56 AM
User is offline View Users Profile Print this message



Jaymack

Posts: 4638
Joined: 07 April 2004

Originally posted by: John Peckham
The very large pumps have VSD speed controllers. When running these VSDs trip the newly fitted 300mA RCD.

I suspect that this tripping is caused on starting surges. If the VSD's are correct including the settings, a time delayed RCD would probably avoid this.

Regards
 06 August 2014 07:57 AM
User is offline View Users Profile Print this message



Zs

Posts: 2899
Joined: 20 July 2006

Something you do have the technology for JP, is to log the earth leakage. Might tell you when things are happening and help to tie them down.

I'll not give all the secrets away on here but give me a ring if you want to hear how I set up on a TPN to achieve this.

Zs
 06 August 2014 10:05 AM
User is offline View Users Profile Print this message


Avatar for Parsley.
Parsley

Posts: 1003
Joined: 04 November 2004

 06 August 2014 10:22 AM
User is offline View Users Profile Print this message


Avatar for davezawadi.
davezawadi

Posts: 2654
Joined: 26 June 2002

I have come across this too John, in many ways RCDs are not really very good. As said above the supply harmonics do not cancel in the leakage transformer, so larger loads instantly cause tripping. If this is the problem then a time delay will not work, but might be worth a try. VSDs do not normally have much 50Hz leakage, but may have quite a lot of other stuff, you could try an FFT of the supply currents on you analyser, more than a few percent of the 50Hz figure will cause tripping.

This section of the BGB needs revision, it seems to have been written by people who do not realise that farms can have very large electrical loads and 300mA is not necessarily suitable for modern equipment. If there were no VSD and a 50kW motor you would not be able to use an RCD anyway. As there is no fire risk, I would go with normal protection here, the RCD is both unworkable and unnecessary so the basic premiss in the BGB doesn't apply. This is a difficult area anyway, if this plant were the other side of a boundary fence and in an industrial park next to a farm, would you fit the RCD? I don't think any reasonable person would, so this can be considered industrial rather than farm equipment. RCDs in areas with inflammable materials or animals are fair enough, but not other areas. This has come up where on farm processing plants are built, it is a case of correct designation of the area by its use rather than its simple proximity to a farm, after all the planners use this to stop development all the time.

-------------------------
David
CEng etc, don't ask, its a result not a question!
 06 August 2014 10:32 AM
User is offline View Users Profile Print this message



whjohnson

Posts: 723
Joined: 24 January 2009

Some of the stuff we work on has a number of large inverter drives, and we have found it impossible to get them to work with any kind of RCD protection.

We took the view that things like this are what the Designer's 'Departures' box is for on the forms!

Oh, and make sure all your earthing and bonding is up to scratch.

-------------------------
Political correctness is a doctrine, fostered by a delusional, illogical minority, and rabidly promoted by an unscrupulous mainstream media, which holds forth the proposition that it is entirely possible to pick up a turd by the clean end.
 06 August 2014 03:35 PM
User is offline View Users Profile Print this message


Avatar for OMS.
OMS

Posts: 19656
Joined: 23 March 2004

What's the rating of the motor, what's the starting current and what's the RCD current rating

This may be no more than simple magnetic saturation of the RCD toroid.

The alternative may be a "proper" circuit breaker with ground fault protection

Either way - if there really is 300mA leaking then it's pretty dangerous to ignore it

Regards

OMS

-------------------------
Failure is always an option
 06 August 2014 04:55 PM
User is offline View Users Profile Print this message



perspicacious

Posts: 7232
Joined: 18 April 2006

"Some of the stuff we work on has a number of large inverter drives, and we have found it impossible to get them to work with any kind of RCD protection.
We took the view that things like this are what the designer's 'Departures' box is for on the forms!"


120.3 reads: Any intended departure from these parts requires special consideration by the designer of the installation and shall be noted on the Electrical Installation Certificate specified in Part 6. The resulting degree of safety of the installation shall be not less than that obtained by compliance with the Regulations.

I'd be interested in reading the signed for technical stance confirming that the omission of a 300 mA RCD will satisfy the requirements of 705.411.1 (iii) and 705.422.7

As I'm protective of my PII, I'd ask the supplier of the VSD etc to either confirm that their device(s) will not trip a 300 mA RCD or to provide their signature on the EIC before authorising their payment.

Regards

BOD
 06 August 2014 05:11 PM
User is offline View Users Profile Print this message


Avatar for mapj1.
mapj1

Posts: 2772
Joined: 22 July 2004

Well, it doesn't of course.
If however there really are some 300mA returning, as part of the normal design, working correctly, as ther makers intended, (I have worked with some EMC chambers and similar where to keep signals out/in this sort of current is found due to very large filter capacitances to ground)
Then the only thing you can do is to re-create an equivalent level of safety, by restoring the "multiple independent faults required to give rise to danger" condition lost when the CPC current is high enough that a single fault is dangerous.
This is no sillier than precautions to be taken for the high earth conductor currents for IT equipment.

The mitigation is multiple independent earth paths, so that any one can fail and it is not dangerous - just instead of one of your 2 or more faults before a danger arises is you RCD being jammed, it is the interruption of one of the redundant CPC.
Problems arise if its TT of course, in that you now need to guarantee a really good rod impedance at 8KHz or whatever not just 50Hz..
At that point, filtering to neutral, rather than earth, and placing the whole filtering thing inside a protective cover, as if it were a live part is an alternative. One can then have a large capacitance between neutral and earth, as there is a much lower voltage there.
This does not meet BS7671, and surprisingly many things could be done under other legislation.
For wilder examples



Chatting to others, it is not of course the 50Hz leakage that is biting you here, but the square wave at 8/16/30KHz or whatever the VSD is chopping at.

There is ~400V or more p-p square waves on each output, and you only need a fraction of a microfarad to earth to explain the current you need to fire the RCD- it also explains how lowering the frequency helps if that is possible in this case. Xc being inversely proportional to frequency and all that.

-------------------------
regards Mike

Edited: 07 August 2014 at 01:46 PM by mapj1
 06 August 2014 08:09 PM
User is offline View Users Profile Print this message


Avatar for slittle.
slittle

Posts: 3507
Joined: 22 November 2007

JP,

Give me a yell tomorrow, been there ;-)


Stu
 06 August 2014 08:59 PM
User is offline View Users Profile Print this message



paulskyrme

Posts: 778
Joined: 12 February 2003

Just use a suitable RCD.

There may, or may not be 300mA of leakage.

Most VSD's are not designed to run on RCD's due to the harmonics filters up front.
Now, the contractor could take a suitable supply from the DB, if it's not a Schneider, or Siemens, then run it through one of their enclosures with one of their 300mA RCD's in, and job done.
After all you don't want to upset the type approval bandwagon do you!

If you want details on the RCD's you know where I am.

I've come across this when I was working for the drives manufacturer.
Our MI's said no RCD's.
Some clients insisted, so it could be done, at a price, and it's NOT cheap!

The leakage, may, well be designed in and deliberate in the drive system.
 06 August 2014 10:18 PM
User is offline View Users Profile Print this message



Zs

Posts: 2899
Joined: 20 July 2006

As a note.. I've seen >300mA leakage before and in that case it was a ground source heat pump at the moment it kicked-in to begin sourcing heat from the ground. no huge motors or any of our usual suspects.

Stu, you've got to tell us all about it....

Zs
 06 August 2014 10:24 PM
User is offline View Users Profile Print this message


Avatar for leckie.
leckie

Posts: 1863
Joined: 21 November 2008

Originally posted by: slittle

JP,



Give me a yell tomorrow, been there ;-)





Stu


Wel Stu, don't keep it to yourself! Spill the beans.
 06 August 2014 10:26 PM
User is offline View Users Profile Print this message


Avatar for leckie.
leckie

Posts: 1863
Joined: 21 November 2008

Oh I missed your request to Stu Zs. I was too busy catching up on the post!
IET » Wiring and the regulations » SECTION 705 OF THE BGB

1 2 3 4 5 6 7 Next Last unread
Topic Tools Topic Tools
Statistics

See Also:



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