Installing gas system

[Bob18]

Extracted, from table "Y" of BSEN1057 (for copper, seamless pipes used in various applications including lp gas)

 

6mm od - 0.8mm

 

1/4in - no definition for this pipe size given

 

8mm od - 0.8mm

 

3/8in - no definition for this pipe size given

 

10mm od, wall = 0.8mm

 

1/2 in - no definition for this pipe size given

 

12mm od, wall = 0.8mm

 

15mm od, wall = 1mm

 

18mm od, wall = 1mm

 

3/4in - no definition for this pipe size given

 

22mm od, wall = 1.2mm

 

1in - no definition for this pipe size given

 

28mm od, wall = 1.2mm

 

 

Table "Y" gives a lot of other info about working pressures and the like, but does not define any Imperial sized pipes in the size range covered (6mm to 108mm), although there are very close near misses for 3inch and 2.625inch.

[MtB]

Agree entirely - I have heard some horror stories, I'll agree.

 

But to me it is obvious that when you see posts from people like "Mike the Boilerman", that their advice is solid. Generally the best of the professionals can back up what they say with the relevant regulations, standards, publications, etc - something I would struggle to do without a lot of research, and then I might still cock up.

 

For example if I trawled enough old posts, I could remind myself why the wall thickness of the generally sold metric pipe doesn't fully meet all the requirements.

 

Yes, I know it's marginal - yes, I know it's not going to kill you, and yes, I know people use it, without problems.

 

But when someone who really knows explains why strictly it should not be used, their arguments are coherent, and I'm happy to follow their good advice.

 

(Then of course we once had "TerryL" - "heating engineer" quite extrordinaire - so you do have a point!

 

 

Alan,

 

Very kind of you to say stuff like this about my posts on gas but I feel duty-bound to point out to the board that I do NOT do 'gas on boats' for a living. I earn my income as a domestic boiler repair technician. (I specialise in all the old, awkward, rare, or generally difficult-to-mend boilers that most ordinary technicians do not understand and cannot repair.)

 

I maintain my qualification to carry out gas work on boats mainly so I can carry out gas work on my own boat without without having to tiptoe around the edges of legality, what with me being a Gas Safe Registered technician for the day job!

 

As you point out, there are plenty of Gas Safe Registered installers out there making a pig's ear of boat gas installations so Gas Safe Registration is certainly NOT a reliable way of identifying someone who knows what they are doing. The problem is that most domestic gas bods (let alone their customers!) are not even aware that there is a separate ACS exam necessary to install LPG on boats along with a whole separate set of regulations, so many are willing to work 'out of scope' unwittingly and give the half-cocked advice we often hear about on here. E.g. the "I consulted a CORGI bod and he said I should put sealant on the olives" comment earlier in this thread.

 

The worst example I ever encountered was a boat gas-plumbed using domestic half-hard (Table X) rigid copper tube and soldered fittings throughout, including an ordinary household LPG cooker with no flame supervision on the hob or grill! The customer had paid top money for this to be done by a (then) CORGI domestic installer not qualified to install in boats. Disgraceful. There was NOTHING recoverable in the installation.

 

Anyway I think I've sent the board off on a wild goose chase by suggesting that the wall thicknesses required for boat gas pipes are defined in BS EN 1057. It appears not. I'm certain that last time I checked (i.e. last time I took my boat gas ACS renewal exams), metric pipe wall thicknesses (with one exception) were too thin. There is a table somewhere in the boat gas regs that demands specific wall thicknesses for boat gas pipes which are not available in metric pipe sizes (with one exception).

 

It isn't that imperial 'must be used', it is that last time I checked, only imperial pipe was commercially available (with one exception) with thick enough walls. If metric pipe was available with thick enough walls, metric pipe and fittings would be fine!

 

I'll have a look later and see if I can find this table of required wall thicknesses. The other possibilty is that BS EN 1057 has changed recently to specify thicker pipe walls on metric pipe, delete all reference to imperial pipe and that metric pipe now complies. But I rather doubt it. It's not the way BS standards are done!

[MtB]

I've just found, in BS 5482 part III, a reference to "BS EN ISO 10239, Small craft — Liquified petroleum gas (LPG) systems."

 

I don't have a copy of this. Maybe the gas pipe wall thicknesses are defined in here.

 

Does anyone have access to a copy?

[ChrisPy]

BS EN 10239:

 

6.2 Piping

6.2.1 Only solid drawn copper piping or drawn stainless steel piping, which are galvanically compatible, shall be used for rigid supply lines. The minimum wall thickness for piping of outside diameter equal to or less than 12 mm shall be 0,8 mm and 1,5 mm for piping of outside diameter greater than 12 mm.

 

 

PS strange that when the EN becomes a BS they can't even be bothered to change the European comma decimal point to a British full stop.

Edited September 5, 2011 by ChrisPy

[Bob18]

Please note: BS5482:1999 is replaced by BS EN ISO 10239:2000 (BSI website, http://www.standardsdirect.org/standards/standards4/StandardsCatalogue24_view_3824.html )

Further note BS EN ISO 10239:2000 has been updated to BS EN 10239:2008 (with another increase in it price to £130...)

[Bob18]

The reason for the use of commas instead of points to delineate a decimal is due to the laziness of members of one of the UK technical committees.

 

 

 

 

In a previous life my boss was the chair of one the TCs, and he landed us with the task of proofing the final, just on its way to the printers to make sure all punctuation, spelling, notation and so on was totally correct. Four of us set to, and two days later really ***d the printer off, there was a howler on about the third page that meant a total re-vote was needed. We certainly earned our ale that week.

[MtB]

BS EN 10239:

 

6.2 Piping

6.2.1 Only solid drawn copper piping or drawn stainless steel piping, which are galvanically compatible, shall be used for rigid supply lines. The minimum wall thickness for piping of outside diameter equal to or less than 12 mm shall be 0,8 mm and 1,5 mm for piping of outside diameter greater than 12 mm.

 

 

PS strange that when the EN becomes a BS they can't even be bothered to change the European comma decimal point to a British full stop.

 

 

Ok ChrisPy has shown it is even MORE complicated.

 

The wall thicknesses required by BS EN 10239 rule out most of the metric copper tube widely available in the UK, and even some of the imperial! Look at the wall thicknesses on the BES website for example. http://www.bes.co.uk/products/139.asp

 

0.8mm wall thickness for tube 12mm and below rules out all the metric tube from BES except their 5mm tube (fat lot of use 5mm tube is for LPG). Their imperial tube is available in TWO wall thicknesses and the thicker option complies up to 3/8" diameter. Their 1/2" tube still fails to comply as tube above 12mm needs to be 12.5mm wall thickness. Table W is fully annealed and hand-bendable.

 

I suspect this is because all the annealed copper tube BES supply appears to be to BS 2871 : Part 1 : Table W, annealed (EN 1057 soft), which is of course different from the Table Y measurements quoted by Bob18 earlier in this thread. Table Y is, IIRC, half hard copper tube i.e. rigid but bendable wuth formers like ordinary 15mm, 22mm, 38mm copper tube etc.

 

So in summary, as always, compliance depends on the standard you have picked to comply with. If you are trying to comply with BSS and/or BS5412:2005 Part III, any annealed pipe to BS EN 1057 Table W will do, metric or imperial as BS5482:2005 Part III does not specify wall thicknesses, only that compliance with Table W is necessary. If you need to comply with BS EN 10239, then you are up a gum tree as any pipe 1/2" or bigger with 1.5mm wall thickness appears to be hard to get, metric OR imperial. In fact metric 12mm tube with 0.8mm wall thickness complies and has a bigger inside diamter than compliant 1/2" pipe with 1.5mm wall thickness! Anomolies Rulez OK!

 

I suspect compliance with BS EN 10239 is only necessary for a new boat - in order to comply with the RCD. But that is just my guess. Maybe someone else can say when compliance with BS EN 10239 is (and isn't) necessary.

 

Mike

Edited September 5, 2011 by Mike the Boilerman

[Speedwheel]

I would assume that the real reason for chandlers to say "metric is the way to go" is that Imperial pipe is getting harder to source in small quantities at sensible prices

 

They don't. All chandlers should still have imperial. It is perfectly easy to get.

Chandlers should not say 'metric is the way to go' because they would be advising someone to fit something that doesn't comply with the necessary regs.

[MoominPapa]

Ok ChrisPy has shown it is even MORE complicated.

 

The wall thicknesses required by BS EN 10239 rule out most of the metric copper tube widely available in the UK, and even some of the imperial! Look at the wall thicknesses on the BES website for example. http://www.bes.co.uk/products/139.asp

 

0.8mm wall thickness for tube 12mm and below rules out all the metric tube from BES except their 5mm tube (fat lot of use 5mm tube is for LPG). Their imperial tube is available in TWO wall thicknesses and the thicker option complies up to 3/8" diameter. Their 1/2" tube still fails to comply as tube above 12mm needs to be 12.5mm wall thickness. Table W is fully annealed and hand-bendable.

 

I suspect this is because all the annealed copper tube BES supply appears to be to BS 2871 : Part 1 : Table W, annealed (EN 1057 soft), which is of course different from the Table Y measurements quoted by Bob18 earlier in this thread. Table Y is, IIRC, half hard copper tube i.e. rigid but bendable wuth formers like ordinary 15mm, 22mm, 38mm copper tube etc.

 

So in summary, as always, compliance depends on the standard you have picked to comply with. If you are trying to comply with BSS and/or BS5412:2005 Part III, any annealed pipe to BS EN 1057 Table W will do, metric or imperial as BS5482:2005 Part III does not specify wall thicknesses, only that compliance with Table W is necessary. If you need to comply with BS EN 10239, then you are up a gum tree as any pipe 1/2" or bigger with 1.5mm wall thickness appears to be hard to get, metric OR imperial. In fact metric 12mm tube with 0.8mm wall thickness complies and has a bigger inside diamter than compliant 1/2" pipe with 1.5mm wall thickness! Anomolies Rulez OK!

 

I suspect compliance with BS EN 10239 is only necessary for a new boat - in order to comply with the RCD. But that is just my guess. Maybe someone else can say when compliance with BS EN 10239 is (and isn't) necessary.

 

Mike

 

Wow, now I know why I do software engineering for a living and not real engineering. That many arbitrary rules is a completly insane. Us programmers have one very simple rule: "If it breaks, you get to keep all the pieces".

 

MP.

[Speedwheel]

Yes, available. I have bought a full set of these quite recently, so recently that they are not yet installed.

Bubble tester a must. Cut corners elsewhere if you must but a bubble tester is just such a good solution to the problem.

Good luck!

 

The downside of a bubble tester is that it doesn't test the high pressure side of the system like a tightness test does.

[MtB]

The downside of a bubble tester is that it doesn't test the high pressure side of the system like a tightness test does.

 

Another piece of misinformation!

 

A tightness test (we have to call it a soundness test with LPG - God knows why) also only tests the low pressure side. A correctly working regulator will be in the 'locked up' condition and maintaining bottle pressure between itself and the gas bottle isolator valve during a conventional soundness test with a manometer.

 

The rel downside of a bubble tester is that it will often not reveal a small leak, just big enough to cause a FAIL on a conventional soundness test, unless you hold the button in for longer than the instructions state. Therefore they cannot be relied upon in my personal opinion. Everyone else in the industry thinks otherwise

[Speedwheel]

Another piece of misinformation!

 

A tightness test (we have to call it a soundness test with LPG - God knows why) also only tests the low pressure side. A correctly working regulator will be in the 'locked up' condition and maintaining bottle pressure between itself and the gas bottle isolator valve during a conventional soundness test with a manometer.

 

Under the test done by a BSS examiner, the soundness test can test all the way back to the bottle shut off valve and therefore does test the high pressure side.

 

Having noted the regulator lock up pressure, then done your let by test, you then test the system at 30mb. You can isolate at the bottles so when the pressure is dropped to 30mb unlocks the regulator meaning you are testing the high pressure side too. If it fails and there is a separate isolation valve in, other than that on the bottle, you can retest with this closed. The isolation valve has to the in the gas locker.

Edited September 5, 2011 by Speedwheel

[mayalld]

Another piece of misinformation!

 

A tightness test (we have to call it a soundness test with LPG - God knows why) also only tests the low pressure side. A correctly working regulator will be in the 'locked up' condition and maintaining bottle pressure between itself and the gas bottle isolator valve during a conventional soundness test with a manometer.

 

???

 

Surely not.

 

If you attempt a manometer soundness test with bottle pressure present on the HP side of the regulator, then any small leak will not register, because the regulator will use the remaining HP supply to bring the LP side back up to 37mBar.

 

I always understood that manometer soundness tests should be carried out having burned off enough gas to drop the LP side to around 25mBar.

[MtB]

???

 

Surely not.

 

If you attempt a manometer soundness test with bottle pressure present on the HP side of the regulator, then any small leak will not register, because the regulator will use the remaining HP supply to bring the LP side back up to 37mBar.

 

I always understood that manometer soundness tests should be carried out having burned off enough gas to drop the LP side to around 25mBar.

 

Actually I agree, my post at 9.34am this moring is wrong. My apologies to Speedwheel. Need to go to work now so no time to edit it. Or maybe I'll leave it as the following posts will make no sense.

 

Mike

[mayalld]

Actually I agree, my post at 9.34am this moring is wrong. My apologies to Speedwheel. Need to go to work now so no time to edit it. Or maybe I'll leave it as the following posts will make no sense.

 

Mike

 

 

Have a greenie for holding your hands up to a mistake.

 

I would add that everything else that you've posted on the subject has been spot on.

[Speedwheel]

Have a greenie for holding your hands up to a mistake.

 

I would add that everything else that you've posted on the subject has been spot on.

 

+1

[rjasmith]

I have read all these threads with great interest and have started to lose the will to live. I’m unable to find anything on the internet that say you must use imperial pipe, can you even get imperial pipe any more. All bubble testers are in metric only and I would imagine mixing imperial and metric a definite no no, introducing adaptors will only lead to additional joints.

If metric pipe is to BS EN 1057 how can it be incorrect to use. These are just my observations from all the above comments, I'am here to learn as are most of us.

 

I've just had a look at BS EN 1057 (via a certain wonderful library's magic website!) and the many tables in there on wall thickness. I have drawn two conclusions:-

 

1. If I have to read any more of that particular document I will also lose the will to live. (It would seem that there is no agreement in all of Europe on wall thicknesses!)

 

2. I now know I made the right decision to go for a gas free boat.

 

(I will now go and hide because conclusion 2 is not politically correct here).

 

Richard

[bizzard]

I would always use the thicker wall imperial tubing as unless using the minature inserts in the 10mm which reduce the ID anyway.the nut and olive of compression fittings just squeeze the tube and you can't get them satisfyingly tight.no good in my opinion. I use a Gaslow regulator with pressure gauge,which i find invaluable for testing both high and low pressure sides.

And have frequently found that on changing gas bottles unless you clean both the mating surfaces of regulator and bottle fitting and test afterwards with test liquid or whatever,people waste a lot of gas unbeknowingly through this as i've discovered whilst working on peoples boats. bizzard

[BEngo]

And have frequently found that on changing gas bottles unless you clean both the mating surfaces of regulator and bottle fitting and test afterwards with test liquid or whatever,people waste a lot of gas unbeknowingly through this as i've discovered whilst working on peoples boats. bizzard

 

 

That is my experience too- it's usually the bottle that has a scratched or damaged seat.

 

Please don't use washing up liquid for testing your bottle connection (or any other fitting) though-it's usually full of salt and sometimes ammonia which can cause rapid intergranular corrosion of brass with the result that the fitting just opens up like a flower. Proper leak testing fluid is quite cheap and easy to get hold of.

 

N

[Bob18]

Having spent much of the last 20 (very) odd years ploughing through a veritable morass of standards there are a couple of things I've learnt.

First, always say which version of a particular standard you are trying to comply with

Second, when working with a multi part standard always say which part you are working to

Third, arguing with an auditor, inspector, or NoBo (and a few other like minded beings) is a bit like wrestling with a pig in mud - its their environment, and they are having fun, and they will win - even when you think you've won, they will win...

 

If you think its bad working with the British Standards you should try working on a project where there standards are drawn from several nations - we did, and there were so many conflicts and contradictions it kept the lawyers in G&T for a long time just working out which clause of which standard had presidence over the other dozen standards...

 

 

Anyway, back to the subject...

One thing that is certain is that the availability of Imperial copper pipe and fittings is going to get harder in years to come, on that front, for NEW boats it makes sense to use metric fittings and pipes.

For existing boats with Imperial fittings then you really don't want to mix Imperial and metric fittings and pipes so stick with what you've got.

[Speedwheel]

Anyway, back to the subject...

One thing that is certain is that the availability of Imperial copper pipe and fittings is going to get harder in years to come,

 

It is not hard to get now and the mills are saying that they have no intention of stopping production so I don't see it being a problem.

 

 

on that front, for NEW boats it makes sense to use metric fittings and pipes.

 

.... except for the the fact that by fitting metric your system will not comply! (Edit to add: subject to the info on some metric pipe complying give by Mike above)

Edited September 5, 2011 by Speedwheel

[smileypete]

But when someone who really knows explains why strictly it should not be used, their arguments are coherent, and I'm happy to follow their good advice.

This is my main beef with standards, they explain what you must do but not the rationale behind it.

 

Anyone know what standards are applicable to LPG in caravans/camper vans, and what pipe thickness they require?

 

.... except for the the fact that by fitting metric your system will not comply!

You can certainly find metric pipe that complies - at a price...

 

cheers,

Pete.

[mayalld]

Anyway, back to the subject...

One thing that is certain is that the availability of Imperial copper pipe and fittings is going to get harder in years to come, on that front, for NEW boats it makes sense to use metric fittings and pipes.

 

Your certainty is misplaced, and I fear that you have missed much of the important stuff in this thread.

 

Despite the fact that copper pipe for water and copper pipe for gas LOOKS the same, it is NOT the same thing.

 

For the most part, if you go out and buy a coil of 8mm/10mm copper pipe, it will NOT meet the requirements for wall thickness required for gas piping.

 

So, whilst there is no requirement to use imperial pipe, there is a requirement to use thick walled pipe, and that is only usually found in imperial sizes. Trying to find metric pipe would be difficult.

[Bob18]

To comply the pipe needs to have certain dimensions.

 

Go back a few posts.

 

Outside diameter 12mm or less, wall thickness = 0.8mm

 

Over 12.0mm (and that includes 1/2 inch, which is 12.7mm) wall thickness is 1.5mm

 

BES are only one tube supplier, Abacus Tubes Limited for instance do a range of coiled tubes with 0.8mm wall (5, 6, 8, 10 & 12mm od), as well as a 15mm by 1.5mm wall.

 

Indeed don't buy your 1/2in or greater Imperial tubes from BES as they don't comply with the standard, being too thin - a 1/2in pipe needs a 16G wall to comply (that's not to say they wouldn't special order if you really wanted it)

[Hannah and Jay]

Ever feel like your post has been hijacked so people can have a dig at each other??!

 

Thanks for all the USEFUL advice but as for the rows about metric pipe, even what is a pipe, how about starting separate threads for these rather dull arguments!

 

And to the poster who didn't have patience for my rather incompetant IT skills and instead of offering friendly advice (surely what this forum should be about?) on how to multiquote, was rather rude, may I suggest the sage advice of my old grannie - If you can't say something nice, don't say anything at all!

 

I'd really like to see these Calor tables that have been mentioned, I'll try and google them, to figure out whether 3/8" or 1/2" would be more suited to our purposes (PLease note - I did not even mention the word imperial!!)

Edited September 5, 2011 by Hannah and Jay