Mind the gap
05 March 2013
A row has developed on the subject of flameproof gaps.
In this photograph, the gap has been opened up to allow a visible flame to exit the flamepath. Note how the gap gives a high-velocity planar flame that will entrain the cold gases
I had always thought that the basic principles of the flameproof concept were clearly understood by those operating in our industry. These can be summarised as:
1. The enclosure must be strong enough to withstand an internal explosion without rupturing.
2. The products of combustion, from such an explosion, must be vented to the outside atmosphere through controlled gaps in such a way that they do not ignite an external explosive atmosphere.
3. Because of the way in which the gaps, as specified in the standard, prevent external ignition, there should be no uncontrolled gaps. The absence of uncontrolled gaps is proved by observation of leakage during the over-pressure test.
The fact that the specified form of gap works is proven not only by the practical testing of a representative sample of each item of equipment, but it also can be demonstrated through the application of the Universal Gas Law:
P.V/T = C (Pressure times volume divided by absolute temperature is a constant)
This basic physics is well understood by the cyclist, who can get his fingers burned at the top of the cycle pump when inflating his tyres. The flameproof gap is the same process in reverse. For most flameproof gaps, the dimensions ensure that the pressure drop of the expelled gases is sufficient to cool them below the temperature at which they will ignite the surrounding mixture.
For plain flange gaps, there is a secondary effect which may or may not be important, but which is recognised in the installation code, where the minimum separation distance between a flange gap and an obstruction is defined.
Just as a Bunsen burner uses the velocity of the gas through a jet to suck air into the gas stream to provide a gas/air mixture, so the rapidly-expelled gases from a flange gap entrain a quantity of the cold gases outside the enclosure that is sufficient to ‘snuff out’ any residual tendency to propagate the ignition. How important this secondary effect is will vary in individual equipment.
So why the fuss? The issue concerns cemented joints in enclosures, such as may be used to hold a glass window into a mounting frame. It has been the practice in all previous editions of the European and International standards to require that there shall be no visible leakage through a cemented joint during the over-pressure test.
However, largely due to pressure from the USA, the latest draft edition of 60079-1 was reworded to allow leakage, provided that the equipment subsequently withstood a transmission test, to prove that the particular uncontrolled gap through the cement did not, in fact, pose a safety risk.
To a certain extent, this is all well and good. Where I take issue with the new edition is that, as proposed, this is only a type test. In other words, if a single sample shows a leakage path during the over-pressure test, but subsequently passes the transmission test, then EVERY subsequent produced item can be despatched from the factory without any further checking.
For me, it is too great a leap of faith to say that, just because one sample leaked and passed the test, every other sample would also have passed. To make that statement, you have to be confident that the cement in every example of the equipment will fail in exactly the same way.
At a recent meeting of the Maintenance Team, a compromise has been sought, whereby we move from one to two type test samples if leakage has occurred. Is this sufficient?
I remain convinced that the only safe solution is to require zero leakage during the over-pressure test, as in that way, you can have some assurance that there will be no uncontrolled gaps which might behave differently from the gaps permitted in the standard.