Connect tube to pipe 2

[azmanvai]

Hi there

We just had an incident where a tubing connected to an end flange snapped and caused a chemical burn to the operator? Does anyone know what is the design/fabrication standard on tube to pipe/flange connection?

Tq

 

[moltenmetal]

I'm assuming you're talking about joining rigid metallic tubing to a piping system.

Generally a tubing compression fitting male connector is attached to the pipe by any joining method permitted in the piping code and your piping specification. Generally a valve is installed at transitions like this such that leaking tube joints can be repaired without draining the piping.

Making the connection at a threaded or flanged joint allows you to easily renew the male connector when it is beyond its service life. All compression fittings have a limited number of assembly/disassembly cycles before they start to leak in a way that can't be repaired just by tightening. Tube fittings welded directly to pipe or to pipe tees can be used where permitted by piping specifications and code, but repair of the tube fitting will then require the fitting to be cut out and a new one welded in place.

Some vendors of fittings sell flange adapters which consist of a flange and male connector body machined from a single forging. It is not necessary to buy these but they are convenient if you have no access to welding and threading is not permitted. Instead, you can use a threaded or socket weld reducing flange, or if the flanged joint is large enough relative to the tube fitting to fall within the limits imposed by ASME B16.5, you can drill and tap or weld to a blind flange.

In addition to the block valve to isolate the tubing from the pipe, a check valve immediately at the injection point is a good idea in cases where the tube is being used for chemical addition. That's of no use for sample points or instrument taps though.

Tubing needs to be physically safeguarded against mechanical damage in areas where such damage is likely. Physical safeguarding is essential if the tubing is plastic.

 

[LittleInch]

If you have a look at someone like Swagelok, it seems the tube adaptor part is type tested and doesn't have a specific standard like ASME B 16.5 or similar EN code.

In terms of design and fitting good workmanship requirements come into play more than a set standard or "code".

The individual fittings are more of a vendor specific design and generally accepted methods of installation. Instrument tubing is not strong and care needs to be taken to ensure the connections are not subject to excess stress, either axial, sheer, fatigue / vibration etc.

It's a big pity the operator has been injured, and hopefully a root cause analysis will show why and then an inspection of other similar connections undertaken to identify any others where poor design or installation has taken place which could overstress the tubing.

Remember - More details = better answers
Also: If you get a response it's polite to respond to it.

 

[azmanvai]

Thanks guys. Sorry for the late reply as I had to fly to another site for a different problem. Anyway it is sort of worrying that there is no specific codes on tubing and fittings as they are primary containment similar to piping and vessels. I wish to correct my earlier statement that the tubing snapped. Based on latest information there is corrosion around the threaded fitting which was welded to the elbow. That corrosion is now a gaping hole which leaked 98% H2SO4. There is no failure between the tubing and fitting connection. The primary suspect now is galvanic corrosion between the SS fitting and CS elbow. Sample is being sent for failure analysis to determine the damage mechanism. Thanks for your responses!

 

[azmanvai]

Photo of leak

 

[LittleInch]

If the contents weren't concentrated H2SO4 it might be possible, but this looks a really stupid place to sit a connection like that. Apart from the trip hazard / risk of damage to the pipe while someone operates the valve, it is very close to the HAZ of the weld of the elbow and if it was a welded fitting there shouldn't be any corrosion. If however it was a screwed fitting which was "seal welded", then I suspect the acid has simply leaked very slight, become contaminated with water and then just eaten the fitting.

The fittings though are normally much thicker and stronger than their connecting pipes hence why there is nothing more than the type testing from reputable vendors.

You do need to be aware of potential fake fittings though - might be worth trying to find where they came from.

Remember - More details = better answers
Also: If you get a response it's polite to respond to it.

 

[moltenmetal]

What is normally in that pipe, and what is in the tubing?

If you are injecting 98% sulphuric acid into water in an arrangement like that, you've had a corrosion-related failure that has nothing to do with galvanic corrosion and everything to do with heat of dilution/hydration of the acid and the lack of resistance of either carbon or stainless steels to intermediate strength sulphuric acid. You need a proper dilution spool and injection quill, made of materials of construction resistant to what you're producing in that spool.

 

[azmanvai]

From another photo which I will upload tomorrow the corrosion seems to be external going it. The tubing is makeshift impulse line connected to a pump. The content of tubing and spool is H2SO4. No water.

 

[azmanvai]

Closeup pic

 

[LittleInch]

Very difficult to see from that photo, but looks to me like it could be the weld metal which has been eaten away.

Too much damage and unclear what the original fitting was to really judge from afar.

Remember - More details = better answers
Also: If you get a response it's polite to respond to it.

 

[racookpe1978]

No water? There will be a little in almost every startup/repressurize situation - if only from humidity and residual.

 

[moltenmetal]

Carbon steels are only resistant to conc acid at extremely low and uniform velocities- low enough to permit a retained iron sulphate layer to passivate the underlying material from further corrosion.

Leakage rapidly leads to dilution due to pulling water from the atmosphere which causes dilution and rapid external corrosion.