Tek-Tips is the largest IT community on the Internet today!
Members share and learn making Tek-Tips Forums the best source of peer-reviewed technical information on the Internet!
-
Congratulations cowski on being selected by the Eng-Tips community for having the most helpful posts in the forums last week. Way to Go!
Spliced Tubes in Heat Exchanger 1
- Thread starter weeeds
- Start date
weeeds;
There are no rules in the Code that prohibit this because it is poor engineering practice. Why?
A butt weld in tubing creates inherent metallurgical notch effects (heat from welding alters the original base metal structure) that can result in preferred sites for fatigue crack initiation and propagation in service. Unless butt welds are absolutely necessary and accounted for in design, why have the increased risk of introducing a potential site of failure in service?
When butt welding in tubing is performed there is the possibility of defects introduced. To account for this, volumetric NDT should be performed to ensure highest quality to avoid Hx tube failure in service.
In your case, if you absolutely need to butt weld a Hx tube section in place, as a minimum, the crown of a full penetration butt weld would need to be removed and the OD surface ground flush to eliminate any stress risers (weld toe undercut). Access could limit this method.
Volumetric NDT of the flush ground butt weld is a must to ensure no weld defects in an otherwise weld free tube.
Is this enough to persuade you that this is not good engineering practice????
There are no rules in the Code that prohibit this because it is poor engineering practice. Why?
A butt weld in tubing creates inherent metallurgical notch effects (heat from welding alters the original base metal structure) that can result in preferred sites for fatigue crack initiation and propagation in service. Unless butt welds are absolutely necessary and accounted for in design, why have the increased risk of introducing a potential site of failure in service?
When butt welding in tubing is performed there is the possibility of defects introduced. To account for this, volumetric NDT should be performed to ensure highest quality to avoid Hx tube failure in service.
In your case, if you absolutely need to butt weld a Hx tube section in place, as a minimum, the crown of a full penetration butt weld would need to be removed and the OD surface ground flush to eliminate any stress risers (weld toe undercut). Access could limit this method.
Volumetric NDT of the flush ground butt weld is a must to ensure no weld defects in an otherwise weld free tube.
Is this enough to persuade you that this is not good engineering practice????
-
1
- #3
I'm going to diverge from the post by metengr by saying that the splicing of Hx tubes is a standard engineering practice. One of the most common practices is what is called safeending a tube to prevent attack either on either surfaces. For many years we would safend, usually 18" to 24", Austenitic SS tubes in vertical Hxs' with some material that was immune to SCC. We did the same thing to prevent damage from MIC on the external surface of Austenitic SS. Recently we safended Hast C tubes 6" to prevent catastrophic attack by the a process material. We also repair/refurbish some steam heated fixed tubesheet calanderia where we have to replace 12" of a 1.125" 10 ga SS tube when it is necessary to replace an expansion joint.
We do all this work in house making autogenous welds using orbital welders. There is generally no finishing of the weld bead required. We individually bubble test every modified tube at 100 psig or hydrotest at operating pressure if higher.
I cannot recall any problems either from mechanical or corrosion issues over the last 20 years.
We do all this work in house making autogenous welds using orbital welders. There is generally no finishing of the weld bead required. We individually bubble test every modified tube at 100 psig or hydrotest at operating pressure if higher.
I cannot recall any problems either from mechanical or corrosion issues over the last 20 years.
unclesyd
Interesting information and perspective. I was unaware of this specific practice in the chemical process industry.
However, I would not permit it on any of our Hx's because there is too much risk for having problems with limited access, cutting your way into a tight bundle and having to weld your way out.
Interesting information and perspective. I was unaware of this specific practice in the chemical process industry.
However, I would not permit it on any of our Hx's because there is too much risk for having problems with limited access, cutting your way into a tight bundle and having to weld your way out.
metengr,
I surmised that was where you were coming from. The largest Hx that I've worked on was about 60" dia, with 3/4" 16 ga tubes. We have also used a lot of inserts on several larger Hxs.
The above referenced to Hast C was casebook example of a scramble to save over half a million dollars worth of Hxs'. Even though I've mentioned this previously I think it's worth a little refresher. In service testing had shown Hast C was a perfect material for our process, both from the process side and it was immune to MIC on the CTW side. Two Hxs' were purchased and one installed to replace a badly leaking one. The new Hast C bundle showed signs of leaking within two weeks and at the end of two months was pouring like a sieve. When it was opened for inspection it was evident what the problem was, the Hast C was almost completely gone to the depth of the tube sheet. 1/32" or less below the tube sheet the tubes were pristine. A frantic effort was made to get some inserts installed. I found AL6XN inserts, this material had never been tested in this process so it was SWAG on my part. The inserts worked and inserts were installed in the second bundle prior to being installed. Later to salvage these bundles they were cut down and the Hast C was safe ended at the top tube sheet.
Here is a little bit of information if anyone ever needs to splice a tube. To make a quality autogenous weld besides being clean the ends of each tube has to absolutely square and welded with absolutely no gap in one continuous pass.
Inserts:
I surmised that was where you were coming from. The largest Hx that I've worked on was about 60" dia, with 3/4" 16 ga tubes. We have also used a lot of inserts on several larger Hxs.
The above referenced to Hast C was casebook example of a scramble to save over half a million dollars worth of Hxs'. Even though I've mentioned this previously I think it's worth a little refresher. In service testing had shown Hast C was a perfect material for our process, both from the process side and it was immune to MIC on the CTW side. Two Hxs' were purchased and one installed to replace a badly leaking one. The new Hast C bundle showed signs of leaking within two weeks and at the end of two months was pouring like a sieve. When it was opened for inspection it was evident what the problem was, the Hast C was almost completely gone to the depth of the tube sheet. 1/32" or less below the tube sheet the tubes were pristine. A frantic effort was made to get some inserts installed. I found AL6XN inserts, this material had never been tested in this process so it was SWAG on my part. The inserts worked and inserts were installed in the second bundle prior to being installed. Later to salvage these bundles they were cut down and the Hast C was safe ended at the top tube sheet.
Here is a little bit of information if anyone ever needs to splice a tube. To make a quality autogenous weld besides being clean the ends of each tube has to absolutely square and welded with absolutely no gap in one continuous pass.
Inserts:
UncleSyd,
You have done some stuff with Hx's that would curl the hair of most Hx folks that I know and I say that with the deepest of respect and admiration for your experience. I value your input. Wow, what a story. How many years did that one take off your life?
Metengr, do you even hava any Hx's in your fleet with 10 ga tubing? (Not counting the top heater on any supercriticals.)
rmw
You have done some stuff with Hx's that would curl the hair of most Hx folks that I know and I say that with the deepest of respect and admiration for your experience. I value your input. Wow, what a story. How many years did that one take off your life?
Metengr, do you even hava any Hx's in your fleet with 10 ga tubing? (Not counting the top heater on any supercriticals.)
rmw
- Thread starter
- #7
- Status
Similar threads
- Question
- Locked
- Question
- Locked
- Question