Not logged inAmerican Welding Society Forum
Forum AWS Website Help Search Login
Up Topic Welding Industry / Metallurgy / Tube Failure
- - By bsalonek Date 10-02-2008 19:38
Looking for help on the following problem   We are stumped by the failure of tubes on our waste heat steam generator.  The attached picture give you and idea of the failure. The unit has 1153 tubes in it, and it appears that 95% of the have failed.  We have never seen this type of failure before, and are looking for possible causes.  The unit opperates at 650 PSI at 557 degrees F.  Our initial thoughts were that teh unit was run without water, however operationg data shows no changes in make up water, or water levels in the steam drum.  The only indication of this failure was a sudden increase in FD fan load.  We are looking for comments from anyone that may have seen this type of failure before.
Parent - - By G.S.Crisi (****) Date 10-02-2008 19:54
Looks like the tubes collapsed due to external pressure, don't they?
Giovanni S. Crisi
Sao Paulo - Brazil
Parent - - By hogan (****) Date 10-02-2008 20:45
could it also be internal vacuum?
Parent - - By G.S.Crisi (****) Date 10-02-2008 22:31
Yes, that's what I wanted to mean by external pressure. The tubes were submitted to internal vacuum and consequently, external atmospheric pressure. How can this happen? The boiler is shut down and let it cool down with all valves closed. The steam condensates, creating an internal vacuum. In order to prevent this, the vent valve should be opened when the internal temperature reaches 100ºC (212ºF), when the water within the boiler won't flash any more.
However, I doubt that a bundle of tubes that were designed to withstand 650 psig (roughly 45 bar) internal pressure will collapse just because of atmospheric external pressure.  
Giovanni S. Crisi
Parent - By Joseph P. Kane (****) Date 10-02-2008 23:09
I tied to make a straight line tubesheet to tube sheet heat exchanger once, and the immovability between the two tube sheets caused the tubes to buckle down by the expanded ends.   So much for amateur engineering!

These collapsed ends look like heart valves. I do not think a 14.7 PSI vacuum caused this type of collapse. I do not know the actual wall thickness of the tubes, but even if they got red hot, they wouldn't collapse like that just from a less than one bar pressure differential!  I think this was a very high overpressure condition.
Parent - - By vittorio (*) Date 10-03-2008 09:53
just one thing.. maybe it's wrong.. you said that there was a sudden increase in FD fan load, this mean, i think, a sudden increase in temperature and this could be due lower heat exchange due to low flow of water.. it could be that the temperature reach  a value so high that the material loses mechanical strength and then collapse under a "realtively low" pressure of 45 bar?
it would be interesting to see the curve of mechanical characteristic for that material and if it's possible to find at what temperature can collapse under that pressure..

vittorio

http://www.corrosionist.com
Parent - - By bsalonek Date 10-03-2008 10:36
Gentleman, I want to thank you for your comments.

Here are some additional facts regarding our waste heat steam generator. We use flue gases from an up stream reformer with an inlet gas temperature of approx. 1650 - 1700 degrees F. The gas flows through the tubes thereby heating the water and generating our steam.  The tube materials are SA-179A with a .120 min wall thickness with a diameter of 1.500".  Actual measurements during fabrication noted that the tube walls averages 0.126 thick.  The tube sheet is 4.125 thick and is made of SA-516-70N material.  The tube holes average ID are 1.520 with an average clearance distance of .010 between the tubes and tube sheet.  The tubes where expanded using a hydro expander at a pressure of 30,000 psi giving us an average wall reduction in the expansion area of 6-7%. The reduction area is approx 7/16" from the face of the tube sheet and continues on for 3/4-1"  in length.  The weld detail has is using a 3/16" j bevel from the face of the tube sheet and then adding a .125 fillet weld over that, thereby designing it around a full strength joint type.

We initially felt that there as a huge temperature problem on the face of the tube sheet, but current FEA analysis indicate that the face of the tube sheet has not gotten over 845 degrees F.  Our current train of thoughts has us believing that some how, water may have gotten passed the expanded area and gotten to the void area that exists between the root of the weld and the expanded area.  Once the liquid is there, is has flashed off thereby creating an external pressure area on the outside of the tube.  However I have calculate this possible void area, and determined that is approx .012 cubic inches in volume.  Based on this volume I do not know if it would generate enough steam to cause the expansion of the tubes.  I also believe that there some how has to be a temperature element to the problem to cause the deformation shown.

Bruce Salonek
Parent - By dasimonds (**) Date 10-12-2008 15:02
Why weld the end of the tubes? We routinely replace generating sections, and I don't recall a time where the tubes were both rolled and welded. As you said, .012 cu.in. isn't a lot of volume, but with a 1300 time volume change, how much volume do you need to have a problem?
Were the welds NDT'd to ensure no leaks prior to service?
Any creep damage on the portion of the tubes in the gas pass?
Hope this helps.
Dale Simonds
Parent - By G.S.Crisi (****) Date 10-03-2008 15:28
Vittorio,
the "relatively low pressure" Joe Kane and me are talking about is not the 45 bar internal pressure; it's the atmospheric external pressure.
Giovanni S. Crisi

PS: parli italiano?
Up Topic Welding Industry / Metallurgy / Tube Failure

Powered by mwForum 2.29.2 © 1999-2013 Markus Wichitill