Seamless vs Welded Tubes 4

[sdwndr]

What is the main critera used to determine whether a shell and tube heat exchanger should have welded or seamless tubes? When can you justify the additional cost of the seamless tubes? Thanks in advance.

 

[btrueblood]

Mainly if there are issues with fouling on the tube side. Seamless tubes are easier to rod out, and will tend to foul more slowly due to better surface finish.

From a heat transfer standpoint, we found better results with welded tubing, which we attributed to increased turbulence in the flow on the tube side, for a specific air/water heat exchanger...YMMV. And the difference in the overall heat transfer coefficient was a few percent, nearly un-measurable without statistical data over a fairly large number of HX's.

 

[metengr]

Unless I am losing my mind, most heat exchangers in the Power generation business use tubes, not pipe, and tubes are seam welded. The rather thin gage is not conducive to seamless tube fabrication.

Most of the heat exchanger seam welded tube product lines are excellent quality, you can't see the seam weld.

 

[rmw]

Seamless is not unheard of for Power Generation tubing but it sure is overkill for the reasons you state about the quality of welded tubing.

rmw

 

[geodesia]

Unless I am losing my mind, most heat exchangers in the Power generation business use tubes, not pipe, and tubes are seam welded. The rather thin gage is not conducive to seamless tube fabrication.

Most of the heat exchanger seam welded tube product lines are excellent quality, you can't see the seam weld.


metereng..
tubes seamless are very common for shell and tubes HE. AS an example, please refer to SA-213 material.
Thickness can be BWG16, BWG18, BWG 20 or even lower...no problem
For some special services, as lethel service, sour service, etc...seamless tubes are specified by many users/licensors

 

[unclesyd]

Another reason for using seamless tubing is that in a lot of corrosive services there could be some preferential attack in the weld area.

In our case of using 304L tubing in HNO3/Organic service we always called for seamless tubes in Hx's due to preferential corrosion in the weld area.

At one time we used 321 SS welded tubing in Hx service and were eaten alive by knife line attack after a small process change.

In a molten organic acid service we experienced preferential attack on the weld in welded tubes when we ventured into the area where we started to corrode the tubes.

We had a new process where some Hx's were improperly stored with bad water during a delay in startup. The SS tube suffered a bad case of MIC mainly in the weld area. The weld line looked like a city street with bug condo's on both sides.

 

[moltenmetal]

unclesyd: if the tubing is annealed after welding and pickled/passivated prior to use, what's going on at the weld area to cause the preferential attack?

 

[unclesyd]

mltenmetal,
A little history background on my previous post and your question.

Our process to make an organic acid by the oxidation of an alcohol with 65% HNO3 with Cu and V catalyst and temperatures up to 120C is probably one of the most corrosive services where you can still use an Austenitic Stainless Steel, which in our case is primarily 304 ELC. Early on when we operated at around 100C there was very few problem with the welds and we even used welded pipe and tube. We always pickled and passivated new pipe and tube materials. As the temperature increased to 212C we started noticing the welds being defined after a very short exposure to the process. At this time we started laboratory and field investigation the sudden change in corrosion resistance of the 304L. We followed several false leads until there was some slight resolution of the problem. By various studies and somewhat shake correlations we decreased the the C and S to lower levels, especially C to 0.020% reported to three decimals along with requiring seamless pipe and tube. We stopped on site pickling and passivating . We saw enough improvement in the performance of material that we allowed the process people to increase the temperature to 215C where we again saw a little improvement in actual performance. Above 215C decreasing the C to 0.015 max has very little benefit along with lower S an P. As the process temperature approached 120C the corrosion rate was asymptotic at 123C.

In our lab work using Huey and Heatflux test and optical metallography were we unable to differentiate the behavior of the material in at the various temperature whether passivated or not. We then had a few samples installed in the process system at various temperatures. The testing results were the same as real time. We weren't able to use actual process concentrations at temperature as the process material will detonate at around 128C.

The only thing we noticed was that grain size and orientation could play a part in the corrosion equation. The SS we used was very susceptible to end grain attack. Immersion samples preferentially corroded on the edge at the higher temperatures before we saw an attack on the HAZ and base metal. As the weld proper corrosion rate was slow to initiate and progressed slowly we instituted a procedure were we made TIG fusion pass on any exposed edges. After the corrosion started on the HAZ it would progress in and then attack the end grains of the base metal. There was general corrosion of the plate material going on at the same time. Even on corroded material a fusion pass would sloe the initiation of corrosion. In starting up of new equipment we initiated a procedure where we started up on 65% HNO3 near our operating temperatures.

During the early part of our testing we saw that in our service was little if any benefit from passivation in this very corrosive media. This was later verified by electrochemical and SEM studies. The same was verified in the 316 SS material mentioned in my first post. This is a process where we melted and vaporized the acid made in the the process where we used 304L SS as the MOC. Once this acid begins to melt it will literally dissolve 304L SS passivated or not. We tried to correlate this corrosion mechanism back to the oxidation process but were unable to do so. We were also assured that if any any material would reach this temperature it would be in the middle of an explosion. The resistance of 316 SS was good to 192C, then we went to Hast C, then to Ti and finally to Copper as the temperature rose.

The only real improvement we saw with either the 304L or the 316 material was when we increased the Cr in the surface to 35 to 40% and the Mo to 15% in the 316% of which neither was practical.

Our overall general conclusion form all our testing was that if you were going to have any corrosion rate of more than 0.001 IPY forced passivation was of little or no benefit. Having said that I know there are some process where it will make a difference and we still require it in all areas other then the oxidation area.

PS:
All this testing was under a plant test procedure that lasted for 10 years and it fell upon we to write the 50 page closing report sans computer. The overall conclusion was that Austenitic SS was not viable as a MOC at a process temperature of over 115C. The only material to use was Ti at any cost, a whole "nother" study.

 

[insult2injury]

Also, some clients in the power industry prefer seamless tubing for high pressure applications (i.e. feedwater heaters).

I2I

 

[rmw]

Seamless tubes are commonly used in the top point heater in supercritical cycles where the water pressure can be in the 5000 psi range, but not exclusively.

rmw

 

[unclesyd]

The temperetures in my first part of the post should read. 212F = 120C

215F = 115C

 

[786392]

Honourable Sir,
Mathematically speaking;Should these not be something like?
115 Celsius = 239 Degrees F
&
120 Celsius = 248 Degrees F

Since 100 Celsius is 212 Degrees F I understand and 0 Celsius corresponds to 32 Degrees F


Best Regards
Qalander(Chem)

 

[moltenmetal]

There's one obvious advantage to seamless versus seamed tubing, and that's the difference in longitudinal joint efficiency. Seamless permits the use of thinner tube for the same duty. But that doesn't necessarily translate to a more cost-effective exchanger fabrication.

Seamless tubes and pipe are "preferred" by lots of people who don't bother to adequately question WHY they're preferred- that's been my experience. Suppliers are usually happy to provide the (usually) more expensive material if you're willing to pay for it in cost and schedule terms.

Exchanger mass manufacturers prefer welded seam tubing because it's cheaper to buy and to use: it can be obtained in continuous reels, straightened and cut to required length, minimizing waste. That's a big advantage if you make a lot of exchangers.

I've seen plenty of expensive seamless tubing and pipe with quality problems which materially affected our ability to use it: excessive ID roughness, wall thickness variation, internal deposits of unknown composition and origin etc. To assume that seamless somehow also implies "higher quality" does not match our experience.

So when somebody like unclesyd comes up with a preference for seamless, I'm interested to know why- because I know it will be for good and well-considered reasons, rather than "because it's in the spec".

 

[geodesia]

Moltenmetal


REgarding the longitudinal efficiency take into account that most materials for welded tubes for heat exchangers requires RT and efficieny is 1, identical to seamless....you cannot use thinner thcikness changing from welded to seamless

 

[EdStainless]

In all cases that are not creep limited, a welded and then cold drawn SS tube will be superior to a seamless product. It will be smoother, more constant in wall, and be testable to higher sensitivities (smaller defects).
We have supplied welded and drawn 304 for many HP FWHs where the original seamless tubing had failed. In fact for super critical HP units w&d is the standard product.

Are you sure that your 304L tubing had NO residual delta ferrite in the welds? I have seen this in a number of acid applications.

Have you ever looked at a true super-ferritic alloy for this service. 27% Cr, 4% Mo?

= = = = = = = = = = = = = = = = = = = =
Plymouth Tube

 

[metengr]

geodesia;
Moltenmetalis correct with some clarification. If you review ASME B&PV Code, Section II, Part D, the allowable stress lines for material are based on product form, in addition to Spec No and Type/Grade. Allowable stress lines are used in all Code calcs - boilers, pressure vessels and Hx's.

So, if you compare the stress lines for, lets say, seamless low alloy steel tube versus welded low alloy steel tube you will see a difference in stress values above 850 deg F. The reason is that the Code assigns a 0.85 factor for seam welded tubing in the creep range (time dependent properties).

One needs to read the Notes in selecting material and understanding the rationale in differences for stress lines.