VFT tube leak

fredrosse · Post by **fredrosse** » Tue Dec 31, 2019 2:09 am

When I had my boiler tubesheets made, also with CNC machinery, the holes were exactly 1.250 diameter, and sliding a tube into the tubesheet was next to impossible, the tubes were almost exactly 1.250 diameter also. My bad for not specifying a couple thousanths diametrical clearance.

I began sanding off tubes on the OD, that worked OK to get a tube into the tube sheet, but the entire length of each tube would need sanding to assemble the VFT boiler. Repeating this for each of the boiler's 48 tubes was not what I wanted to do. I have a 1.250 reamer, so I wrapped a short piece of 0.004 inch (0.1mm) steel shim stock about half way around the reamer cutting teeth, and used that setup to dress all the tubesheet holes on the drill press. This beautifully opened all the holes to about 1.253, and boiler assembly was then very easy.

Lopez Mike · Post by **Lopez Mike** » Tue Dec 31, 2019 9:39 am

You're right about A178 having a seam. I forgot.

I must give this some thought. . .

I don't think I buy the idea of work hardening such low carbon steel with such a small expansion. The yield point must be around a strain of .001 or the tubes wouldn't stay expanded in the tube sheet. The distance from the yield point to the the tensile limit must be about twice that I'm pretty sure.

I think this will be an experiment for some rainy day. Swage out a tube a large amount, say to a strain of .020 or more. Bore a piece of scrap steel plate to a clearance of the usual .001/.002 and then roll it in. It will be easy to test the grip with a press and a scale. How to decide if there has been work hardening is a bit harder. Maybe a crude tester (Rockwell B?) made from a punch. In fact, comparing the size of the marks from my automatic punch might be a good way for a comparison even if it doesn't give an exact number.

I am often amazed at how little practical testing has been done in many engineering fields. I was chasing back some testing data on some radio antennas and found that the original tests has been done 90 years ago and nothing since. Every text book quoted the same data set! When I bothered to try replicating the test I found significant discrepancies. I then found the actual test logs from long ago and guess what? The engineers notes indicated that they were most dissatisfied with the narrow range of their tests and urged later experimenters to enlarge on their work. It never happened at least not in a published way.

Don't be afraid to experiment. But don't go blowing yourself up!

RGSP · Post by **RGSP** » Tue Dec 31, 2019 10:43 am

Lopez Mike wrote: ↑
Tue Dec 31, 2019 9:39 am
Don't be afraid to experiment. But don't go blowing yourself up!

Surely the second half of that line can be reversed when politicians are the experimenters?

Willingale Tubes are possibly the main suppliers of finished boiler tubes in the UK, and they go to great lengths with their tube bending systems so that wall thickness is NOT significantly reduced, even with fairly tight bends. I don't know how they do it, but they're only 10 miles away from here, so I must go and talk to them some time and find out. From talking to someone casually (who worked there as a labourer), they can, and sometimes do, roll the centre sections of straight boiler tubes so the the diameter is reduced a little, the wall thickness increased a little, and the ends left at full diameter, untouched and in a good state for expanding into a tube plate.

Sounds like a nice solution, but I suspect there are hidden drawbacks to it.

Lopez Mike · Post by **Lopez Mike** » Tue Dec 31, 2019 3:51 pm

I never thought of reducing the diameter in the center portion of a tube. Interesting.

Further thoughts . . .

Why are we trying to minimize the wall thickness of our tubes? I've always assumed that it was to get more heat transfer from the flame to the water. But I strongly suspect that a much lossier interface is between the surface of the tube and the water or gasses. Surely steel conducts heat far better than water and even better than a layer of steam bubbles or hot gasses flowing past.

When I did the calculations for my small locomotive boiler (fire tube type) the weakest link was the collapse point of the tubes. I'll have to go drag out my calculation sheets but if I remember right the material data entered into the equation wasn't the tensile strength of the tube but its modulus as the critical mode was Euler failure (Getting out of column). I cannot remember but I suspect that there would be disproportionate payoffs in pressure holding ability from increases in boiler tube wall thicknesses.

Commentary please!

RNoe · Post by **RNoe** » Tue Dec 31, 2019 5:01 pm

Mike:
Interesting statement that your locomotive fire tube boiler's weakest link was the collapse point of the tubes.
On the several boilers including (two of mine) that had the calculations done per the Washington State supplied strength calculations, the weakest point was always the back head, which typically has the least amount of staying. That said, the safety factor was still at least 5 times the MAWP.

Considering we are sitting behind that backhead with legs spread, we have high confidence in the calculations and fabrication of the boilers...
And in 100 years of State pressure vessel records, there has never been a hobby boiler explosion, or injured operator from failures of those pressure vessels.

About 7 years ago I was asked by the WA State Pressure Vessel Inspectors' Association 50th anniversary meeting to present a talk on Hobby Boilers. My research for that presentation found the above stated information. I brought a completed boiler (from my 1/8 scale Mikado, 300+ lbs.) for them to inspect, along with examples of the safety valves used (Superscale), and projected lots of photos and slides of my research and our hobby. It was a very popular talk with them. And good for our Live Steam Hobby, which has since been mostly deregulated by WA State because of our perfect safety record.
So far, so good, if we maintain our criteria for design, fabrication, operation and maintenance of said boilers.
RussN

Steamboat Mike · Post by **Steamboat Mike** » Tue Dec 31, 2019 5:16 pm

<t>Hello, 
This is a very interesting discussion on tubes and rolling them. I would add the following thoughts: 
 
I seem to remember advice from the Original Book of Steam of my youth, the Audel's Marine Engineers Guide, where the author, Frank Graham, goes through the design of the steam launch Stornaway, with a vertical firetube boiler with 1" tubes. The tubes were so small because the intended fuel was to be sootless anthracite coal. In his plan he specified a hole in tubesheet hole of 1-1/32" (1.032") for the tubes to make insertion easy but more importantly to make removal possible. .032" oversized is only a gap of .015" on the diameter but makes a huge difference on the ease of tube handling. I have always believed that actual boiler tube is made from steel with a ductility specification that allows for significant stretching and thinning without problems, as distinct from schedule 40 pipe, either welded or seamless, when used as boiler tube. 
 
As to rolling the tubes initially, I remember something about there being a specification for how much to thin the wall of the tube when rolling them in. This can be measured simply by knowing the actual tubesheet bore and the thickness of the unrolled tube wall and the doing the appropriate math to arrive at the finished opening of the rolled tube. Using a telescoping gauge and micrometer can easily check progress and end result of the rolling. Once this is established I think that the torque reading of the roller can be used as a reliable guide to the proper amount of expanding of the tube for successive tubes. The beauty of the wall thinning measurement is its repeatability. It does take some time but if I am working on my own boiler I really don't care how long it takes, I just want the best outcome possible. 
 
For locating weepers I have found drying the boiler and applying compressed air and using a spray bottle of detergent and water to be very helpful in locating the source. With water everything gets wet and shiny making pinpointing the source difficult. 
 
On three vertical firetube boilers I have had so far in Catawissa, one had seal welded 1-1/4" tubes (Benson Mountain, 1999) one had 2" rolled and beaded tubes ( Kalamazoo, 1924), and the current boiler (Boschan Boilerworks, 2009) has 1-1/4" rolled and beaded tubes. 
I have never had a rolled joint issue with any of these boilers, though I am anticipating the "Firebox Shower of actual tube failure at ant time soon, based on observation of the tubes when seen through the handholes. I use raw river water with only minimal treatment because the water is quite pure but has a lot of oxygen in it, but that is another topic for another thread. 
 
 
Hope this information is of use, happy steamboating, best regards, Steamboat Mike</t>

fredrosse · Post by **fredrosse** » Wed Jan 01, 2020 2:45 am

From Lopez Mike:

"I am often amazed at how little practical testing has been done in many engineering fields. I was chasing back some testing data on some radio antennas and found that the original tests has been done 90 years ago and nothing since. Every text book quoted the same data set! When I bothered to try replicating the test I found significant discrepancies. I then found the actual test logs from long ago and guess what? The engineers notes indicated that they were most dissatisfied with the narrow range of their tests and urged later experimenters to enlarge on their work. It never happened at least not in a published way."

In fact, much detailed research and testing is usually performed throughout, however the companies that get the good data now keep it to themselves, rather than sharing it with easily obtained references. I work in the utility power industry, and most of this industry consisted of electric utility companies that inherently had a local monopoly with their captive customers. This was the case from about 1900 to 1994, and technical information was developed with generous research in almost all facets of power generation. This information was readily available throughout the industry, because the monopoly companies had no motive to keep anything private.

That all changed in 1994, when the government decided that the electric grid was big enough to handle routing of large amounts of power. The utility monopolies were turned into independent competing organizations. "Independent System Operators" were formed, and they were charged with getting competitive pricing for any utility that offered power to the grid. From that time on, utility companies have hidden technical data, not wanting to give any advantage to their competitors. Unfortunately the government scheme to bring free competition to the industry failed to give any rate reductions to the consumers, and the free enterprise manipulators soon learned how to manipulate the system to maximize their profits, sometimes obscenely (ie ENRON for one).

wsmcycle · Post by **wsmcycle** » Thu Jan 02, 2020 10:43 pm

Re: VFT tube leak
Post by Lopez Mike » Fri Dec 27, 2019 2:10 am

A 1/2' tube sheet would be fairly startling. How big is this boiler anyway!!!

If it's as much as 1/4" I'll be interested to hear. More like 3/16" I suspect.

Steamboat Mike · Post by **Steamboat Mike** » Sat Jan 04, 2020 1:52 pm

Add on to my comments above:
With regard to tube sheet thickness, on booth the Bensom Mountain and the Boschan Boiler Works boilers, the tube sheets are
SA 516-70 steel 1/2" thick. I never thought this thickness was anything unusual. On the Benson I cut the tube holes with a trepanning head in the Bridgeport, I believe the Boschan boiler was cut with a slug cutter. Both methods give a "bored finish". The Benson is rolled, seal welded and re rolled both ends, the Boschan is rolled and beaded. I have never had an issue with weeping or leakage on either one.

Lopez Mike · Post by **Lopez Mike** » Sat Jan 04, 2020 3:27 pm

I'm amazed.

Unlike many situations where the loads build up exponentially as with a beam, a tube sheet on a VFT only has local loads that are resolved in tension of the tubes. Just a bunch of hollow stay bolts in effect.

Unless the rolling and welding cannot stand loads of a couple of hundred pounds, I'm not sure why the sheet needs to be that thick.

Not being a structural engineer I think I must be missing something here.