Kelly Anderson,

Quick question about welding the old hole up. When welding is there a particular method such as uphill welding, or can you continuously weld around the circumference of the hole without stopping?

Thanks,
Austin Bryant

While I'm not Kelly Anderson (nor do I play him on TV) I would be willing to bet his answer will sound something like this (from earlier in this thread):





mld

Hi All

The Laying out for Boiler Makers and Plate Fabricators Fifth Edition on page 273 of the book has a basic description of repairing over sized staybolt holes. The book has been scanned by HathiTrust/Google Books and the page may be found here https://babel.hathitrust.org/cgi/pt?id= ... up&seq=283
The book has a lot of good information and I encourage those interested in boiler work to look at it. It has to be a good read Amazon says it is #20,396,838 on the best sellers list!

Robby Peartree

5P = all position = go on around

Weld procedures are specific to the organization and application. There may be several alternatives to a desired end. At NSRM our procedure to welding up a staybolt hole is as follows:
Prepare the hole
Weld the hole
Clean the welded surface
Inspect the hole for sufficient build up and inclusions
Ream and tap
Reinspect for flaws in the weld.

The hole preparation is straight forward, either ream the hole to remove the threads or die grind the hole to remove the threads. If the position is awkward counter sinking the hole can improve access. Clean the surface surrounding the hole as well.

Welding is performed using E-6010, the diameter and amperage is at the discretion of the welder but 1/8” is found to be satisfactory. The welder reaches to the inside of the hole at the water side edge and starting at about 11:00 welds down to the left, across the bottom, and up the right side to the starting point. This first pass is done in such a manner as to leave a bead on the water side of the sheet higher than the surface of the sheet. Sounds odd but it is doable. This insures that there is more than 100% of the sheet thickness available for the threads in the resulting hole. Clean the weld either with a needle gun or die grinder.

The second pass is started next to the first weld at about 6:00 and welded up to the right and around. Do not repeat the starting point for any two or more adjacent welds to eliminate a line of craters. The following welds are made starting at different points around the hole until you lay down the last bead on the outside corner of the hole.

Clean the welds between passes. Start the procedure again, starting from the water side and working out until there is enough material built up to ream a clean hole. Grind the outside surface weld flush with the sheet. We typically ream the hole to ¾” with a bridge reamer to accept a 7/8” staybolt tap. Inspect the reamed hole for inclusions in the weld and repair as needed. Following tapping a careful inspection for inclusions is made and repaired as needed. If an inclusion is found in the threaded hole you must remove all the threads and start over again.

Inspect the water side with a borescope for a bead on the water side.

This is our procedure. Yours may and can vary. It isn’t difficult and with a little practice can be not a time-consuming procedure. There may have several holes in process at once to expedite the procedure. Al the weld is done with E-6010. We have tried using E-7018 but the chance of inclusion isn’t worth the effort.

Tom F/CCDW,

I appreciate you sharing your procedures on this. Thank you for your input.

DC

Here are a couple of interesting pages from the International Library of Technology series, Steam Locomotives and Boilers book, by Charles A. Norton from CNR, 1923.
Those flexible stays are interesting -- did they see much use?
Also, using a nut and copper washer on the fire side of the crown stays is of interest.

The only split and twisted stay that I recall ever seeing was a new old stock one that Linn used to keep on his desk.

Washers and nuts on the fire side was fairly common in British or European practice, as well as in Scotch marine boilers, but I don't recall seeing any in US locomotives. PRR used them on the outside of their notoriously thin roof sheets.

Calling Dave Griner...

The copper washer looks like it would seal up very nicely, so long as the engine stayed hot all the time... but in typical railway museum service, with about 700 fire-ups and cool-downs over the service life, I bet they take a set and leak like a sieve when cold.

I own most of a british austerity 0-6-0st, currently nearing the end of a major boiler overhaul; it has a copper inner firebox and has had (among many other things) a full new set of crown stays, with stay nuts applied inside the firebox, without washers, which is as specified in the original design. Our boilersmith specialises in this particular locomotive type
- he owns one himself and has repaired several others, and confirms that this setup has no particular tendency towards leaks from crown stays. I tried to send a photograph but at the moment it won't stick.

From a book soon to visit the binders Locomotive Boilers By Calvin F. Swingle. Mr Shaffer's Improved Stay Bolt is pictured.

Robby Peartree

I love the goofy stuff people came up with to address problems like stay bolts. But it has been said that there is little satisfaction with gimmickries, and I believe that.

Perhaps this is of interest. I wrote this article a few years back:

A few years back, whilst poking through a local flea market, I chanced upon a couple of booklets put out by the Penberthy Co. called “Penberthy Engineer & Fireman”. Inside, it said they were published monthly by the Penberthy Injector Co., Ltd., Windsor, Ontario, Canada. The booklets are about the size of a Reader’s Digest magazine and filled with articles and interesting ads. Price was twenty five cents a year. One is dated August, 1907 and it contains a very interesting article on high pressure hydrostatic tests done to an old industrial boiler.

The Cheney Box Factory of Lowell, Massachusetts had elected to replace a twenty year old boiler in its works. This company was owned and operated by Frank P. Cheney and worked out of a factory previously operated as the Merrimac Croquet Co. The earlier company had made croquet sets and other products but had closed in 1896. The building was purchased by Cheney. Since the boiler had been installed in 1887, it must have been in situ when Cheney took over. The factory was on Tanner and St. Hyacinth Streets in a section of Lowell known as Ayer’s City. Daniel Ayer worked to develop farm lands to the south of Lowell as an industrial hub. Tanner St. was sometimes shown as Tanners, reflecting an association with the tanning industry. Frank Cheney’s main product line was wooden crates of various sizes used in shipping.

The boiler had been built by the Scannell & Wholey Boiler Works. Bartholomew Scannell and Denis Wholey were both born in Ireland and arrived in the United States in 1843 and 1842 respectively. Scannell had worked at the J. C. Hoadley Boiler Works in Lawrence, Mass., and later at the Stewart & Allen Boiler Works in Worcester. In 1875, he entered into partnership with David M. Dillon making boilers at Fitchburg. He married Mary Wholey. Mary’s brother Denis had been running a grocery business at Lawrence. He and Scannell decided to form a partnership and in 1880, they erected a wood frame boiler shop, forge and storage facilities in Lowell. The site chosen was on Tanner St. directly across St. Hyacinth St. from the Merrimac Croquet Co., later Cheney Box Factory. Railway service was provided to the site by the Old Colony Railway.

This brings us to the events recorded in the “Penberthy Engineer & Fireman”. They were reporting from an article written by E. A. Mores of “National Engineer” magazine. The 1887 boiler was replaced in 1907, after twenty years of continuous service. It was not stated in the article but the replacement was very possibly another Scannell boiler. Frank Cheney was persuaded to allow the old boiler to be put to the test. No reason is given for conducting the tests, other than perhaps a general interest in what might be learned.

On the afternoon of May 11, 1907, a group of some two hundred persons, described as engineers from the Lowell area, gathered to witness the tests. Included among them as invited and official observers were Joseph H. McNeil the Chief State Inspector, local inspectors Ferguson and Hinkley, Inspector Molloy of the Manufacturers Mutual Insurance Co., and Joseph Jones, a former inspector for the Hartford Steam Boiler Inspection & Insurance Co.. Frank Cheney, incidentally, was a member of the National Association of Steam Engineers.

The following data was provided for the 1887 Scannell boiler:
- boiler shell in three courses with single plates, lap seam.
- boiler diameter - 66 inches.
- average thickness of plate as measured from dry sheet - .3885 (approx. 25/64).
- overall length of boiler - 16 feet, 6½ inches.
- length of tubes - 15 feet.
- diameter of tubes - 3 inches.
- number of tubes - 120.
- tubes were lap joint.
- tubes in front head (tube sheet) expanded.
- tubes in rear head beaded.
- number rows of rivets - 2.
- diameter of rivets - ¾ inches.
- diameter of rivet holes - 13/16 inches.
- rivet pitch - 2-3/8 inches.
- number of diagonal bar braces on each end - 1.
- number of head to head braces - 6.
- diameter of braces - 1.25 inches.
- tensile strength of plates assumed to be 55,000 lbs. ( no marks could be found on plates and no record available - inspector McNeil assumed the 55,000 lbs. figure to be accurate.)

Steam restorer Robert Bryce from Manitoba has provided calculations from the data that is available. Accepting the information given, a ballpark working pressure for this boiler could be 95 psi (pounds per square inch). Variables would be that the average thickness was given at .3885 but we do not know what some of the shallower points may have measured. Nor do we know the thickness of the heads (tube sheets). A normal hydrostatic test of this boiler at the time would been in the 140 psi range. Given the date of construction, it is not known or stated whether the plate was from wrought iron or regular boiler steel.

Out in the yard of Cheney’s works, the boiler was readied. Mr. Cooke, the engineer from the Cheney works had set up a Worthington Duplex pump with dimensions of 6 inch by 2-1/8 inch by 6 inch with an outside packed plunger pump connected to it. Cooke would operate this pump which was powered by steam from the works, stated as providing 110 psi. J. J. Markham, chief engineer of the Lowell Electric Light Co., was appointed engineer in charge of the tests. A tested gauge was employed to verify pressure readings.

Upon a final look over the boiler to verify measurements, Mr. Markham gave the signal to Cooke to proceed. The vessel was brought up to 125 psi whereupon a pin head sized stream of water started on top from the rear girth seam. A small weep was also noted on the middle longitudinal seam. It should be noted that the boiler was cold and leaks would be more possible than with a warm or hot boiler. Pressure was then taken up to 250 psi. This caused the middle longitudinal seam to leak badly, beginning near the centre of the seam. Both end plate joints began showing small leakage.

Chief Inspector McNeil had passed a steel measuring tape around the middle plate of the boiler in order to measure any possible increase in the shell diameter. Cooke was directed to continue raising the pressure with the following observations being made:
- 265 psi - middle longitudinal seam leaking badly.
- 290 psi - some leakage in top row of tubes in front head.
- 295 psi - small leakage in top row of tubes in rear head and around fusible plug.
- 305 psi - leaks at front top row of tubes, longitudinal seams, around front head where riveted to shell.

At this point, the leaks in the longitudinal seam were letting out streams of water, some extending as far as 20 feet. It was decided to release the pressure and try caulking the seams before trying higher pressures. Bartholomew Scannell may have been in attendance as he ordered two men sent from his boiler works across the street. They worked for about a half hour and re-caulked the offending seam. The second test was started:
- 310 psi - re-caulked seams started leaking again.
- 325 psi - water from longitudinal seam leak shot away 30 feet.
At this point, it was deemed impossible to raise pressure further and so the pump was stopped and it would be left to the Scannell crew to attempt further repairs. The middle longitudinal seam was re-riveted with new rivets and all seams re-caulked. A larger steam pipe from the Cheney works was sought out. A one inch diameter pipe had been laid, which was full size for the pump. But at 100 feet in length with several elbows, it was understood that full performance could not be achieved given the power losses, so a 1½ inch diameter pipe was laid.

The boiler was drained for inspection. It was found that some of the rivets in the longitudinal seam had been sheared and some broken. The head to head braces showed some small loosening; but there were no signs of bulging on the tube sheets. It was assumed that the rear tube sheet, having been exposed to the hot fire, should have showed more leakage but this was not the case. A straight edge was laid across the top of the boiler and this showed no deflections from the original. By this time, it was the end of the day and it was planned to take up the project another time.

It was May 18th before they could return for the third testing. It was around three in the afternoon when the trials began. Chief Inspector McNeil again placed his tape around the middle plate. These observations were made:
- atmospheric pressure - circumference of middle plate - 17 feet, 4-3/8 inches.
- 375 psi - 17 feet, 4 and 15/16 inches.
At this point, with a 9/16 inch increase in circumference, some rivets were sheared on one end of the man-hole ring, cracking the plate on each side of the ring lengthwise into the rivet holes. The cracks were opened about 1/16 inch. This was the final straw for the old Scannell and the pressure could not be raised any further even though the Worthington pump was working full out. Pressure was released and the boiler drained. The man-hole and hand-hole plates were removed for inspection. Four of the head to head braces had broken and the other two showed bad straining. However, no signs of cracking was observed in the longitudinal seams.

As time moved along, the Cheney box works remained in operation until 1920. Eventually, the property was taken over by Scannell. The Scannell and Wholey partnership was dissolved in 1900 with Scannell taking full control. The company was then called, simply, Scannell Boiler Works. Along with boilers, it made other iron products including penstocks, stairs and fire escapes. In all, the works covered over fifteen acres. Cornelius Scannell, brother of the owner, was a long time foreman of the works. It built many boilers for residential heating and also secured many municipal contracts such as school boilers. Bartholomew Scannell was also appointed municipal inspector of boilers for Lowell. Scannell passed away in 1920 and his sons took over the business. His son, Bartholomew Jr. acted as president and Jr.’s brother Phillip was company treasurer. The company is still in business today and still family owned, a remarkable run.

At the end of the article in Penberthy’s “Engineer & Fireman”, there were no conclusions or opinions stated regarding the tests. Given that the old Scannell boiler had been in continuous service for twenty years, it did not fare all that badly. In design and construction, it was somewhat dated even by 1907 standards. Robert Bryce’s calculations suggest that the theoretical ultimate failure pressure, given factor of safety, could have been around 500 psi. The theoretical yield (where the material would distort but not return to original shape) would then be around 250 psi.; though it appears the boiler exceeded this pressure in the tests. Under very extreme conditions, it retained its core integrity and this testified to the workmanship of Scannell & Wholey back in the earlier days of boiler construction.

For those interested in the contemporary account of the Shaffer staybolts -- an early account is in the Brotherhood of Locomotive Firemen and Enginemen's Magazine, volume 38 (1905). D.L.Shaffer was a brother of Lodge 318, Pittsburg (which is how the city in Pennsylvania was spelled at that time).

The same four cuts, labeled the same #59 to #62, appear in Swingle's Modern American Railway Practice (vol.1) on p,163. This is also downloadable from Google Books if one wants a reference copy.

I have discovered how to obtain a PDF copy of Swingle's Standard American Locomotive Engineering (1244pp, 1912, ed.1914) which he wrote to supplant Forney's Catechism of the Locomotive as a training tool for future Locomotive Firemen and Enginemen. This was scanned at the Library of Congress, but its digital identifier (https://hdlk.loc.gov/loc.gdc/gdcscd.00337757947) returns only 'digitization in progress' and all three physical copies show as 'on internal loan - overdue'.

Fortunately, pandemic or not, you can get a downloaded copy, albeit as a very large file -- call up the 'gallery' view of the pages, and at the bottom you will see a link to download the 'entire PDF'. You have to click the actual link, and then click the little 'go' button beside it. (It is just under 957MB on disk as downloaded!)

What these references appear to show is that brother Swingle was mentioning the Shaffer staybolt as the production of a fellow member of a BLF&E lodge, rather than due to its phenomenal service -- note that it follows discussion of the Tate flexible staybolt in 1908, but has disappeared from the larger volume's discussion four years later. I suspect this is no particular accident.

Aside from ignoring some of the more obvious questions about the true action of the Shaffer 'universal joint', the inventor appears to have labored under the idea that the hollow, larger portion of the bolt went into the fireside sheet, so that it would be 'open at its inner end to the furnace...' and thereby supposedly increase the amount of exposure of the hot gas in the firebox to the water in the stayed space. You, I, and the fellow behind the tree know it would do no such thing for any particular length of time. He also has a version where the entire length of the hollow portion is threaded, putting heaven knows how many stress raisers and corrosion-crack initiation points into the bolt in the waterspace. This frankly doesn't bode well that the universal joint that has just been used to drive the (I thought it was the inner, but apparently it isn't) smaller end of the bolt to where it can be 'riveted' is going to provide the proper combination of spacing and bending action to give flexibility in a wet environment with typical scaling contaminants present. How he proposes to clean the joint in service is even less clear; perhaps he assumes differential expansion will keep it cracked free. Personally I see tremendous issues with wear at the mating surfaces of what will be a stressed universal connection as set. Did he think it would unwind as the end was riveted over? -- perhaps it would.

I'd be interested to see the long-term serviceability and failure modes of those twisted-slot staybolts, perhaps especially the types forged with two 'legs' that were then twisted to shape. All this was in the age before practical NDT; how would you know the iron had been shaped correctly without defect, or opening up surface cracks for corrosion or stress raising? Or make sure all the little twisted voids didn't fill with scale? Somebody also has to explain to me how telltale holes work with bolts made this way... I don't see how.

Now, I presume the idea is that a universal-joint design that is 'fillet-welded' with good full penetration (e.g. with properly blanketed laser keyhole welding with adjusted preheat) from both 'outsides' of the waterspace could be aligned to produce hinging in the 'correct' plane of calculated differential expansion. (To get this you would matchmark the outside of each 'half' along its axis of action, and fix first one end, then rotate the other and fix it, via tack welding before proceeding with the circumferential automagical weld). The problem is that I think you'd have the same trouble with water-treatment action in the hinged joint, and much the same difficulty with cleaning or inspection. You might get around some of this with selective plating or hard coating of the joint, or using manganese or other sacrificial anodes distributed in the boiler structure. I'm sure Mr. Peartree has some reasonable answers for how this might be made metallurgically practical. I see in the '47 Cyc that molybdenum steel was touted as a material for staybolts and also ideal for welding fabrication (p.241)