Success eludes; but progress is plentiful on the cannon project

Thursday’s pour was one of those half-full, half-empty glass days. But, in the end, I am certain that the half-full was way more important than the half-empty.

We had a great crew at the furnace all day. As you can see in the video, there was a much greater mix of modern and period clothing than in the past. In addition to Stephen Williams, foundry metallurgist at Newport News Shipbuilding, two apprentices from NNS also were there: Sean Massey and Jonathan Johnson as well as a member of the foundry’s melt crew, Andrew Piskorski. Phillip E. Harrison, industrial designer and operator of Penumbra Design Studio, who also is involved in historical foundry work, came along at Stephen’s request, and Nick White, of Mendip Fireplaces in England, just happened to be in town. Enthusiastic supporters Billy and Nancy Stark were there, along with some other volunteers.

So, there were a variety of other sets of knowledge and experience to add to that of our Historic Trades founders, Suzie and Mike, and others from the Department. All-in-all, it was a day packed with high energy, hard work, free flowing questions, answers, and discussions, and a lot of congenial camaraderie.

The Shipbuilding crew had previously washed the interior of the mold parts with a refractory coating, and the day started out with setting the mold in the casting pit, tilting its top about 10° toward the furnace so that the bronze would run down along its side rather than straight down the middle—should be like pouring a good beer, one of the guys informed us.

As that was being done, we began to ramp up the heat in the furnace, taking its temperature with one of several pyrometers along the way and methodically recording time intervals of fueling, amount of fuel added, and the resulting effect. As this process went along, we learned a lot more about how the furnace was working and just what we needed to do to keep its temperature rising in a gradual curve. We also added a fire in the furnace archway under the melting area to make certain the melting floor was raised to a temperature that would preclude any bronze freezing to it.

In addition to the temperature-measuring equipment, the Shipbuilding folks and Phillip also added a couple other high-tech ingredients. One that proved very helpful was a ceramic cloth insulator donated by Specialty Foundry Supply that allowed us to open and close the fuel feeding opening more readily and also insert behind the feeding door to more precisely control the air intake there. The second was the use of FOSECO insulating sleeve tube, cut in half longitudinally to use as the pouring trough. It is such a good insulator that it doesn’t pull heat from the bronze as it runs, and it is much more reliable and cleaner than the heated clay troughs we have used in the past.

All of these modern bits significantly reduced the variables with which we were contending and let us concentrate on the basic furnace operation.
Morning fueling of “10 at 10”—ten sticks of oak kindling at ten minute intervals—soon changed to varying amounts of wood at experimental intervals. When things really got rolling, it was “10 at 4”.

A couple of times, the furnace temperature seemed to plateau out, but a bit of experimentation got things moving again. A couple of times it dropped, but experimenting with the amount of air drawn through the ash pit, allowing air to enter via the charging door, and the setting of the chimney dampers again led to everything moving nicely along.

This Tin Bronze (90% copper & 10% tin) begins to liquefy at 1550 degrees and then reaches its liquidus at 1850 where it is completely liquid. Stephen and Phillip decided that the best time to pour would be after the metal reached and stabilized between 1900 and 2000 degrees F. We had hoped to do that about 1:00, but given temperature stalls and ups and downs, it actually was approaching 7:00 pm before we decided to go. The last couple of temperature checks using a pyrometer with a probe to dip into the metal gave the desired results, but something was wrong. Stephen and Phillip noticed a difference in how the inside of the melting pan looked.

Just as we were about to pour, they realized that the level of metal was lower than before and dropping ever more rapidly. The melting chamber had developed a leak. We decided we had to pour right then. As seen in the video, the bronze came rushing out of the tap hole, stalled for a while when the iron plug inside blocked the flow, but resumed again when Stephen raked it out of the way. It was a beautiful pour—just what we had hoped for—except that we had only about half, maybe even less, of the bronze needed to fill the mold.

Everyone was disappointed—such a last minute unforeseen—but encouraged nonetheless. On Friday morning, we dug up the mold, broke it away, and found we have what looks like a really nice back end of a light three pounder, from cascabel to about where the trunnions begin. The surface looks good and the casting rings solid. We’ll know more when we cut into it.

In the meantime, though, we are realizing we are way ahead of where we were when we began this pour. We know much more about how to heat the furnace and its idiosyncrasies. We have a better loam mixture for the molds, and using the modern mold wash resulted in a good cast surface. The poling process to burn off impurities worked well, and the mechanics of tap hole, plug, and pouring trough worked extremely well.

In addition, we have been fortunate to form an awesome team to continue the project. All are ready to get the furnace repaired (which looks like it may not be all that difficult) and making another set of molds to try it again this fall. Being out there on the site all day, feeling the anticipation and curiosity, listening to the exchange of ideas, knowing that we were approaching things with much more insight than in the past, and, despite the hard work, seeing people enjoying themselves, each other, and the project is a great example of what the preservation aspect of Historic Trades is all about.


  1. says

    Just saw the video-directed here from the armoury blog. Pretty cool to see a pour in action. I live in PA, surrounded by historic iron furnaces. Good luck-success isn’t too far out of reach at this point!

  2. Christine Hansley says

    To the Cannon Crew,
    I guess the saying “If at first you don’t succeed, try, try again” is in order. I’d like to thank NNS and all the others for their continued giving of help and knowledge with this project. A partial cannon is better than none. It showed you’re on the right track to do this.
    What I don’t understand – is why wait until the fall for another pour? Has it been determined what went wrong with the furnace? Why did a leak occur? Why not try again as soon as the furnace is repaired? I guess I’m just like a little kid. I just want this project to work for you folks.

    Have a good weekend,

    • says

      Hi Chris,

      A couple reasons for the wait. The first, and most important, is the time it takes to make a new set of molds. There are three large, complicated pieces, and it is a long, involved process. If the founders could concentrate on just that, it wouldn’t take so long, but they also have to make and finish other castings in the shop and, of course, welcome and talk to all of our guests. A couple months for the work is not unreasonable. The second is that Williamsburg summers can be steaming, miserably hot, and it’s just a lot more pleasant to try again after the weather cools down a bit. Fortunately, we have some flexibility in our schedule–those eighteenth-century founders producing directly for a war effort didn’t have that luxury.

      We won’t know for sure just why or how the furnace leaked until we start poking around in it, but our guess is that the expansion (and maybe some contraction, too) of the masonry work as it was brought up to temperature caused cracking in the melting area. We expected external cracks, but thought that the refractory construction of the melting area would preclude ones there. We learned otherwise…..

      According to the Art of Gunfounding by Carel deBeer , “…the surface near the fire hole the metal should either have the appearance of being covered with a blanket streaked with branching cracks, or by a film indicating imminent oxidation.” Musly Adds that a greenish-white vapor should be discernible over the bath. (Musley p.151) However that may be, it must have been one of the secrets of the trade for the founder to judge the right moment for the pour. Judging the temperature is by color and requires long experience of looking at lots of molten metal.

      • Christine Hansley says

        To the Cannon Crew,

        Thanks for the info. I do understand summers in williamsburg. Heat and hummidity and more of the same.

        With the war effort, what was the turn around time in the 18th C. for the founders to produce a cannon? Did they have multiple molds to pour at one time?


Leave a Reply

Your email address will not be published. Required fields are marked *