Thread Tools
Dec 04, 2009, 01:15 PM
Damp and Dizzy member
Brooks's Avatar

Live steam A.J. Goddard sternwheeler

This will be a pond-scale model of A.J.Goddard. The sternwheeler was used on the Yukon during the gold rush, 1898. The ship sank in a storm on the river, and was recently found, more or less intact. The nautical archeologists have made interpretive reconstruction drawings of the ship, and there is one photo before she sank, which are my sole sources of information. More of the ship's history, and the plans drawings url, can be found in this thread (thanks to Ken NJ for posting the information about this ship):

The model is 1:25 scale (2' hull), and will be live steam powered using the Midwest Heritage single-acting engine and boiler. The boiler will be converted to the "pot boiler" style that has proven effective in other steamers I have built or assisted with. Firing is still under consideration: Sterno, alcohol, or gas. I will use a chain & sprocket setup similar to the BAGRS G-scale locomotive. Gears will allow me to experiment with gear ratios. The chain noise will be a bonus for the intended skipper, I think.
BAGRS instructions:

The BAGRS uses about 1:7 gear ratio (wheel : engine). It's desirable to keep up the speed of an oscillator engine like the Midwest. Real steamers of this era used direct connection of engine to paddle, via a Pitman arm. Later steamers used chain drive, so the use of chain for this model is not entirely bogus - the real vessels so driven were called "bicycle boats" since they used roller chain. "Bicycle" paddlewheel towboats on the Mississippi were powered by gasoline, diesel, and diesel electric. A steamer over on Paddleducks (kiwimodeller Ian) uses 1:3.5 ratio, with a double acting engine; his steamer is hard to slow down, he says..I'm hoping thats a good omen :-)

I will simplify a lot of the detail. In particular, the canopy will not be installed. I hope the lack of canopy will make the boat less suseptible to wind (which sank the original boat). The paddlewheel framing is made of wood, rather than metal like the original. The spokes are more robust than typical wood spokes, partly for durability, and partly because I've never made a paddlewheel before.
Last edited by Brooks; Jan 03, 2010 at 10:33 AM.
Sign up now
to remove ads between posts
Dec 04, 2009, 04:16 PM
Retired for now
You may leave off the canopy but you would want the framing for it as it is also the hogging equipment. Pete
Dec 04, 2009, 05:49 PM
Grumpa Tom
Kmot's Avatar
Cool. Way, way, cool!
Dec 04, 2009, 07:47 PM
Damp and Dizzy member
Brooks's Avatar
Norgale - The Goddard seems to use pipe for the uprights for the hogging chains; they don't seem strong enough to be really effective, so I wonder if the hull really needed support. It was an iron hull, and only 50 feet long. I have a photo of steel hull 50 footer, made 1886, and don't see any hogging chains (the photo is not that clear, however).

Neither boat has the wooden truss structure on the paddle support rails - this was to act counter to the chains (which prevented droop). Paddles slap the surface, so tend to bend the paddle support rails upward, which the truss was designed to oppose. Since there is no truss needed for these small iron/steel hulls, maybe there was no need for hogging chains either; in which case, the pipes and bracing were just for the canopy. But I agree, they would be nice to add to the model.

Kmot - after blowing my whistle about how capable the Midwest engine & pot boiler conversion are (as demonstrated by John's Steam Tramp), I thought I should make a boat myself :-).

One question about the Goddard - what did she use as a splash wall? Sternwheels throw up a lot of water, and the big wall is a characteristic of the breed. The underwater photo only shows pipe frames aft, where a normal steamer would have a wood board wall. The reconstruction drawing claims the canopy was wood. But, I am wondering if the wall, and the canopy, were made of canvas. Since the whole craft was disassembled and packed over the pass from Skagway, perhaps they chose to go this minimal equipment route. They'd not have to pack wood, of course (the owner even packed a sawmill over the pass). But a roll of canvas would be something sellable even if the skipper later decided the boat didn't need it.

The Rose photo is on page 378 in "Navigating the Missouri", by William Lass. "Packets to Paradise" is by John Lepley.
Last edited by Brooks; Dec 04, 2009 at 08:01 PM.
Dec 05, 2009, 06:19 AM
Registered User
-kno3-'s Avatar
I would add the trusses and chains,because they make the ship look more interesting, even if in model size they are not really needed for supporting the hull.
Did you consider providing for some means for adding weight at the bottom of the hull? You might need this to achieve stability, and unfortunately you won't know how stable the hull is until you have all spuerstructures and equipment on.
You could cut a recess in the middle of the deck, covered with a lid, and put some lead in there when needed.
Dec 09, 2009, 10:06 PM
Damp and Dizzy member
Brooks's Avatar


She ran fine on air. I have not tried her in the water, yet, will need to do that to fine tune the gearing. The long drive chain from idler sprocket to the paddlewheel ran ok. I am concerned that she might throw that chain, so will install a chain saver; a loop of brass rod encircling one leg of the chain, the other end attached to the hull. The machinery is moved forward for weight and balance, otherwise I could put the engine and bits farther aft and shorten the drive chain.

The paddle is heavy, I am pretty sure I'll need to use the 1.5" foam hull, and may even have to add a blister of foam to the bottom, aft. The towboat "Sprague" had a 40'x40' paddle that weighed 160 tons. Her hull was not flat along the bottom, but was 10 feet deep at the bow, 7 feet deep amidships, and 12 feet deep at the stern.
I'm not sure now if the Sprague's excess hull depth was applied to the bottom, or the top. The measurements reported may refer to the sheer line, which would make the hull flat along the bottom.
Last edited by Brooks; Jan 03, 2010 at 10:31 AM.
Dec 10, 2009, 02:29 PM
Damp and Dizzy member
Brooks's Avatar
A November 1988 "Scale Ship Modeler" magazine has an article on sternwheeler Bertrand. Beautiful RC model, electric powered. The modeler added a full length slab of floatation under his scale hull. Scale effects strike against us when it comes to wind and water :-). I'll likely need to do the same. The article is available for download over on Paddleducks:;down=98

Kno3 - I'm not sure what extra devices I'll need for stability. I might add a fin keel, with pb. The keel would increase maneuverability, a lack that real and model sternwheels alike suffer. I discovered the uses of a fin on a motor boat when I needed to sharpen the turn radius of a K&B outboard: I was running in a ditch, and could not turn around with just a flat hull and the outboard. But by adding a 2x4" fin keel (technically a leeboard, since I just screwed it to the side of the boat), the rudder/outboard had something to "pry against". I could then do donuts on the ditch w/o impacting the bankside. From sailing, I knew that a sailboat w/o any centerboard or daggerboard down would not turn, but would just skid. Same is true of flat-hulled motor boats, looks like.
Last edited by Brooks; Dec 10, 2009 at 02:39 PM.
Dec 10, 2009, 03:58 PM
3 Blades to the Wind
Shaun Hendricks's Avatar
You could always install a nylon chain guide/tensioner to minimize the chance of dropping the chain.

I agree the piping would be great for the canvas but I think the hull looks internally braced and supported. I don't see the framing or trusses either for hogging. This looks like a sturdily built little flat hull boat.

Added idea...

I also had a thought about your keel idea. It might be better to put two shallow skegs in front of your monkey rudders to handle the drift factor. If you neutral force the monkey rudders (put the steering shaft down the middle of the rudder instead of at the leading edge), you could 'blend' the skegs into the rudders when going straight so the rudders add to the stability of straight cruising. In a turn, the monkey's would bite at 'clear' water and really turn the boat.

It might not be scale but it would probably maneuver like the dickens.

Another thought...

Give any thought to using a pivot arm (instead of a chain) like the full scale ship used? That pivot arm and push/pull shaft would look really cool in motion on the scale version.
Last edited by Shaun Hendricks; Dec 10, 2009 at 04:44 PM. Reason: Had a thought...
Dec 10, 2009, 07:45 PM
Damp and Dizzy member
Brooks's Avatar
Hi Shaun - Using a pitman arm to actuate the paddlewheel, in place of the chains, would be cool, no question. Translating rotary motion from a motor (Midwest oscillator or an electric motor) into the linear-with-a circle-motion of the pitman arm is complex. It does not seem like it should be, afterall, it's just the opposite of what a loco does. There have been many schemes proposed by modelers, though. A scotch yoke works, but can wear out quickly, according to one builder, and has some geometry issues. I'll definitely try one for the next model. This model is a) experimental (keep is simple, go with what you know) b) built under time some simplification has been applied.

Chain tensioner, good idea. There is already, due to the long catenary, some tension on the chain. I am not sure what the failure mode will be (looseness, misallignment, waves rocking+inertia, etc.), I'll just have to run it enough to find out. The keeper was to reduce salvage expense :-)

Skegs fix the problem of a boat not going straight, think feathers on an arrow. I'm interested, based on the experience of other paddlers, in the opposite problem, ie. makeing her turn. Balanced rudders are supposed to help there, as you pointed out, and I plan to make the 3 rudders balanced (as were the real A.J.Goddard's). On the river, balanced rudders were subject to jamming by river trash: The skegs + regular unbalanced rudder design was to help avoid jamming. The model should not have jamming problems, unless we run thru seaweed :-).

The steering problem of sternwheel models has been approached via the "install more rudder area" philosophy. I think, based on my outboard experience (you won't get a more effective rudder than an outboard's spinning prop), that the solution lies elsewhere, namely to stop the skid of the yawed hull. Thus my musings on adding a keel or leeboard to the Goddard.
Last edited by Brooks; Dec 10, 2009 at 11:29 PM.
Dec 11, 2009, 04:30 AM
Registered User
-kno3-'s Avatar
Don't put the cart before the horses
Build the hull and see afterwards if you need a keel.
I would just allow for one or several covered deck recesses (where the hatches are in the plan) to add ballast if necessary (for balancing the hull).
Good luck!
Dec 14, 2009, 01:31 PM
Damp and Dizzy member
Brooks's Avatar

Ceramic burner

Following tips from John, who made a ceramic burner for his Steam Tramp, I made one too. Ceramic burners are very quiet and need little gas to operate, thus extending fuel supply (butane). I used a 6 oz pineapple juice can for the body, a Softbrick for the ceramic part, an Accuraft Ruby poker burner and jet (thanks for the loan, John), and Permatex Ultra Copper RTV sealant. Softbricks are used by pottery kilns, and kiln supply houses stock them. Don't get the hard kind of refractory brick, it's supposed to be very difficult to cut.

The can was cut 1 1/8" high; 1" interior height plus the 1/8" can bottom rim.

My softbrick burner disk is 1/2" thick. This material is as soft as chalk, very easy to work with, handtools suffice (I used an electric drill for the holes, it's like drilling butter). The only difficulty is in hacksawing off a 1/2" thick slice from the softbrick - don't twist the saw, or the slab may crack, it has very little tensile strength. Working the brick makes a lot of dust, so I worked with the brick in a cardboard box (Kinkos). After drawing a circle the size of the can, I laid out a 1/4" grid with pencil. Intersections were drilled with a 1/16" drill; I left a thick rim, not trusting the brick to standup to holes near the edge, so only got 5x5 holes plus 4 extra, totaling 29. A triangular file was used to make grooves along the lines of the holes. This makes little peaks of ceramic, which will glow red hot, yielding radiant heat. To finish the ceramic, I rotated the can in the brick; if your can has smooth sides, you can simply use the can as a holesaw, which is what John did. My can had ridges, so I could only start the cut. I finished rounding the ceramic with a coping saw and sandpaper. The resulting plug was a press fit to the can.

I covered over some of the poker's slits with 11/32" K&S brass tubing (this was because the poker is longer than the can's diameter). I sealed the tubing at the jet end to the poker with the Ultra Copper, and sealed the tubing to the can with the same. I put Ultra Copper around the circumferance of the brick, and pressed it into the can. In testing, I see that my seal was not complete - this does not seem to pose a problem, other than the flame burns the sealant; I don't know if all the sealant between brick and can will combust, eventually, or just the part that burned today. John, due to his using the can itself as a saw, got a nice snug fit, and did not seal the circumferance of his brick at all. David Wegmuller, who wrote of his ceramic burner, used regular RTV. The can itself is gascooled, so does not get red hot as a rule.

Firing - Ceramic burners don't like, or need, a lot of flame space between themselves and the boiler. In this they are opposite to most firing systems that I'm familiar with. Sandy Campbell, ACS Engineering, Scotland, gave me a drawing of his model boiler/burner system - he only allots 3/8" between top of the ceramic and bottom of the vertical boiler. I found, while testing my burner, that I got more consistent lighting, and a fuller burn pattern, when I lowered the test boiler (tin can with some water) to about 3/8-1/4" above the burner. Radiant area increases as boiler is lowered over the ceramic. If you go too close, the flame will sputter, if you raise too high, the flame will abandon the outer holes, and radiance will decrease.

A source for soft brick, if you can't find them locally:
Last edited by Brooks; Dec 14, 2009 at 02:00 PM.
Dec 14, 2009, 04:31 PM
3 Blades to the Wind
Shaun Hendricks's Avatar
Well, my idea of the balanced behind the skegs was to give the rudders clear water to bite into. It's my understanding that rudders behind keels (or skegs) that are hinged to the keel or skeg are less effective at steering and they induce additional 'roll' to the ship, which is sometimes wanted and sometimes not. You're totally correct about the jamming issue though. The skegs can help and hurt there.

I love your burner. It seems to me that design could be combined WITH the boiler, making a single unit, perhaps even adjustable with a servo so you can remotely increase or decrease the steam output. Just another thought.
Dec 14, 2009, 05:20 PM
meatbomber's Avatar
great project brooks ! i`m just wondering what you had in mind for speed control if at all or just single speed like the Tramp (which is beyond cool i might say )?

On the control issue, have you thought about not adjusting the heat input but the steam output to the engine ? The amount of steam not used would be vented into the exhaust stack.
Dec 15, 2009, 12:00 AM
Damp and Dizzy member
Brooks's Avatar
Shaun - thanks for the skeg information. I had never considered the flow-straightening aspect of a skeg. Jet engines use vanes to straighten flow between stages, so what you said makes perfect sense.

MB - I made a simple steam throttle for a BAGRS. I might do the same for this boat, have to see how she performs on the pond. Generally, you try to conserve steam, so if you were generating too much, you'd cut back the fire, not vent the excess pressure (the safety valve will do that for you, but that's for safety, not speed control).

Venting steam up the smoke stack, if done properly, will increase draft to the fire, helping you to make steam. Steam locomotives did this all the time; in fact, it's so important, that much research was made to determine the best way to use the steam to get more and more draft. There are a plethora of designs from the 1800's and early 1900's of smokebox appliances, some with science, some with just practical men trying something they thought would work. I used a 1900's, 3-stage blower design on one BAGRS to get more draft :-).
She ran on steam this evening. The bench test went pretty well. I am still learning operation of a ceramic burner - always something new :-). I could maintain pressure for the bench test, paddle whirling, for as long as I had water and gas in the tank. Longest run was 9 minutes - I only stopped to check water level. Other runs were 3 & 6 minutes, again stopping to record water usuage rates. Based on the bench, I'll have enough water for a 12 minute run. Water will slow the paddle, so the water run will likely be longer - we saw that with John's Steam Tramp. She would respond to changes in the gas flow. I was too busy to get a measure on RPM, but judging by sound, cutting or increasing the fire via the gas control knob had an effect.

So I am pleased that this design worked :-). I am not sure how much actual power there is, though. I could stop the paddle fairly easily with my fingers rubbing on the blades. Hopefully there will be enough power to spin the wheel in the water. Dog pond testing over at John's garage Tuesday I hope :-).
Last edited by Brooks; Dec 15, 2009 at 12:05 AM.
Dec 15, 2009, 06:52 PM
Damp and Dizzy member
Brooks's Avatar

On Fergus Pond

John and I ran a sucessful test of the Goddard, using Fergus' pond (Fergus is the Hartleys' Chesapeake Retriever). After one field modification, A.J.Goddard ran great. We had great steam pressure, and strong paddling the whole cruise. Boy, I am sold on ceramic burners :-).

The modification was to increase the gear ratio from 1:4.6 to 1:7.35 (7 revs motor to 1 rev paddle). The initial ratio of 1:4.6 proved too high for the motor - the motor and paddle spun like crazy on the bench, but the motor died as soon as the paddle hit the water. We were operating with the safety blowing, so were getting as much out of the boiler/motor system as possible.

We switched the gear on the idler rod from a 15 tooth to a 24 tooth. That did the trick, the boat worked fine; the paddle never stalled. Gearing now is: motor=7 t, 24t -idler shaft- 7t, paddle=15t. Interestingly, this new gearing duplicates that found in the BAGRS loco - Mike Martin (BAGRS designer) had commented elsewhere that most BAGRS locos seemed to gravitate back to that gearing, showing that it must have some optimum performance for the motor.

Run was 9 1/2 minutes, (about 80ml initial water charge, with 10ml left in the boiler). I expect we will work up to a longer run with practice. We were running the burner at max; John doubled the initial runtime of his Winifred tramp by discovering that the tramp's burner could be turned way down.

The main noise was from motor whirring along and paddle splash. I have not yet discerned chain noise. The 1 1/2" foam hull worked fine, draft was acceptable and stability was also. Paddle depth was good. The bow rode a little high, but not too bad. I'll have to add a strongback underneath the deck - it's developed a sheer line - pretty in itself, but plays hob with the bulwarks I wante to attach. An Al L-shape, bolted to the bottom of the 1/4" plywood, should level it out.

The boat wanted to scoot. Corrugations in the sides of the plastic pool kept her from racing around. The tripple rudder seemed to work, though there was not really enough room in the pool to get a good look.

A movie will be posted.

Thread Tools

Similar Threads
Category Thread Thread Starter Forum Replies Last Post
Off Site Live Steam Pop Pop Boats. a lot of 19 boats! 1987tc Boats (FS/W) 0 Nov 08, 2009 05:16 PM
Idea Live Steam tdonily Dock Talk 17 Feb 03, 2005 03:30 PM