Steamboat Arabia 1856 by Cathead - FINISHED - Scale 1:64 - sidewheel riverboat from the Missouri River, USA

[Cathead]

Steve, you're closer than you know. When I was out on the porch yesterday taking photos for that update, I had my Kindle as well, loaded with the reference photos I was using to get those paired images from the same angle. I set it down on the railing above the model and almost lost my grip; it would have dropped a couple feet straight down onto the port wheel and machinery, bouncing into who knows what else. I really don't know what I would have done in that circumstance.

[druxey]

But you have to admit, a dual set of models would be novel and intriguing. Besides, in the 'finished' version you could omit a ton of internal detail that shows so nicely in the 'wreck' version. Just sayin'.

[Roger Pellett]

Beautiful work, Eric!

 

Roger

[cog]

On 5/5/2019 at 5:12 PM, Cathead said:

Steve, you're closer than you know. When I was out on the porch yesterday taking photos for that update, I had my Kindle as well, loaded with the reference photos I was using to get those paired images from the same angle. I set it down on the railing above the model and almost lost my grip; it would have dropped a couple feet straight down onto the port wheel and machinery, bouncing into who knows what else. I really don't know what I would have done in that circumstance.

I tend to agree with the missus Eric, and Steve has a point, but I wouldn't do it for that reason. So if you start a second one you only have to wait for that fatallity, and still be able to work on a "backup" ...

 

On 5/5/2019 at 6:32 PM, druxey said:

But you have to admit, a dual set of models would be novel and intriguing. Besides, in the 'finished' version you could omit a ton of internal detail that shows so nicely in the 'wreck' version. Just sayin'.

Very intriguing indeed, and you really have some very persuavive arguments, Druxey. Nearly twisted my arm, don't know about Eric's ...

 

However, if you do decide on a duo build, you could choose another way to lay your deck planks at the stern ... just a thought ... 

 

 

I think I'd better go look at some birds in the garden now ...

[mbp521]

On 5/4/2019 at 6:38 PM, Cathead said:

Also, assuming that anyone reading this is interested in American riverboats, please go check out this new build log of the Chaperon, the only accurate kit available in this genre. Brian's doing a great job with her (some poor advice from me notwithstanding) and it's well worth a look.

Cathead. Thanks for the kudos, it is very much appreciated. While my talents aren’t quite up to your level, I think she’s coming along nicely. 

 

While can see Mrs. Cathead’s point, unfinished models do have a certain intrigue to them, (like your Bertrand) the beauty of the finished product is also worth the work.

 

Besides, you always have the build log to go back and look at to see all of the internal details. 

 

- Brian

[Cathead]

You people are evil, encouraging a second model. It ain't happening until this one is done at the very least.

 

Quote

Besides, in the 'finished' version you could omit a ton of internal detail that shows so nicely in the 'wreck' version

That's not quite right; all the machinery I've built so far can be seen on a fully finished model because the main deck was mostly open to the side. It'll be somewhat obscured, but if I left out any of the detail of what I've built it would look wrong when you gazed on the model from the front or side. Even in the shadows,  you need to be able to see the engines, driving arms, etc. The only thing I could skip in a fully finished model is a full wheel, and I already did that on the port side. So the only difference in a "wreck" model would be that you'd see the machinery more clearly without a deck above it. I wish it were otherwise!

 

Moving on, here's the design question I mentioned last time that I need some input on; it relates to the internal bracing. Riverboats used extensive networks of "chains" (actually iron rods with turnbuckles) to keep the hulls from warping. On sternwheelers, the main concern was "hogging", or longitudinal warping, in which the hull tended to flex down at the bow and stern and rise in the middle, especially due to the weight of the engines and wheels right at the stern. Thus the main set of bracing on sternwheelers were hog chains, running longitudinally along the hull and rising above the decks.

 

However, on sidewheelers the biggest concern was sideways warping due to the weight of the wheels hanging off the sides and the wide decks, especially the guards (the deck extensions beyond the main hull). So sidewheelers had a lot of lateral bracing, depicted by Alan Bates like this (I hope it's ok to post this image, I don't know how to explain this without using a Bates drawing).

So you see two kinds of visible bracing: the knuckle chains that support the edges of the hull, and the cross chains that support the guards. Pretty straightforward. My question is, how were these spaced along the length of the hull? Although  I can find versions of this cross section in various sources, I can't find anyone explaining how many of these were used along the hull's length. Every 10'? Every 20'? How did they interact with major features like the boilers and engines that block their path as shown above? I can't find a diagram or text that deals with this question.

 

Here are some mockups on the Arabia to show what I mean (the boiler deck is just set in place for reference, not attached). Here I've placed scrap wood at the approximate angle of both knuckle and cross chains, at an approximate 10' spacing along the forward hull. But I have no idea if this is right. Note how close the knuckle chains come to the boilers and how the engines essentially block their use.

 

Kurt, Roger, anyone else knowledgeable, I'd love some perspective on the right way to arrange these.

 

[kurtvd19]

Eric:

I have talked extensively with Jack Custer (owner and Editor of the now defunct Egregious Steamboat Journal and expert on riverboats) about hogging chains.  Jack said that while fore to aft hogging chains have been well documented - because they are very visible) there was actually not all that much written about the transverse hogging chains.  He spent quite a bit of time encouraging me to model a cross section to show how transverse hogging chains were used.  This is something I can't do w/o a lot more information from guys like Jack.

 

Looking at the drawings and writing on pages 23 - 30 of the Engineroom Cyclopedium and Figure 3 on page 7 and remembering that ARABIA was not a cotton transporting boat - I don't think the transverse hogging chains on ARABIA would have been located anyplace other than close to the sidewheels.  The photo of the SF Museum's steamboat EUREKA being rebuilt in 1924 shows the hogging chains in front of and behind the sidewheels.  It's not a riverboat as such but it does show the transverse hogging chains only by the wheels.

The typical "cotton" boat shows the transverse hogging chains clearly in this photo of the large sidewheel boat - note the cotton waiting to be loaded.

The next photo of the BEN CAMPBELL shows a boat of similar configuration to the ARABIA and there are no visible transverse hogging chains in the forward section of the boat.  In the last photo note there are visible transverse hog chains on the CHALAMETTE and the boat in the background - but both are "cotton" boats.

I hope this helps a bit

Kurt

[Cathead]

Kurt,

 

You're certainly right that Arabia wasn't a cotton boat but one built for the upper rivers; if nothing else this is shown by how narrow her wheels and guards are. So your thought is that I shouldn't bother with any transverse chains except maybe in front and back of the wheels? And those would be cross chains (going out to the guards) rather than knuckle chains, since the cylinder timbers block the latter? 

 

I'd looked at that Ben Campbell photo and those I have of the Mary McDonald because these (as you say) were pretty similar to Arabia, and hadn't seen anything, but thought that was just because of photo resolution/detail. It'll certainly make the build easier to not have to add these!

 

What about longitudinal hog chains? Bates says sidewheelers usually had a singe one running down the centerline, just below the boiler deck, making it nearly invisible. Is it worth trying to add that detail? I assume it'd have to anchor just behind the main staircase, run up between the chimneys, then hang from the center ridgepole the rest of the way aft. It'd be really hard to see but if it was likely there, seems to me I should try to simulate it. What do you think?

 

Also, am I missing any other main deck detail I should include before attaching the boiler deck and intervening superstructure? I'll add hatches and winches to the bow later, and that's open anyway.

Edited May 11, 2019 by Cathead

[kurtvd19]

Eric:

I wouldn't show transverse hog chains except by the wheels.  The writings after Figs 43 and 44 say that there were two sets, just ahead and just behind the wheels.

The only evidence I saw of fore to aft hogging chains on a sidewheeler are on the J.M. WHITE.  This was a Mississippi River boat.  And was quite elegant inside and out.  I can't find it but I have a photo of the interior of the J.M.  WHITE showing the transverse hog chains in the dining room and how the passengers and crew had to be aware of them so they didn't hit their heads on them.

As to fore to aft hogging chains I think there would have been two sets as shown in fig 7 going to the lower hull as shown in fig 30. 

Kurt

[Cathead]

Kurt, here's what Adam Kane (not Bates, faulty memory on my part) says on the subject:

 

Quote

Side-wheelers typically had one hog chain running along the center-line of the vessel...the after end...was mounted into the keelson just forward of the sternpost, while the forward end was fastened one-quarter of the vessel's length aft of the stem...these rose to a level just below the boiler deck

 

That's what I was basing my plan on; it puts the forward end of the chain right at the main stairwell. What do you think? This makes sense to me as it places the line of support directly along the keelson (though I'm sure it was done various ways as with most things on these vessels).

 

Another thought on the transverse chains is that I'm building Arabia, like the Ben Campbell, to have a boiler deck that extends out to cover the main deck over most of the length (unlike the big cotton packets), which also provides more bracing and support overall for the guards even while it adds a bit more weight as well. Which again implies, as you say, that I just need a few chains around the wheels.

Edited May 11, 2019 by Cathead

[cog]

When I read your initial post on the hog chains, I thought it would be as Kurt answered: Before and aft of the paddle wheels. However I thought it strange not to use the transversal hogchains, as the hull would warp when being loaded. Reading Kurts posts, I now do wonder why only one longitudinal hogchain would be used, as that would probably cause transverse warping again. Crossed hogchains would in fact be logical for a boat carrying heavy cargo, and if the beam extends a certain width, I would expect mulitple longitudinal hog chains as well. (Just a thought to make things slightly more complicated)

[Roger Pellett]

Eric,

 

I hope that you find the attached helpful.  This is from a digitally remastered daguerreotype photo taken in 1848 of the Cincinnati, Ohio waterfront.  The remastering was done by the Cincinnati Public Public Library.  The steamer in the photo would appear to be a typical Ohio River boat intended fou use in shallow water.  The vertical white posts are for the transverse chains supporting the paddle boxes.  The vessel does not appear to be fitted with longitudinal chains.

 

Roger

 

Edited May 12, 2019 by Roger Pellett

[Cathead]

Roger,

 

That's a great and rare early image. The resources I've read (like Bates and Kane) all note that early sidewheelers like this one had tall support posts as shown here, but that by Arabia's time these were mostly kept short (i.e., under the boiler deck) and so wouldn't have been visible in any exterior photo. Same goes for longitudinal chains, one of the reasons it's so hard to use later photographs as evidence for anything in this regard.  I think that photo nicely documents the fact that not all boats inherently had a full set of transverse chains, as in an early craft like that they would be visible as higher posts. That vessel seems to have narrow wheels and guards, like Arabia, thus needing less lateral support. I'm a little surprised it doesn't show longitudinal chains given how long that hull is, but it's also early enough to predate the full adoption of hog chains and still falls in the highly experimental period of steamboat design.

 

Right now my leaning is to put in two sets of transverse chains fore and aft of the wheels, maybe 10' apart. I still feel like some kind of transverse chains would be necessary to help strengthen the hull further forward for similar reasons as Carl's post, even if it's only one or two sets, so I'm playing with the idea of putting in a couple where they can been seen as visual interest.

 

I'm going back and forth on the longitudinal chains; I like Kurt's suggestion of two (one on each side of the hull) despite Kane's claim that this was mostly done as one down the middle, because it actually makes more sense with the geometry of the hull and superstructure.

 

Overall, as Arabia was built for the Ohio River but later transferred to the Missouri, it wouldn't surprise me if she was initially built with minimal hog chains but was retrofitted with more for the latter river. Navigating the upper Missouri required extensive encounters with sandbars and extremely shallow water, meaning that the hull of such vessels was subjected to extreme flexing as they literally slithered over the bottom. Thus I think it's reasonable to conclude that she may have had more chains than a lower-river boat to help hold the hull together under such conditions.

 

Thanks to everyone for the input, it's been extremely helpful in thinking this through.

 

 

[cog]

We are nicely talking about hogchains, but did you consider diagonal bracing? Adam Kane The Western River Steamboat pg 110 ...

[Cathead]

Carl, all that sort of thing refers to interior framing within the hull, beneath the main deck. None of it would be visible on this model's closed hull.

[cog]

I know, Eric,  but it would explain e.g. missing longitudinal hogchains as rigidity would be established through the braces ...

Edited May 13, 2019 by cog

[Cathead]

Carl, from my perspective (and I'd be interested in Kurt's and Roger's input), the internal bracing and the hog chains serve slightly different purposes. Internal bracing makes the hull more rigid, which is particularly useful in ocean-going vessels but potentially problematic in riverboats because too much rigidity means you can actually break the boat's back in shallow-water river conditions (this would be less of a concern for deepwater steamboats on the lower Mississippi like the big cotton packets). The chain system allowed for a certain amount of hull flexibility that was needed in these conditions while still providing support; it also meant you could readjust the tension needed as the boat worked and aged. So most boats, especially by the 1850s, used a combination of both as relying on one or the other exclusively would be problematic.

 

Also, extensive wooden bracing gets heavy and expensive rather quickly, and upper-river boats needed to be light and cheap. A few iron iron rods were a lot more practical in this sense than lots of custom-cut oaken braces.

 

Finally, upper-river boats tended to carry proportionally more cargo in their holds than the bigger lower-river boats with their huge stacks of cotton on deck. Among other things, this kept their center of gravity lower, an important consideration in the very windy conditions of the middle and upper Missouri River. Thus, transverse bracing especially quickly eats into both the available space and the navigability of the already shallow and narrow hold, whereas chains are smaller and easier to work around.

 

So while there's no doubt something like Arabia had internal bracing in her hull, my instinct is that it wasn't sufficient or even desirable to have a full web of bracing down there, making chains a realistic alternative.

[mtaylor]

Eric,

Wouldn't need for hogchains be more important for down river with the heavy cotton bail loads then up north?   I'd think on Missouri, that a bit of hull flex might have been good thing just because as you state, they had the sandbar problem.  The Ohio has it's shallows upriver as I recall so what worked there might be good for the Missouri.  

[cog]

Thank you, Eric. Another lesson to be stored

[Cathead]

Mark, as I said above, upper-river boats needed hull flexibility and lightness to deal with shallow-water conditions, which is why I would expect them to rely on chains as much as or more than hull bracing. However, that doesn't mean they needed lots of them, given their relatively light hulls and cargoes. Lower-river boats could afford to have more interior hull bracing (and thus more rigid hulls) because of calmer and deeper river conditions. They still needed chains to help support those huge hulls and their extreme length-width ratio, but especially required lots of transverse chains to support their extra-wide and heavy guards (see Kurt's images of the J.M. White, above).

 

My conclusion from all this is that something like Arabia wouldn't have had as extensive a network of chains as you see on the J.M. White, but likely still needed a few to support the needed hull flexibility of an upper-river boat. I.e., Arabia shouldn't look like a spiderweb, but should probably have a few transverse chains near the wheels and possibly one or two further forward, then one or two longitudinal chains to help support the hull's flexing over sandbars and such.

 

I had no idea this would generate so much discussion, maybe I should have put this question in the general riverboat thread. Oh well.

Edited May 14, 2019 by Cathead

[kurtvd19]

Eric:

I doubt there is anybody there that isn't following the Arabia build.

I think you are 100% on track.

Kurt

[Roger Pellett]

The longitudinal  load distribution on the hull of a ship is determined by subtracting the upward forces of buoyancy from the downward forces of weight at each point along its longitudinal axis.  For sternwheel vessels, the paddle wheel represented a heavy weight at a point where there was little or no displacement and therefore buoyant force to support it.  This would produce a large hogging moment required to be offset by the chains.  

 

You are modeling a sidewheel vessel.  The heavy side wheels and engines are located at a point along the hull where the shape of the hull provides a full upward buoyancy force to offset their weight which is probably why the steamer in the photo that I sent you does not show longitudinal chains.

 

Transverse chains are a different matter, as they would have been necessary to support the paddle wheel’s outboard bearing.

 

Chains were only effective in limiting hogging forces as compression forces caused by sagging would cause them to buckle.  Sagging forces in the calm water environment of the River could be limited by hull shape.

 

Roger

 

 

[Cathead]

Roger,

 

Both Louis Hunter and Adam Kane note that longitudinal hog chains were first developed in the 1840s and probably weren't in widespread use until the 1850s. So their apparent absence in your 1848 photo isn't necessarily indicative of their use or not. Both of those resources, as well as Bates, make blanket statements that sidewheelers used longitudinal hog chains, but that once the engineering had been worked out (i.e., probably by the mid-1850s), they tended to remain below the boiler deck and thus wouldn't be easily visible in photos. But they'd be visible on a model viewed from the side. So I think I'm well within reasonable modeller's license to install them on Arabia.

 

Bates also notes, in a passage I missed until rereading carefully, that some boats' knuckle chains (those supporting the hull) remained below the main deck, which is a good enough reason for excluding them above-decks on Arabia.

 

In several re-readings, I can't find a clear answer on the use of cross chains (supporting the guards) for narrower non-cotton-packet boats like Arabia. It's obvious they were abundant on wide boats, but no one really discusses their uses on other boats. A clear argument could be made that Arabia didn't need them other than near the wheel, but also that a cautious builder or owner might have had a few installed further forward for greater strength. As for the hog chains, I think reasonable modeller's license allows the judicious inclusion of cross chains in the model for visual and historical interest.

[cog]

And then there are the Sampson post, and the beam and truss arrangement ... Bates p 28 - 29 ... it seems you have some more delving to do

[Cathead]

Sampson posts went out of use before this period, they were an early style only. There won't be any trussing above deck other than the regular superstructure.

[cog]

Any other options?

[Roger Pellett]

Eric,

 

The Car of Commerce photo shows four vertical white stanchions protruding thru the deck in way of the paddle boxes.  If you wished I believe that it would be accurate to show the same arrangement on your model with the transverse chains supporting the paddleboxes running above the deck.

 

Roger

[Cathead]

Roger,

 

I think the reasoning I posted when you first shared that image still stands:

 

Quote

The resources I've read (like Bates and Kane) all note that early sidewheelers like this one had tall support posts as shown here, but that by Arabia's time these were mostly kept short (i.e., under the boiler deck) and so wouldn't have been visible in any exterior photo...That vessel seems to have narrow wheels and guards, like Arabia, thus needing less lateral support.

I don't think it'd be wrong to do as you say, either, as Arabia does fall into that grey zone of development (early in the period when designs really stabilized for the next few decades). But overall I seem to be modelling her as a relatively modern vessel incorporating most of the new design ideas of the 1850s onward rather than as a throwback to rapidly evolving 1840s designs, so in that sense I think I'm comfortable keeping the transverse chains below the boiler deck.

 

I really appreciate your input, though, it's helpful to be challenged to think through and justify the choices I'm making (if only to myself).

[Cathead]

There will be an update on the Arabia soon, as I've been making progress on the chains, but first some Missouri River news.

 

The river is flooding heavily now, following a spring of heavy rains throughout much of its huge drainage basin (it collects water from about 1/6 of the continental US). For example, May in Kansas City has already been the 8th-wettest month in the 131 years of records (for any month of the year), and there's nearly a week left to go with lots more rain in the forecast.

 

One of the basic problems in this basin is that the river's channel has been heavily modified. In Missouri, for example, the river's floodplain is around a mile wide, and the natural river had many channels and easy access to its floodplain, so that when spring floods happened the water easily took multiple routes downstream and could spread out into the floodplain. It takes a LOT of water to cover a mile-wide floodplain with even a few inches, and when that effect is extended for hundreds of miles along natural floodplain lined with trees and vegetation, even major floods can be partially absorbed by the landscape.

 

However, the river was heavily modified for navigation and development. Now, throughout Missouri, it follows a single narrow & deep dredged channel with high levees on both sides that trap all the river's water in that single channel. It doesn't have access to its floodplain anymore, so when floods do happen, the actual water level in the channel rises a lot faster and higher than it would otherwise, creating an arms race between ever-higher levees and ever-worse floods. Worse, the destruction of prairies and deep prairie soil for modern agriculture leads to far more runoff from the landscape, as the former held lots of water on the landscape while the latter sheds most of it straight into drainages.

 

The few exceptions to this are a string of state conservation areas and federal wildlife refuges along the river that are managed for natural habitat and to absorb some floodwaters in a more natural way. With the river near its highest peak since the record floods of 1993 this weekend, Mrs. Cathead and I hiked out to a blufftop overlook above one of these areas (Eagle Bluffs Conservation Area) to get a look:

 

Above, you're looking across about a mile of floodwater that I would guess ranges from a few feet to at least ten feet deep, based on my extensive knowledge of the landscape here (I go birding here all the time, this area is one of the top birding destinations in the state). You can't see the actual river; the current channel is at the far side of the valley, just below the higher bluffs on the horizon (the thick, horizontal line of trees below those bluffs are the riparian forest on the near riverbank). Eagle Bluffs covers around 4400 acres (~17 square km) and I'd guess that at least 2/3 of it are underwater. Below is a view looking downriver; the river still can't be seen (here it's cutting across the valley behind the thick line of trees at center) but there are some typical limestone bluffs shining in the distance in the late day sun.

The reason I post these, in addition to just loving the Missouri River and both its natural and human history, is that this is a rare chance to envision what this valley might have looked like during the steamboat era (before modern development destroyed its natural state). A flood this big would have shut down steamboat navigation, but you do get a sense of what the landscape would have looked like in 1856, and there would have been various channels and lakes in the otherwise tree-covered floodplain even during normal navigation conditions.

 

As a final note, the bluff we took these shots from was topped by a large Native American burial mound, likely from the Woodland Period (about 1000 BCE to European contact). These mounds are ubiquitous at high points along the river and its major tributaries in this part of Missouri, likely forming a line-of-sight network in which multiple mound sites could be seen from any one location along the river. The biggest earthworks (and biggest known Native American city) in North America lie just downstream of the confluence between the Missouri and Mississippi rivers near St. Louis (about 120 miles from here). Just a reminder that this river supported civilization long before Europeans got here; diseases brought by early explorers (including early steamboats) played a major role in destroying these cultures before they had a chance to adapt or resist.

 

Thanks for reading (or at least skimming) my lecture. In another day or two I'll have proper photos of the actual model progress.

Edited May 26, 2019 by Cathead

[Roger Pellett]

Thanks for posting this Eric.  “Keep em’ coming!”  Mother Nature always has the last word.

 

Roger