Jump to content

mbp521

NRG Member
  • Posts

    936
  • Joined

  • Last visited

Everything posted by mbp521

  1. Glad to see that I was able to spark an interest in this magnificent boat. I will be following along intently! Best of luck. -Brian
  2. Alrighty folks, the brace is finally off and the new build has begun. Come on over and take a look. -Brian
  3. Greetings everyone, I am finally ready to get going on my next build. This time I am going a bit more modern and taking a step into the 20th Century. A friend of mine asked if I minded building him a model of the Caroline N, a Mississippi River Towboat that his father-in-law named after his daughter. This sounded like nice challenge, and an excellent chance to hone my scratch building skills, not to mention I have never built any boat newer than the late 19th Century, so I accepted. The Caroline N was built by the Jeffboat Company out of Jefferson, IN. She is 134' long with a weight of 946gt. She was originally laid down in 1973 as the Louis H. Meece for the American Commercial Barge Line. She was later purchased by the Triple M Transportation Company and renamed the Caroline N in 2004. She was sold to SCF Marine in 2012 and in 2018 she was leased by Marquette Transportation and renamed Titletown USA. As of the start of this build, she is still being operated by Marquette Transportation. She is powered by two GM Diesel engines, a 4.192:1 reduction gear box, turning two fixed pitch screws in kort nozzles and rated at 6,140hp. I will be building this model at 1:64 scale which will put the model at just over 25" long. The hull will be POB and most of the construction will be basswood & MDF. I'm sure there will be other materials thrown in for good measure as I go along. Her paint scheme will be the blue and white from her Triple M Transportation days, as pictured below. I am currently working on getting the plans scaled to size and will be using reference photos from the actual Caroline N, as well as her sister boats the Charles Lehman (now Limestone Lady) and Richard C. Young (now Miss. Kris) for additional details. The plans that I will be using are the original Jeffboat General Arrangement plans of her configuration in 1981 for sizing. These match up best with the photos as far as placement of critical features such as windows, doors, stacks and other details. With all that being said, it's about time to get this build started. I'll hopefully be back shortly with the scaled plans and ready to start making some sawdust. So pull up a chair and stay tuned. -Brian
  4. I am still around, just a couple of more days and I should get this contraption off my arm and start physical therapy (torture). With the brace gone I should be able to at least get things ramped up on my next build, At the very least I can get the build log started. Stay tuned. 😁 -Brian
  5. I’m not really surprised that these would have gotten lost, at the small scale I would’ve had a hard time keeping up with those as well. Nice job on the scratch built ones though. Yours are probably better than the kit provided ones anyway. These were the races that the tiller rollers rolled along. I can just imagine the fits you would have if the rollers were included with this kit. They would probably be the size of a pin point. 😁 One of the issues that I had in research was trying to find if the hull was red or black. There is no information available that I was able to dig up that showed either way. When I was discussing this with Ray Hamel at Vicksburg NMP he said that tests had been run on pieces of the wood fabric but the results of paint color were inconclusive. So I just went with personal preference to break up the coloring a bit. To be honest though, the all black scheme looks pretty good too. -Brian
  6. Just squeaked that one in there! Beautiful mount, I love the idea of using reclaimed drawer pulls. One thing of concern that comes to mind, would the sulphur in the rubber o-rings cause a reaction with the poly coating of the hull? Just curious. I’d hate for something to damage the beautiful work you’ve done. -Brian
  7. Eric, I admire your tenacity with sticking with this build. I for one would have shelved (or trashed) it a long time ago given your frustrations with it. I have a couple of builds that have been shelved for years, due to frustrations with the instructions and the kit itself. I, like you, didn’t just pitch them due to the fact that I have a few hundred dollars tied up in the kits and felt it was a waste of my hard earned money to just toss them. I figure one day that I will dust them off, and give them another go, but for now I have plenty of other projects in the works to keep me busy for a while. Stay strong my friend, you will persevere! -Brian
  8. I wrestled with the black color of the boat and went through several versions before I settled on the dark gray that I used on my Cairo. It was just seeing the boat in the different lighting and dependent on the time of day that would frazzle my brain. -Brian
  9. Greg you are cranking out these boats faster than I can get through your build logs. Almost hard to keep up. Beautiful work on fleet! The grunginess and weathering is all too realistic looking. Just amazing! -Brian
  10. Speaking of following instructions, this conversation reminded me of another area of where the instructions are wrong and I didn't realize it until it was too late. When you get to Stage 4, the instructions show to mount the bull rail braces, part 48, to the stationaries with the opening outboard and placing the bull rails from the outboard side.. To be historically accurate, they were actually mounted the other way around to where the bull rails could be installed from the inboard side. I found this out later on when I was studying some of the contemporary photos of the boat. In the below photo, the Chaperon is on the left. If you zoom in to the starboard side where the bull rails are mounted you can just make out the framework of the braces that show them mounted this way (I thought I had a better picture of this, but I can't seem to find it right now). It would also make sense that they be installed from the inboard side, due to the fact that if a person or livestock were to lean against, or run into them they could knock the rails out of place and fall overboard. -Brian
  11. Beautiful work Tom, and congratulations! So sad to see the build come to an end but can't wait to see what's next. Thank you for putting this log together, I have it bookmarked for when I finally get around to building my kit. -Brian
  12. It's that fine line that we dance on with the Admiral between work and play. A couple of years ago I bought the Admiral a Cricut for Christmas. She had said she wanted one and I thought, why not, it would keep her busy enough that she would leave me alone with my shipbuilding. Well so far, I think I have used the Cricut just as much as she has. And the honey-do list continues to grow. -Brian
  13. John, you absolutely got it right. I ran into the same issue with my build and placed them on the 2nd & 3rd posts for that same reason. Also good thinking on reinforcing the boiler deck, a decision you will not regret. I went as far as using scrap pieces inside of the walls as well to keep them even at the tops and bottoms since the walls were so long and flexible. It added a lot of stability to the structure. She is really coming along nicely. -Brian
  14. Thank you all for the well wishes. Even though I am not able to work on an actual wooden model ship build, I am able to work on wooden puzzle of a ship. 😁 While entertaining, still nothing like making some sawdust. Hopefully only a couple of more weeks in this contraption and I can get back to it. -Brian
  15. So glad to see you back at the bench again Keith! I can’t wait to delve into the upcoming updates. Kind of the opposite with the sunny days, I tend to find myself wanting to work on a build more on the cloudy days and wanting to be outside on the sunny ones. Of course wintertime sunny days can be a bit deceiving sometimes here in Texas, I can only imagine a sunny day in Michigan in the winter. I’m sure that would change my thinking in a hurry. -Brian
  16. I have a whole box of unused drawer pulls sitting in my barn collecting dust. Now I have a prospective use for them. What a unique idea Keith! -Brian
  17. Man that structure is tiny! I don’t think there is any way that I could do a full build at that scale. Even the tiny bits on my boat gave me fits. Definitely like the added details. -Brian
  18. Looking good Tom. I’ve never been a fan of the Britannia parts MS uses in their kits, and I often replace them with scratch built pieces when I can. Needless to say though, you did a great job on straightening and cleaning up the kit pieces and the bird cage looks wonderful. It will be a bittersweet day when “Finished” is added to the title of this build log. -Brian
  19. Playing catchup on the build log entries. There seems to be a lot of information posted lately, Roger, to answer your question about whether the Cairo use rope or chain to attach the anchors, I would say that it was chain. The below pictures were screenshots that I took from a conference Edwin Bearss gave just before he passed away. In his presentation he had these photographs taken during the recovery efforts up on the screen. I searched in vain to find the actual photos and finally settled on just taking a picture of my computer monitor (hence their graininess). One picture shows the starboard chain going through the hawse pipe and the other shows the chain draped across the foredeck. The port hawse pipe empty so it would somewhat be hard to tell if there was a chain or rope attached to the portside anchor. There is documentation from journal entries of sailors aboard the Cairo and the other boats in the flotilla, at the time she struck the torpedo that tell when the explosion occurred the port anchor was launched from the deck. If I remember correctly, one of the flukes from this anchor is on display in the Cairo museum. So the possibility of the port chain being broken and with the force of the explosion could have cause it to recede through the hawse pipe and would not have been visible during salvage. Or this hawse pipe could have simply been strictly used for rope to tie the boats off to the shore. During my build, I struggled with the same question, how did they haul in the anchor with a capstan that was designed for rope. I came up with one possibility that they would use a length of rope attached to a hook or a loop. They would secure the hook through the chain loop and run the rope through the hawse pipe and around the capstan and haul the chain up a little at a time. This by no means would have been an efficient method of doing it, but as with John Howard and the rest of the St. Louis team have stated, documentation is scarce. So with the limited information on this subject, this is the way I incorporated it into my build. As also stated previously, most of the time these boats were tied up to the shores of the rivers, but there were times that they were at anchor. Given that many of the Western Rivers were not known for their great depths at the time, there was the possibility that there would not have been much anchor chain to haul in. Just my two cents. -Brian
  20. This was a real noggin scratcher when researching how to build the roof on the forward and aft structures of the paddlewheel house, and to be honest I can't remember if I covered it or not. Since there is only one known existing photograph of the Cairo it is really hard to tell how this was constructed. I stared at the picture for hours on end trying to figure it out. The HSR plans show the roof to be flat and I think they were drawn that way going off the Cairo photograph. But the more I researched and studied the picture I came to the conclusion that what it looks like in the Cairo picture that makes the roof look flat is the loading booms that were sometimes attached to the aft mast pole were stored on top of the forward roof when not in use. These booms can be seen in the Cincinnati photo in my previous post, and I added them as a deck detail on my build. Since the Cairo, Mound City and Cincinnati were all built at the Mound City shipyard, I took it upon myself that they would have all been built the same way. The contemporary photos of the Cincinnati and Mound City, show that the rooftops on these two structures do have a curve to them as well as the four other boats built in the Carondelet shipyard, so why would the Cairo be the only one with flat rooftops. This was my reasoning for building my version with the curved roofs. I may have gotten it wrong, but I think it will definitely be hard to disprove my theory that the Cairo had round rooftops as well. 😁 In the back of my mind I would sure like to know for sure, but there is just not enough information out there on it. I'm sure the guys in St. Louis have run into the same issue, but since there is a photograph of they aft end of the St. Louis, they had an easier time with this part. -Brian
  21. Eric, your solution to your pilot house conundrum worked out great. This was indeed an unusual shape to get correct and I struggle with it a lot. I ended up making several mock-ups with card stock for the sides trying to get the angles correct, and the forward three panels with their added thickness just made it that much more difficult. At least the kit provided a resin mold to somewhat ease the pain of getting the shape correct. For your comment above, I know I am a little late in responding, my apologies but my arm was giving me fits this past weekend and my pain medicine tends to make me a little loopy. However, the curved section over the paddlewheel was not "roofed" like the rest of the structures built around it. From the contemporary photos of the City-Class boats I was able to discern that this area was planked with much thicker timbers (much like the deck material) and the rest of the structures were thinner planks and tarpapered. This can be seen in the photos of the Cincinnati and St. Louis. The pitch of these roofs was minimal. Basically just enough to shed the rainwater off. I believe is set my pitch at about 3-4 degrees. Hope this helps. -Brian
  22. Another suggestion I could offer up would be, if the wood "armor" is even all the way around and matches up at the corners of the resin form, you could possibly glue them all in place on the resin cast and line them up at the top. Once everything dries, you could then do as Ken @Canute suggested and sand the bottom down on a piece of glass or other hard flat surface until the bottoms all match up. As a side note, to entice you on a previous post about a railroad ferry project, I took this picture in the Vicksburg Rail Road museum on our visit last year. I thought it looked like a pretty neat project, and something that I had thought about doing some day. My great-grandfather worked for Union Pacific years ago and was in charge of shuttling the RR cars on a ferry similar to this one between Baton Rouge and Port Allen. -Brian
  23. "Resistance is futile" 😁 It is a contraption alright, and real fun to try and sleep with at night. I've been dealing with the issue since early November, and since we had already met our medical deductible for the year I figured I'd go ahead and get it done. It was also under Admiral's orders. -Brian
  24. That would be an almost unimaginable sin to drill holes in such a fine hull. Glad you took the alternate route and went with the cradle idea. -Brian
×
×
  • Create New...