Jump to content

thibaultron

NRG Member
  • Posts

    2,903
  • Joined

  • Last visited

Everything posted by thibaultron

  1. Look up Barbatos Rex on YouTube. He does extensive paint testing, and has some on AlClad..
  2. Thank you for posting this tread! I'm sure what solutions you find will help the rest of us, that plan to buy one of these printers in the future.
  3. The interesting thing about the leather suspension, unlike a modern car spring, it was not for the comfort of the passengers, but to reduce the shock loading on the horses, as the body bounced around.
  4. Part 009 Next I had to clean out the inside corners of the upper piece for the slots for the false keel, keel, and chocks, to remove the rounded corners left by the milling bit. After further cleaning up the false keel, I used it and the keel to check my progress on cutting out the corners. Next I used the false keel as a jig to align the upper and lower frame slots, and glued the upper section in place, then removed the false keel and placed weights on the upper section and along the exposed lower half to keep everything level while the glue dried. Next came a step that should have been in the instructions, before the frame assembly. There is a temporary resin fill piece between the upper and lower halves of the frame, that positions the upper tips of the frame pieces, away from each other. After the frame is glued, two small pieces of wood span areas of this gap, and form part of the finished frame. The only reference to this is the graphic of assembling the upper half, were it is shown in place between the frame sections. This shows the two pieces that will be installed when the inserts are removed. The entire upper tip piece will stand proud, of the lower frame half, held in place only by the chock and the two inserts, once the frame is finished. After I install the two wooden support pieces, I’ll trim the head off the insert, and reinstall the tail piece, to support the gap, until I reach a point where it is no longer needed. The problem, is that both the upper and lower tip pieces have to be fitted to the resin piece, just like the chocks and keel parts. With the lower half assembled this proved awkward. One thing to be aware of, is that the groves are angled, not straight across the frame section. Yes, I didn’t notice this on the first two pieces. The resin piece required minor cleanup to remove some small vent nubs along the flat tail piece. This is a picture of one side cut to fit the insert, and the other before I started. Here is both sides fitted. Notice that the lower frame half tip (the upper one in this shot) “re-kitted” itself during the process. The same thing happened on the other side. So, after fitting the parts, I reattached the two lower frame pieces, and, after the glue dried, proceeded to glue one of the upper pieces in place. I shortened the tail piece of the resin spacer a little, to make sure that it was not hitting the adjacent frame section. I found that modified clothes pins had just the right spacing to clamp the parts parallel. I reversed the spring so that the handle ends were the side closed by the spring pressure (see the photos). I used a clamp on either side of the chock, and at the top of the frame to hold the tips in line. I used a file to add weight to hold the tips tight to the insert. This method does not normally work on a single chock holding two frame sections as the pressure tends to push the parts away from each other. For this use though, everything is locked in place by the insert and the previously glued sections, and the clamp ends are spaced by the lower frame surfaces. Yes, the third photograph shows a nice lump of excess glue at the chock, but that end of the chock will be filed down to match the frame section, which is thinner than the mating end of the tip piece. If you look at the joint to the right in that picture, you will see that the same thing will have to be done to one side of that chock, also. The forum software only allows a maximum of 16 pictures per post, so I will continue when I get a bit further along is completing this frame.
  5. Part 008, or “How to continue to do it wrong!” During the build, I will continue to show mistakes, why I made them, and (Hopefully) how I fixed them. This is my first POF build, and basically my first wood sailing ship kit, though I have built HO scale wood buildings and wood RC warships. By showing my mistakes, I may help others avoid them. “Good judgement comes from experience, experience comes from bad judgement. " In retrospect, I should have started with one of the inside frames, where mistakes could have been more easily hidden. As I mentioned before, Frame 6 would not fit back between the shims when I tried to put it back into place. After getting everything un-glued, I rebuilt the lower section of the frame first, so far so good. Then, I said to myself, “Self, you know this frame has to fit onto the keel, doesn’t it?”. Self replied, “You’re right, better try that!" So I found the two pieces for the keel, the keel, and false keel. The first thing I found was that both pieces had been machined incorrectly!! As you can see above, the blank apparently shifted a little when they turned it over to mill the back, and the upper and lower surfaces came out misaligned. Not by much, but after cleanup both the parts were too short by twice the difference in misalignment (had to correct both top and bottom). At least the relative depth of the lands on the false keel are correct. So, I thought, “I can just add a shim between the keel and false keel, to make up the difference.” Looking at the keel. I then discovered that the two rabbets are now also misaligned, height wise, and the keel will sit lower in the assembly jig. Not good! I’m going to write to CAF, and see if they will replace these two parts. Luckily, these were the only parts in that blank. After some cleanup it looked like this, Now, to continue the frame 6 story, it turns out that, no, the keel would not fit in the frame slot! ********!! I tried widening the slot with files, etc., but was just marking up the underside of the chock, and cutting the frame piece unsquarely. So, out came the Isopropyl alcohol again! I cleaned up the unsquared area with my disk sander and got everything put back together, with the keel now fitting nicely. Then, for fun, I cutout the bottom piece of the upper half of the frame, and placed it on top of the lower pieces. Double *******! The lower frame curve was too shallow! So once again, do over. If you look closely you can see that the front assembly, does not have the same curve as the back piece (6A-6A-1). Note however how nicely the edges of the wider sections match. Sigh. I cleaned up the outside surfaces of the upper piece to remove the attachment points. As can be seen in the above picture, I then applied tape to the mating surface of the part, to prevent it from sticking to the lower pieces. Then I glued the lower pieces in place and clamped the upper piece in place to align them. I had to sand the wedge areas of the chock to lengthen it to fill the longer gap between the two frames. There is enough meat in the part to do this. After everything was correctly positioned I clamped the upper section to the lower ones, until the glue had grabbed, then removed the upper section, and placed a weight on the lower parts, to keep them flat. Finally I got the two curves to match, unfortunately, now the keel slot in the bottom half is too wide. Also unfortunately this frame is an end frame, and, naturally, the exposed side. I’ll figure out that problem later. Here is a picture with the bottom section glued back together, and the upper piece set on top. Now the curves match! And, also now, the slots in the lower half are wider than the ones in the upper section. After everything had dried, I glued the rest of the lower Frame 6 parts in place. Finally done with this assembly!
  6. Great looking model, well done!
  7. Just found this build. Sorry to hear about your back trouble, but I'm glad you are feeling better! Interesting subject.
  8. Many years ago, I took one of the DUKW tours out of Tampa, FL. After we got out a ways they asked if anyone wanted to drive it. My hand was up so fast, I think I got friction burn from the air resistance! No one else was interested, so I got to drive it for about 20 minutes. Had a wonderful time!
  9. On Sinagots (a small French fishing boat) Have two masts with lug sails. They are set up with the top spare for one mast on one side, and the top spar for the other on the oposite side. Thus one sail, or the other, is against the mast depending on which tack they are on.
  10. Looks like an indexing setup. Place a piece between the centers, and cut X number of flats, or drill X number of equally spaced holes around the circumference. Used with the milling machine. For example make a round bar, into a hex bar, then drilling a hole centered on each face, Drilling four holes in a mast, at 90 degres apart, etc.
  11. I don't know how sharp the teeth are on your model, but T-Rex teeth close up were a blunt triangular cross section shape, with serrations on the edges. Designed to hold up to their massive bite, and bone crunching ability. More like a triangular punch. Look up pictures of them.
  12. I built a similar one. Use CA for the joints. The press fit joints between parts, are not strong enough to support the weight of the pieces, especially the skull. I tried Canopy Clue, the first time, but it failed after a few weeks. It makes a great model!
  13. I watched a lecture given by the curator of the Wasa museum. They had a replica of one of the cannons cast (I think a 24 pounder), and test fired it against a replica of the side of the Wasa.The cannon ball passed completely through the side of the hull, even if there was a frame in the way! The difference was fewer splinters (the main killer in such a situation), if the ball only hit an area without a frame. He said they calculated that the ball would likely also go through the other side!
  14. Even if you have paint chips, for modern equipment, you may still have problems. Gray paint, in particular, can vary between batches, even from the same manufacturer. A story I heard in the 70s is: An upper ranking officer was going around and comparing the gray paint on radar units, and almost all of them varied from the paint chip he had . Afterward he sent each installation a paint chip, and ordered them all to repaint the units in the proper color, or else (he used different words, but this is a family forum). At the appointed time he reinspected the units, and all but one still did not match! The other unit maintenance personnel were demoted. Later some of them asked the person in charge of the one that passed, how he did it. He responded that he repainted the chip, at the same time he painted his units.
  15. Vallejo makes two types of paint, Model Color (for brushing), and Model Air (prethined for airbrushing). The Model Color can be airbrushed with thinning, but the Model Air has a finer pigment flake size. A majority of the two paints have matching colors. Their site has charts for color matching the two, as well as matchings for several of the other brands.
  16. Nirvana, could you post a link to your T37 build, a search of the forum did not turn it up.
  17. varnish should work, as you've already varnished parts of the interior. As far as the PVA/sawdust mixture, others will have to chime in. I've never tried it.
  18. Use very light cloth on the outside, and at least a coat or two of resin on the inside, after the outside is finished.
  19. Use a marine epoxy type resin, not the polyester (automotive) type resin! The polyester has a tendency to pull away from wood. The marine epoxy type is designed to stick to wood and will stick much better, in the long run. If you find either has trouble setting, heat the hull in a semi sealed container (plastic tub, etc.) in the sun. The resins use chemical heat as part of the setting process, and thin sections may not generate enough heat to do the job. You may have to paint on a second coat of just resin, after sanding.
  20. I should have started with the bottom parts, and fit them between the shims, then built the upper sections.
×
×
  • Create New...