Photo #1 The hoop is shown here tack welded around the inside edge and then removed from the rest of the skeleton for final welding. The hoop will be clamped solidly to the table for this stage of the welding. If it is not clamped well it will warp something fierce...don't ask how I know.
Photo #2 Here is the completed hoop with the welds ground smooth. Note: you will also need to touch up the outside (non welded) face of the tube with the grinder because of slight kinks that result when the tube is bent to shape. Also, if you decide to do your bending with this method you better LOVE grinding - because you get to do a lot of it. Without a good tube bender and the right dies, however, I don't believe there is a better way to get the job done on this tight of a curve.
Photo #3 The finished hoop is positioned back on the skeleton and is being readied for welding.
Photo #1 shows the deck hoop positioned on the body skeleton and the two ends curved around and clamped onto the rear door jambs.
Photo #2 is a close up of the curve. This is a 90 degree bend with a 6" radius.
Photo #3 The hoop is clamped in place and ready to be tack welded. At this juncture I will only tack weld the cuts around the inside face of the tubing. This will prevent the tube from warping up or down if I applied too much heat to either the top or the bottom sides of the tube. I will take the hoop off and clamp it to a solid flat table to weld the top and the bottom and to finish the welds on the inside face.
Next I need to make the hoop/curved piece which wraps around the top back of the seating area of the roadster and attaches to the rear door jambs on each side. I am using 1x1 square tube for the hoop and have decided to fabricate a curve at each corner of the hoop with a radius of 6".
Based on my experience bending the top ribs of the rear deck, I am going to use a different method - the "slice and dice" - for bending the Deck Hoop since it is a much sharper bend. This method simply puts a series of cuts along the inside face, the top, and the bottom of the tube but leaves the outside edge in tact. By calculating the inside and outside circumference of this curve (pi X Diameter/4) I know that I need to remove just a bit over 1.5 inches of material from the inside face of the curve. I do this by making a series of 13 cuts using my chop saw which has a blade width of 1/8".
Photo #1 shows the set up to make the cuts. I learned from some earlier experimentation with the "slice and dice" method that the tubing gets VERY flimsy after you make a few cuts in it and it will droop down in the wrong direction unless you hold it or keep it supported. If you look closely at this photo you will note a second piece of 1x1 "support" tubing below the one I am going to be cutting. The two piece of tube are clamped together near the ends just to keep them in line as I move them through the chop saw.
Photo #2 Here are the 13 cuts completed by the chop saw. However, this does not quite complete the job. Since the chop saw blade is a circle, it does not cut the front side of the cut quite as deep as the back side.
Photo #3 To remedy that problem I used my skil saw with a metal blade set at the proper depth. You need to go completely through both sides of the tube all the way down to just touch the inside face of the final side of the tubing (the one which is left in tact). If you do not make a clean cut totally through the side it will cause kinking when you bend it. Also, you need to adjust the blade from time to time to compensate for wear (the wheel gets smaller as stuff wears off it so you need to lowering it now and then).
I then laid the driver's side skeleton on the work table and positioned a piece of 3/16th flat stock about midway up on the body section (running horizontally). I then put the sheet metal panel on top of that and clamped all around the edges. It seemed to work fine so I decided to try it with an even bigger crown by placing two pieces of 3/16 flat stock on top of each other - for a total crown of 3/8" - as shown in Photo #1. [My '32 pickup was built with approximately a 1/4" crown in the doors and they are about the same width as this rear panel of the rat is high - so I'm hopeful that a 3/8" crown will do the trick and eliminate any concave effect.]
Photo #2 shows the amount of crown from the end view (shot from the door jamb looking toward the rear of the panel). I will need to figure out how to fill this area so that it does not show but that will be a small price to pay if I can actually get a crown in the side panels.
Photo #3 shows another view of the panel clamped in place over the double piece of flat stock. With the panel firmly clamped down I could not see any visible distortion, wrinkles, or ripples. Based on this experiment it looks like I will be able to get a crown in the side panels, doors, and cowl after all.
My sheet metal order came in yesterday (up here in the boonies ya' gotta special order 18 gauge because nobody stocks it). So I'm getting a little ahead of myself and jumping into a bit of sheet metal work today. I want to try to see if I can put a slight crown in the side panels of the body in order to eliminate the concave effect which was discussed in journal entry #71.
Photo #1 shows the very low tech job of tracing the outline of the driver side skeleton section onto a piece of 4x8 18 gauge sheet.
Photo #2 I then scribed a "cut line" 1/2" inside the traced line along the top of the curve. By cutting the panel a bit short it will put my sheet metal joint right in the center of the 1x1 curved rib of the side skeleton. I'm hoping this will act as a heat sink and help prevent warpage where I have to weld sections of sheet metal together.
Photo #3 I used a nibbler to cut out the section. If you look along the bottom edge of the panel you can just make out a line about an inch from the edge. I added this inch to the panel so that it can be bent and welded to the bottom side of skeleton.