The build of the Prototype II has been completed with great success on some fronts, and failure on other fronts. The machine was successful in that it shears even more than I’d specified, and that it actually works. The failures were in how long the machine took to produce (see my initial build proposal), how much the materials cost, and that I didn’t finish the punch element.
The current state is that we have a functional prototype, but still don’t have CAD or full documentation on how to build it. My Sketchup model is not entirely up to date either; it needs to have the updated shear tables, updated clamps, the reinforcement plates on the back side, and parts of the vertical linkages which had to be torched away. If anyone is interested in building, I can tell you the changes I had to make. Just email me at brianna at kufadesigns dot com.
I do not have any plans for finishing the remaining documentation, as I have moved back to California and will be working full time in the fabrication industry. OSE is looking for a draftsman to finish the CAD and 2D fab drawings for the machine. See the summary of existing documentation.
First, for the successes: the machine has been tested and can cut up to 1”x10” flat steel, and 3/8”x4”x6” angle steel. It should cut up to 3/8”x6”x6” angle steel, but we didn’t have any on hand, so we didn’t test it. Here’s a video of the testing:
The cuts were far closer to perfection than I’d hoped. What I’d been looking for in the cut was that there be no burr, and minimal deformation along the length of the cut. The cut was cleaner than any cuts of similar magnitude which I had seen on an industrial machine. The cleanliness of the cut is very difficult to achieve, as it requires the blades to be perfectly parallel, and to have a gap between .007” and .010”, which required very precise machining.
The testing went smoother than expected, with only one failure. When I tested the first 1”x6” flat cut, the clamp failed, because the welds weren’t strong enough. It was initially threaded rod welded to the top of the shear table. After the failure, we drilled holes for the rod, and plug welded them from the opposite side. This has held up well. Here is the test in which the clamp failed:
It is very likely that the machine could do even larger flat cuts, probably up to 1”x12”, and possibly more. We stopped testing after 1”x10” though, as any failure would be very time-intensive and costly to repair, and the machine had already far-surpassed it’s goal of cutting 1”x6”.
The other major success was the fact that I, a person relatively inexperienced with machining, was able to build it, and learn all necessary skills. This means that the ironworker is buildable for somebody who hasn’t had much machining practice. I outlined most of the machining operations in my blog, so that any other beginners could learn from my mistakes.
Now for the failures of the machine. The angle shear doesn’t shear the thickness we were aiming for, the punch isn’t complete, and the machine took too much time, resources, and material to build easily.
First off, I designed the machine to be able to cut up to ½” thick angle steel, and did not meet this goal. For one reason or another, the machine will not open far enough to insert a piece of ½” thick angle. This could be an issue with the design, or I could have messed up machining somewhere. However, as far as I can tell, I machined it to the specifications of my design. I think it could be remedied by simply mounting the upper blade 1/8” or so higher on the upper arm. This needs more review.
The other design/build error was the lower arm. After having the arm professionally torched from an AutoCAD drawing, the linkage interfered with it, and I had to torch more away from it (which is no easy task for 3” thick steel). Yet again, I do not know if it was an error in execution (the company which did it will only guarantee a ¼” tolerance), or an error in my design.
Another failure is that I didn’t produce a punch for the ironworker. Marcin decided not to have me build the punch I had initially designed, because it would need to be removed if you wanted to shear, otherwise the parts would interfere with each other. I couldn’t see any other way to design it. We discussed re-doing the design to resemble a Piranha punch, but I simply ran out of time to do it before leaving Factor e Farm.
The other issue was the time and resources to build it. I do not believe it is cost effective to build your own ironworker using this design.
Ironworkers with similar capabilities cost $17,000+ to purchase used. However, these ironworkers usually include a punch, a bar shear, and some sort of coper-notcher. It would be possible to retrofit all of those to my machine, as I left the necessary space to do so, but it would cost $800+ extra for the necessary parts to do so, plus probably a weeks worth of machining.
All in all, it cost about $6500 in materials to build the machine. It requires many machines and tools as well, which only increases the cost. It took me about 4 months working full time on it to build it. If personal labor is worth $2K per month, this puts my machine at $14.5K+ to build. For the marginally lower cost than an industrial version, you really don’t get much more out of it, plus you lose the network of industry professionals to help you if anything breaks. It does cost less, but not significantly so.
If I were to re-do it, knowing what I know now, I suspect it would still take 1.5-2 months to build, which theoretically would lower the price to $10.5K. It would likely take a professional machininst or fab shop the same amount of time, or possibly down to 1 month, but they cost far more than $2K per month.
One of the reasons it took so long was that I had very little experience on mills or lathes, so I learned many things the hard way. One example is that I chose to torch out, rather than Mag-Drill, the holes for the bushings for the machine. The reason I didn’t Mag-Drill it, is that the bit it would require would cost $200+, and require a bigger-than-standard Mag-Drill. This choice meant that I spent far more time machining them, as torching hardens the metal; It took me 4-8 hours to bore out each of (12) 3” diameter holes. Had I Mag-Drilled them, it would take less than 1 hour each. I didn’t realize just how difficult it was going to be to machine thru the torched metal.
Another reason that it took so long is that I finished the machine at Factor e Farm, which really didn’t have the infrastructure I needed to build it. I spent a few weeks setting up the necessary infrastructure, like 3 phase power and wiring for the welders. If I hadn’t had to do this, it would have shaved off at least 2 weeks from the build time.
Overall, I am very proud of how the ironworker turned out. It ended up with very high-quality construction, as was proven with the cleanliness of the cuts. The major disappointments were that it took so long to build.