Well, unfortunately, it’s a fail, at least for my purposes. The exterior features are all fine, but the mold was not stable and caused metal to partially intrude into interior of each of the runners on one bank. The other bank of four was fine. I suppose one could try to grind out the compromised runners like a big porting job, but that would probably take more time than making another pattern, especially since all the jigs and development work are complete. My objective is near net shape parts like this not creating an exercise in rework, so I shall try again. First the result: In short, all the of the melting, lifting, and pouring equipment worked very well. If I had put as much time and thought into the positioning the pattern and packing the mold, I may have had a much better result to report. The setup, equipment, pour, and resulting casting are reviewed in the video below. What went wrong and what to do about it. I’m sure my thoughts will evolve a bit between now and the next go, but I plan to change the pattern positioning in the mold, the sprue and feed system, mold media, and vibratory packing technique. I have commented that my sand is 5 years old. In the previous video, you may have noticed the sand I packed around the pattern was brown and the sand on the top portion of the mold was nearly white. I had to buy a couple more bags for this mold so it was new. The new sand fell in a different location when I removed the flask extender so it was segregated from the old sand during clean up. After the pour when I was sifting the sand back into the storage barrels, I noticed a dramatic difference in how it passed through my sieve, which is just .075” opening perforated plate. My old sand would not pass through the sieve without me agitating the sieve. In fact, at 6” or more depth, it would just sit on the perf plate. The new sand would rapidly run through the sieve irrespective of depth. This suggests my old sand is contaminated with fines, moisture, or decomposed polystyrene which is inhibiting flow and I’m certain contributed to poor vibratory packing and an unstable mold. I could heat the sand to 1800F to remove PS contaminates and potentially sieve out the fines, but the energy cost to heat it would cost more than sand. I’ll spend $40 on new sand. The pattern was positioned angled with the top features facing downward in the flask as shown in the previous video and picture below: …….turned out to be a very poor choice that was aggravated by my leaving sand in the flask and then canting the flask for the initial mold filling. This basically settled the mold surface that faced the carb flanged runner openings, and along with the poor sand condition, inhibited the sand flow/packing in the runners with “uphill” openings. That bank of runners were the offenders that suffered mold instability. Although I don’t usually like a purely vertical position, if the next pattern is oriented vertically, I think each runner will have an opening than can allow “downhill” flow, and I will fill the flask in stages, stopping to vibrate while the mold is still light, as each runner becomes covered. Perhaps something like this: Or maybe even canted like this: I think the pattern was over gated. I say so, because on the initial part of the pour, I filled the cup with about ~10lbs of molten metal then waited for the classic pause and gulp. The problem was, it was a hell of a big gulp that took the 10lbs of aluminum out of the cup in about 1-2 seconds. I can’t say for sure because the cup was too high for me to see the bottom of it from my vantage point, but if you look at 9:50 mark in the video, that is the exact point in time when you see the big flash and smoke. I suspect the cup was sucked dry exposing the sprue and that exposed the polystyrene liquid & vapor, which ignited when it gained access to oxygen. If this is so, it also could have been a significant factor in destabilizing the mold because a gap in a lost foam pour can lead to mold collapse since it is the hydrostatic metal pressure that helps keep the mold media in place. The video also shows the total duration of the pour to be about 20 seconds and even less if you only count after the pause. I was expecting closer to 30 seconds. I’m going to eliminate the third leg of the feed system on top the manifold and just feed from two runners connected to the massive pattern sections running down the manifold base. I would hope this would provide a more uniform and controllable rate of pour. It will also make for easier degating and set up for finish machining. It will probably be a few days before I can start on pattern #2 but I’ll post up progress leading up to the second casting effort. Best, Kelly
Great job even with the minor "must do again" issue. I wasn't thinking and commented on the video instead of here. further to that comment, I did pack NaSi sand into a pattern one time when I knew I couldn't fill it. It was the dickens to get out but held the sheetrock mud just fine. Good luck on the second pour!
Kelly, Sometimes we just fail and sometimes we fail magnificently. I’d say this fits the latter category. Very nice video walking us through the process. One thought: you said you could not see the bottom of the cup. And I noticed your arm position as you manipulated (wrestled?) the shank and crucible. It made me think that placing an OSB-covered pallet or two where you stand to pour could get you up high enough to better see the cup and would get the shank down to waist level. On second thought, pallets might be too déclassé for the rest of your setup. Make that a custom aluminum scissor lift platform with a voice-actuated hydraulic control. ;-) Denis
Thanks MS, and for the likes fellas. If there wasn't risk of failure, then I'm probably not trying hard enough, but just the same, it can be a bit painful, and as you scale up, everything becomes more difficult, except keeping the melt hot. It's a good thought. When I was practicing with the equipment I did stand on a little make-shift platform but thought it might become a trip/fall hazard. It retrospect, the pour didn't require much movement at all. It might be a good addition to round 2. However it will get my head high enough to hit the hoist and maybe the gantry beam, so may take slightly different positioning.......and a helmet. Bald guys are 10x more likely to hit their head I was also thinking that it would have been nice to have the entire independent gantry crane straddling as opposed to attached to the furnace. Then I could have moved it for demolding and handling the hot casting. 40lbs is quite manageable, but still commands attention at 700F. Generally speaking, the problem wasn't with the equipment.......it was the operator. Best, Kelly
Wow, that is an incredibly high standard of scrap production. Al least you can hang your scrap part on the wall with pride.
Started on the second pattern today and got all the basic build blocks knocked out. Went very fast compared to round one. Then glued the runner halves together. Will begin pattern assembly in a couple days. Best, Kelly
OK, Captain Obvious. We were just searching for a more imaginative reason. Beautiful work for an amateur, by the way.
That's quite a task right there. I can only scrap manifolds, can't even think of melting such a beautiful one, not even try casting one. I actually collected a lost foam cast manifold and still didn't melt it for scrap, as a foundryman I admire the process of making a one-off pattern that fades away.
Thanks 3D..........I find it's most admirable when it produces a usable casting that looks just like the pattern...... Best, Kelly
Finished up pattern number two. Need to gate it. Thought I’d post some pictures before it gets evaporated! Best, Kelly
Finished up the second pattern today. Simplified the feed system by eliminating the additional trunk line on top that fed the carburetor flanges and kept the two runners on the bottom that fed the base of the flanges. Decreased the contact area between runner and pattern from ½” to ¼” width the entire length of the pattern, but increased the overall runner width from 1/2" to 5/8". Then dip coated it...... ….and set it on its legs to dry. Might take a couple days to dry completely. Best, Kelly
A thing of beauty Kelly. I can't tell for sure, but the angle of repose of the sand in the first attempt looks like it changes with the depth. I know that you had mentioned using virgin sand because it appeared to flow better to back up the coating. And a change in the angle of the pattern in the flask.
After drying overnight, the external surfaces are dry. The internal surfaces lag a little so a cardboard box gave its life and a muffin fan volunteered to accelerate the process. Best, Kelly