I have been fooling around with this part for a year or more now, when I get some spare time, but it has been good for my brain to arrive at a workable plan. The original part was badly corroded and it took quite some thinking to work out parting-lines and how to get the pattern out of the sand. Step 1 was to make the pattern in four pieces. This proved later, to not be the best method. Step 2 was to saw the pattern into 4 pieces and add locating dowels. Step 3 was to saw the original in half, repair the corrosion with filler and make two core boxes. Step 4 was to waste time and money trying to make sodium silicate cores. In the end I found it absurdly easy to make baked sand cores instead. Step 5 was to make the patterns for and cast two chills. Step 6 was to do a trial pour and the photo shows a mini disaster. One quarter of the pattern moved, whilst being moulded and created the holes on the base. Not sure why the outlet pipe is holed but it is probably because the pattern is not aligned with the core. Lesson learnt, is to next time make the core-box first and then align the pattern to the core, when gluing the parts, Cheers Charlie
Learning seems so expensive. But maybe that is what makes it so valuable. I love/hate learning. I'm betting the next iteration will be spot on. Thanks for sharing your learning. Maybe we won't have to also go through the identical (mis)steps. That is a pretty complex piece. Sorry the silicate did not work out. Denis
Thanks Dennis. I will try the sodium silicate again one day, when I feel the need. For now I am enjoying molasses, water, flour and sand as it works so well, Cheers Charlie
It's interesting to see someone's process of getting to an acceptable result. I don't cast often and typically not very complex parts so I feel I am still quite behind the curve yet and have much, much farther to go. I would appreciate hearing your recipie if you're willing to share.
Dtsh, my baked core recipe is 40:1 silica sand to lindseed oil with 10% molding sand and some flour to add green strength. I bake at 350* on core plates in a toaster oven for about an hour for thick cores. I glue with flour paste and bake for a short time to drive off the moisture.
Fine sand as required and measured by weight. 1% water. If the sand is damp, add only as much to make a fistful stay balled. 1% corn flour/starch. Starch and flour are the same thing. 2% molasses. Method Weigh the sand and flour and mix dry in a container. Weigh the molasses. I do this by mixing the molasses and water in an old milk jug. Use hot water to help dissolve the molasses into the water. Add the 2:1 molasses/H2O solution to the dry mix and mix until no dry flour or sand, can be found. If necessary add H2O if the mix can't hold a ball, when squeezed in the hand. Measure the thickest part of the core to the nearest 1/2". Bake at 150C for 10 minutes for every 1/2" of core thickness. This works very well for small cores; on very thick and heavy cores I bake for 20 minutes for every 1/2" of core thickness. Allow to cool after baking. I bake my heavy cores on a piece of steel 1/8" tread-plate. The plate must be flat and not susceptible to warping in the oven. Glue cores with PVA glue. Store in a warm dry place until needed. Advantages Cheap ingredients No special tools required Smells nice when cooking – unlike linseed which nearly caused a divorce! Easy to strip core with masonry bit and drill Disadvantage Time delay between making and use, requires prior planning. Accurate scales required Core-box can be anything that holds the shape; e.g. plastic, wood, metal, paper-mache. Sharp corners tend to fret; generally the core is more friable than sodium-silicate, or resin-bonded sand. Cheers Charlie
One thing I have done with delicate features easily damaged on silicate cores is to reinforce/strengthen the delicate feature with thin cyanoacrylate glue. I just drip some onto the cured area until it looks like the area sand has all been affected. The glue seems to kick off well on its own (maybe the sodium silicate tends to catalyze the glue???) and the area that was easily damaged is now pretty darned hard and robust. The glue seems to have no adverse affect on the casting. Similarly, I have used the same glue to reattach pieces of SS cores that I occasionally bumble and break or just to strengthen a vulnerable area of a core. I have never had a glued feature break off when filling a mold with iron. Most likely (don't know for sure, of course) it would also work well on the baked cores. Denis
Good idea Dennis. I have been thinking about it since you PM'd me and I might trial some PVA glue, or a solution of same, to see if it has any useful effect, Cheers Charlie
Please also try water-thin cyano glue. I like it because it cures instantly, adds no moisture, penetrates the sand for at least an eighth of an inch, and it has worked. Special emphasis on the last quality. Denis
Success, now that I have taken a more serious approach to keep my multi-part pattern together. Typically for similar parts , I can get away with a 2-part pattern, but this one needed 8 parts and two chills. Good for my brain to work it all out and it gives me confidence to try something even more challenging, Cheers Charlie
If I make another one I will try to remember to take some photos. Just as tricky is devising ways to machine it, but now that I have a milling machine, drilling holes on the PCD is a piece of cake, Cheers Charlie
Cyanoacrylate adhesives are activated and cure with moisture. "The cyanoacrylate group in the monomer rapidly polymerize in the presence of water to form long, strong chains."
