CASTING DEFECTS


Casting is not a perfect science, defects will happen even when you think you did everything right. On this page I'll try to explain some of these defects and maybe how to avoid them. This will be difficult because the same defect can have different causes. All the steps used in creating a casting will effect the quality of that casting. It will be up to you to determine which step in the process caused the defect, then correct it before continuing with the next casting. To explain this I'll start with pattern construction and work through to the finished casting, hopefully it will help.

PATTERN DESIGN AND QUALITY:
An improperly designed or constructed pattern is the first indicator that you get a defective casting. Sometimes you can't do anything about the actual design of the casting you'll have thick and thin sections, if it can be redesigned to avoid the do so, if not make sure you have a fillet in the corners of the pattern. The quality of the pattern can have a great effect on the casting. The casting will never have as good of finish as the pattern, so the smoother the pattern the better the casting. Things to avoid would be too little draft, deep pockets or holes, rough finish or no finish if a wood pattern, loose fitting pins on split or mounted patterns and adjacent thick and thin metal sections.

GATING:
Sometime gating doesn't seem like much, cut a channel in the sand and the metal flows into the mold, and your done. In practice, this may work for most home casting but, the reality is you may have just been lucky enough to get a good one. The gating system controls more then just physical defects, but more on that later, right now, it's just what we can see. I have discussed gating design in the GATING page, here we'll discuss the possible defects of incorrect gating.

MOLDING AND MOLDING SAND:
I have only used greensand at home so I'll keep this to what I know. To start with greensand has to be the proper "temper", this comes from experience. Sand that's too dry or too wet will give you a bad casting. I'll buy premixed olivine sand 'cause I'm too lazy to mix my own. Ok, I also have a foundry supply nearby. Ramming a mold too hard or too soft can cause problems. Normally a drag can be rammed as hard as you can, the cope, depending on the shape, can be rammed too hard. As an example, a plate shape could be rammed hard cope and drag, a deep bowl shape, open side up, should be rammed softer allowing the casting to shrink properly.

CORES:
Cores will present new problems to the casting. While working in the pattern shop I often got called into the foundry because something didn't fit just right, or I have a scrap casting. Only to find out that the molder had filed the core to make it fit easier, or he'd rapped the mold too hard and made the core prints bigger causing the core to move. If not properly vented, cores will cause defects. A pattern without enough clearance or draft for a proper fit will also be a problem.

METAL, MELTING and POURING:
Most of the problems encounter here can be eliminated with a few simple rules. If melting aluminum, brass and bronze use a different crucible for each. This will avoid contaminating the other alloys. The next is harder, try not to over heat the metal, for copper alloys this will result in burning off the lighter metals. Aluminum you don't want to pour too hot, if the heat is too hot remove the heat and let cool to pouring temperature, then pour. A pyrometer will help with this. I have yet to get one, but maybe soon. I'll try to get pictures of all the defects as they happen, and they will.

Casting shift can be caused by loose pattern or flask pins. Or possibly sand shifting during the roll over.
Changes in wall thickness could be rapping to hard to remove the pattern, core being made oversize or the molder filing too much off the coreprint. In this case I used a greensand core, rapping the corebox too hard caused the core to shrink some in diameter. Using a co2 core or some other harder core material could have eliminated this problem. A check of the core prints showed it should have been less than .032 clearance.
Crush, the cause could be coreprint with too little clearance, cores rammed over size by improper clamping, cope that is too heavy or the molder setting the cope too hard.
Sand wash is easy, make sure you have a smooth pattern, good draft, a fillet at every sharp corner, a properly rammed mold and good sand. Sounds easy, right.
Flash or Fin are caused by rapping too hard on the mold while trying to remove the pattern, setting the mold on a surface that isn't flat or if the pattern is mounted, the mounting board in not flat. The cope and drag were machined as one piece, in this case the thickness between the cope and drag of the machined pattern was thicker than the board it was set into. The only way to correct this fin would be to remove it from this board and insert the pattern into a board the right thickness. Luckly this was a one casting part and the flash was mostly on the ingate.
Shrink cracks & cavities, is the metal pulling itself apart while cooling. It's a rough surface inside the casting caused by thick sections of casting that are not risered or chilled properly. Or possibly too hot of a pouring temperature. In this case the wrong size riser was used, I increased the riser diameter by .25" and eliminated the problem.
Cold shut, improper gating, not enough cope height to get required pouring head, cold pour, lack of proper venting and wet sand are some of the causes.I poured too cold here, the right side is correct the tab didn't fill on the left. I increased the tempurature and got a good casting. The casting is .090 thick in this area.
Scab, is a rough area on the surface of the casting, sand may also be embedded in the casting. Causes are, mold that is not rammed hard enough or sand is worn out or improper temper, either way the sand lacks the ability to bond properly.

Paying attention to detail and develop a routine for your casting session will make thing go easier. Knowing the cause of a casting defect will help you eliminate the defect, next time.

· HOME · PATTERNSHOP · FOUNDRY · CNC PROGRAMMING · LINKS ·

All content on this web site is copywrite.
2006-2008