Question 1 of 6: Casting defects in a no-bake sand-moulded cast-iron tee

Question 1: Casting defects in a no-bake sand-moulded cast-iron tee

Question text not reproduced: the examination questions are © Engineers and Geoscientists BC. Open the official past paper (linked at the top of this page) to read the question, then follow the worked solution below.

Approach. All three defects are read off the same mould: locate each defect on the casting, ask what is locally different there (gas source, metallostatic head, section thickness), and name the process variable that controls it.

(a) Why the gas bubbles appear only at that one location

Air-set (no-bake) sand is bonded with an organic resin — a furan or phenolic urethane cured chemically at room temperature. When the mould is filled at roughly 1350 °C, that binder pyrolyses instantly, and every gram of resin releases a large volume of hydrogen, carbon monoxide, carbon dioxide and hydrocarbon vapour. The core is the worst offender because it is surrounded by hot metal on all sides while the mould is heated on one face only, so the core reaches decomposition temperature throughout its section and its gas has nowhere to go except through the metal or out along the core prints.

That is the whole explanation for the localisation. The gas evolved in the core travels to the metal–core interface, and the bubbles that break away rise through the liquid iron by buoyancy until they are trapped by the first solidified skin. The location marked in Figure S3(b) is the upper surface of the horizontal branch of the tee, directly over the core — it is simultaneously (i) the roof of the cavity above the gas source, so buoyant bubbles collect there and cannot escape further; (ii) a cope surface, where the solidifying skin forms early and caps the bubbles in place; and (iii) remote from the core prints, which are the only vented path out of the core. Everywhere else the bubbles either escape into the mould through the permeable sand, are flushed out ahead of the advancing metal front while the cavity is still filling, or rise clear into the sprue and riser and leave the casting entirely.

The controllable causes are therefore core gas evolution and core venting, not the melt itself: excessive binder content (a resin level above about 1.5 % of sand weight is generous), incompletely cured or too-fresh cores, moisture pickup in storage, core sand that is too fine or too tightly compacted to be permeable, and cores whose vent holes or coke/wax vent ropes do not run out through the core print into the atmosphere. The remedies follow directly: cut the binder to the minimum that gives handling strength, allow full cure, vent the core positively to atmosphere through the print, raise sand permeability, and keep the cores dry.

(b) Factors that caused the penetration defect

Metal penetration is liquid iron infiltrating the pore space between the sand grains, so that the casting surface becomes a rough metal–sand composite that has to be ground off. It occurs where the liquid pressure exceeds the capillary pressure that keeps the metal out of the pores. The governing inequality is a comparison between metallostatic head and surface tension over the effective pore radius, so every factor below either raises the left-hand side or lowers the right:

• High metallostatic head. The defect sits near the bottom of the inside diameter — the deepest point of the cavity — where the ferrostatic pressure from the whole column of iron plus sprue is greatest. The pressure at that point is simply ρ g h, so a tall sprue or an unnecessarily high pouring head aggravates it.
• Coarse or under-compacted sand, giving large inter-granular pores. The core surface is the loosest sand in the mould, and no-bake cores are often blown or hand-rammed to a lower density than the mould faces.
• No refractory mould wash. A zircon-, graphite- or magnesite-based core coating fills the surface pores and is the single most effective cure. Its absence on the core is the most likely proximate cause here.
• Excessive pouring temperature and long local solidification time. The bottom of the inside diameter is a hot spot fed continuously by descending metal; the longer the metal stays liquid against the sand, the more time it has to infiltrate.
• High fluidity of the alloy. Grey cast iron with high carbon equivalent is exceptionally fluid and low in surface tension compared with steel, which is why penetration is a classic grey-iron defect.
• Low binder level or burnt-out binder at the hottest region, which destroys the sand’s hot strength and lets the grains separate under the metal pressure.

(c) Factors that led to the enlargement at “C”

Point “C” is a local swelling of the casting outline — the cavity grew after the mould was closed. This is mould-wall movement (swell): the metallostatic pressure pushed the sand face back before the casting skin was strong enough to resist it. Note where it occurs: at the base of the tee, the deepest part of the mould, where the pressure head is highest, and on a heavy section that stays liquid longest, giving the pressure the longest time to act.

The contributing factors are all mould-strength and mould-restraint variables. Insufficient ramming or blowing density leaves the sand face compliant. Under-dosed or incompletely cured air-set binder is a particular risk, because a no-bake mould gains strength on a chemical clock: if the mould is poured before the resin has developed full strength, or if the sand–catalyst ratio drifted, hot strength is low and the wall yields. Sand that has been over-worked or contains excessive fines behaves the same way. Erosion by the incoming stream can enlarge the cavity at the same location if the gating directs metal against that face. Finally, if the cope was not adequately weighted or clamped, the whole cope lifts slightly under the ferrostatic pressure, producing an enlargement plus flash at the parting line.

The corrections are to raise mould hardness and hot strength at that face, allow full cure before pouring, reduce the pouring head (or pour with a bottom gate to cut the pressure surge), weight or clamp the flasks properly, and reposition the ingate so the stream does not impinge on the swollen face.