Blowholes and boss porosity in a grey iron pulley casting
A foundry pours grey cast iron pulley blanks of about 8.2 kg finished weight in green sand moulds on a jolt-squeeze machine. The pattern is a two-part split pattern with the boss in the drag. Pouring temperature is 1390 °C, sand moisture 6.0%, green compressive strength 0.55 kg/cm², permeability 60. The last two lots show two problems: (a) clusters of smooth round holes on the machined face of the web, and (b) a spongy porous zone at the top of the boss next to the riser. The riser is 55 mm diameter × 70 mm high on a boss of 90 mm diameter, and the sprue is 18 mm with the runner entering at the parting line. The customer's order for 200 pieces is due in ten days. Blowholes on the web are found at the customer's boring operation; boss porosity is usually removed during boring. Scrap is running at 14% against a 5% allowance in the quotation. Reused sand is shovelled and wetted but not mulled.
1. Identify the mechanism behind each defect, explain why the remedies for the two are not interchangeable, and say what you would fix first given the ten-day deadline.
Read the worked answer
Two different mechanisms - one change will not cure both.
(a) Smooth round holes on the machined web = gas porosity (blowholes). The sand is the suspect, not the metal. 6.0% moisture is well above the 3.5–4.5% normally run, permeability 60 is low for a jolt-squeeze mould, and reused sand that is only shovelled and wetted is not mulled, so clay is not coated onto the grains and the mould is both wet and tight. Steam and gas generated at the mould wall cannot escape through the sand or through vents and are trapped against the cope surface - the face that is later machined. The 18 mm sprue entering at the parting line adds a turbulent, poorly vented fill.
(b) Spongy zone at the top of the boss = shrinkage/feeding porosity. It sits at the last point to freeze, and the riser (55 mm diameter) is smaller than the boss it is meant to feed (90 mm diameter) and offset from it, so it freezes early and stops feeding. This needs more thermal mass, not drier sand.
Why the remedies differ. Drying, mulling and venting remove gas but will not feed a hot spot; enlarging the riser or adding a chill feeds the hot spot but does nothing for gas. Raising the pouring temperature would improve feeding and worsen the blowholes, so it is a fix for neither.
Fix first: the sand. Gas porosity reaches the customer's boring line unseen and cannot be inspected out after shot blasting, whereas boss porosity is normally removed in boring. Correct moisture to about 4%, mull the sand, vent the cope and pour a five-piece trial. Change the riser only after that trial, so you can see which change did what.
2. Quantify what 14% scrap means against the 5% quotation allowance, and set out the production and dispatch decisions for the next two lots.
Read the worked answer
What 14% costs. For 200 good castings you must pour about 233 instead of 211 at the 5% allowance - 22 extra castings, plus the machining time spent on pieces scrapped after machining. At a raw casting cost of, say, ₹260, that is roughly ₹5,700 of extra castings before machining; the 5% allowance in the price has already been consumed, so every point above 9% comes straight off the margin.
But the decision is not only arithmetic. Web blowholes are discovered at the customer, after boring - a line stoppage and a returned consignment cost several times ₹5,700.
- Pour a five-piece trial with corrected sand moisture and venting, then section each one: cut one web face to look for gas and one boss to look for feeding porosity.
- Ship nothing from the trial until the sections are clean.
- Hold the two suspect lots. Visual inspection after shot blasting will not reveal sub-surface gas, so either machine the web face in-house before dispatch (adds cost but moves detection upstream) or ask the customer for a concession limited to boss porosity.
- Log scrap by defect type over the next two lots. If gas has not fallen below 5% after the sand change, reopen gating and riser design rather than normalising a permanent 14%.