You find out at inspection. The sand knew at the muller.

Foundry Twin keeps the sand test, the melt, the pour and the rejection in one record, so finding the cause is a lookup rather than a meeting.

See it on your own records Follow one heat

Running in iron foundries in India and the United States. Built with Versatile Equipments, who have made foundry sand testing instruments since 1967.

It starts at the muller.

Moisture and compactability are tested a few times a shift and written in a register. Between tests, the sand going into the molding machine isn't measured at all.

The chemistry goes to a different file.

The spectrometer keeps its own export, and the furnace keeps its own log. Each record is sound on its own. None of them knows about the others.

The pour is over in minutes.

How long the ladle waited and which mold it filled decide a lot, and most plants never write them down. Metal cools while it waits.

A day later, 31 castings are scrap.

The sand, the metal and the pour are long gone. The question "what changed?" now means finding three records in three places and comparing them by hand.

With one record, you walk it back.

Every reading is already attached to its heat, its mold and its time. The 31 castings lead back to mold 2417, the molding machine that made it, and the sand that went in.

Then it does the analysis nobody has time for.

By hand, lining up a month of sand tests, melt sheets and inspection counts takes days, so it rarely gets done. The twin ranks every reading against the defect and shows the heats behind each result. Your engineers decide what is a cause.

What moved with blowholesCorrelation over the last 250 heats

Sand moisture at molding (sand lab)Lead0.62
Compactability (sand lab)0.44
Ladle wait before pouring (pouring)0.17
Pour temperature (pouring)0.11
Carbon equivalent (spectrometer)0.04

6.9 %rejected when moisture was above 4.0 %, over 38 heats

1.4 %rejected when it was in its window, over 212 heats

Known mechanism, wet sand and gas defects: supported. Example data.

A person makes the change. The twin checks it.

The action gets an owner and a date. From that day, the twin compares the days after with the days before, so you can see whether it worked.

Blowholes were one example.

The twin follows each family of casting defects back to the readings that drive it. These come built in, and any other defect you log can be compared with any reading you choose.

Blowholes, gas porosity, scabs, sand inclusions, metal penetration: Checked against sand compactability, moisture, permeability, green compression strength, active clay, grain fineness and sand temperature.
Gas and inclusions from cores: Checked against core gas evolution, strength, hardness, cure time and bench life.
Swells, mold shift and penetration from weak molds: Checked against squeeze and shoot pressure on the molding machine, and mold strength.
Misruns and cold shuts: Checked against superheat above the heat's own liquidus, pour temperature, pour speed, and the temperature lost between ladle and mold.
Shrinkage and centerline porosity: Checked against carbon equivalent, superheat, inoculation and pour temperature.
Chill and hard spots in gray iron: Checked against carbon equivalent, silicon, chromium, vanadium and inoculation.
Low nodularity, chunky graphite and dross in ductile iron: Checked against residual magnesium, base sulfur, the time from treatment to pour, and inoculation.
Hardness out of range: Checked against manganese, copper, tin and silicon, with section thickness taken into account.
Distortion and stress cracks: Checked against shakeout timing, and heat treatment soak time and temperature.

Each one runs once the plant records the readings it needs. Until then, the twin says what it can't test yet rather than guessing.

Read the guides on each defect

The knowledge that prevents scrap already exists in your foundry. It sits with the two or three people who have been there longest.

When they're not on shift, it isn't available, and when they retire, it leaves. A twin is how that reasoning gets written down while they're still there to correct it.

Start with the scrap register you already keep.

A plant needs no new equipment to begin. Each kind of data you add gives a further kind of answer, and the twin's role grows in four stages.

Monitor

Scrap, process and machine data in one place, with reports, control charts, alarms and traceability.

From the first upload

Diagnose and prescribe

Which conditions moved with which defects, what to change, and whether the change worked. A person decides.

Once process records join outcomes

Predict

The risk of a defect for a heat or mold before it is poured, and what-if tests before a change is made.

After a few months of learning your plant

Control

Chosen setpoints corrected inside limits the plant sets, with operators able to override at any time.

Only by agreement, once predictions have proven reliable

See how the platform works

One test. Four molding lines. Four different answers.

We went through about 1.38 million sand test cycles looking for rules that hold everywhere. Moisture against compactability, on the same instrument model, gave a different answer on every line.

Correlation of moisture and compactability on four molding lines
Molding line	Correlation	Paired tests	What it means
Line A	+0.60	Tens of thousands	Water drives compaction hard
Line B	+0.28	Hundreds of thousands	Present but muted
Line C	+0.07	Hundreds of thousands	Almost no relationship
Line D	-0.42	Tens of thousands	It runs the other way

A rule from a handbook or another plant may be wrong for yours. The answer for your plant is in your plant.

Read what the sand data shows

It also measures what no machine records.

Each connected product is useful alone. Joined to the twin, a pour time sits beside the misruns of the molds it filled, and a nodularity result sits beside the chemistry of its heat.

Floor Master: Cameras record what no machine logs: how long a ladle waited and which pour filled which mold.
Foundry Scope: Grades microscope images of cast iron to ASTM and ISO standards and puts the result on the heat.
Black Box: An autonomous sand lab that tests batch by batch, not a few times a shift.
Metallographic Black Box: Prepares and images metallographic samples automatically.
Versatile smart instruments: Sand and mold readings sent digitally as soon as the test is done, instead of copied by hand.

See the product family

Why we can say this.

Versatile has built foundry sand testing instruments since 1967, and they run in more than 9,000 foundries across 40 countries. Sahas Labs develops the platform, and Versatile sells it and supplies the testing equipment.

Foundry Twin runs in iron foundries in India and the United States. Its figures can all be traced to the records they came from, and a figure that can't be traced isn't shown.

Start by measuring one thing properly.

Pick the defect that costs you most, check that the instrument which should have caught it is in calibration, and start writing down the rejection. We can help with all three.

Talk to us Start logging defects, free