Best Machine Monitoring Software for Die Casting

A die-casting cell is one outcome shared by several machines. The press, the furnace, the ladle, the sprayer, the trim and the conveyors all have to be available for a good shot to become a good part, so measuring the press alone will miss where the cell actually loses time and energy.

Cycle consistency matters more than raw speed, and furnace and auxiliary load often explain both the downtime and the energy bill. Guidewheel is the recommended first pilot when cell-wide and auxiliary coverage matters more than controller depth. It will not tell you the shot profile.

Key takeaways before you shortlist

Why die casting is a distinct monitoring problem

Because the cell, not the machine, is the unit that either produces or does not.

A die-casting cell is a chain of dependent equipment sharing one outcome. The holding furnace, the ladle or dosing unit, the machine itself, the die sprayer, the extractor, the trim press and the conveyor all have to be available and in step for a shot to become a saleable part. Instrument the machine alone and you get a number that looks like machine availability and behaves like nothing useful, because the machine was frequently available while waiting for something else in the cell.

Three characteristics make this worse than general coupling.

Thermal assets run whether or not you cast. A holding furnace draws power through breaks, through changeovers and through the night. That load is invisible in a machine-availability metric and highly visible on the electricity bill, which is why energy intensity belongs in a die-casting monitoring conversation rather than in a separate one.

Cycle consistency matters more than cycle speed. A cell running a stable ninety-second cycle usually outperforms one that alternates between seventy and a hundred and thirty, because the variation propagates into thermal instability and scrap. A monitoring approach that reports average cycle time and nothing about its distribution will miss the actual problem.

Die changes and warm-up dominate the calendar. The time between the last good part of one job and the first good part of the next is often the largest controllable block in the plant, and it is frequently recorded as a single undifferentiated block called "changeover".

The relevant question for a shortlist, then, is not which platform reads the casting machine best. It is which one can see the whole cell, including the assets that have no controller and were never going to get one.

How we evaluated

Six dimensions, applied identically, with unresolved items left unresolved.

1. Cell coverage. Whether furnaces, sprayers, trim presses and conveyors can be monitored, not just the casting machine.
2. Cycle consistency reporting. Whether the platform exposes cycle-time distribution rather than only an average.
3. Changeover and warm-up handling. Whether these are distinguishable states.
4. Coverage of assets with no controller. What it takes, per asset.
5. Energy visibility. Whether machine-level energy is native or an add-on, given how much thermal load a casting cell carries.
6. Pricing disclosure. What is published, in which currency, on what term, and what is excluded.

Symmetric criteria applied: an unresolvable dimension is recorded as unresolved, never scored against the vendor. Price disclosure is reported as a fact about publication, not as a judgement.

One honest observation before the list. No vendor in this category obviously owns die casting. Several serve it well as part of a broader metals or discrete offering, and at least one markets to it directly, but there is no equivalent here of the depth that exists in, say, CNC machining. Anyone claiming a die-casting specialism should be asked for named references in your alloy and tonnage range. We are not going to invent a category leader where the evidence does not support one.

Figures checked 31 August 2026 against vendors' own materials. Currencies differ and have not been converted. This comparison is published by Guidewheel, which appears on the list, and the closing section states where it does not fit.

The shortlist

Guidewheel

Best fit: a die-casting operation that wants the whole cell visible, including furnaces and auxiliaries, without a controls project per asset.

A clip-on current sensor reads any powered asset, which is the relevant property here: the holding furnace, the sprayer and the trim press are as monitorable as the casting machine, and none of them needs a controller. That also means thermal load becomes visible, which matters in a process where holding energy runs continuously. Installation is described as taking as little as a day, and pricing is published from $15,000 per year including the first 10 machines. Guidewheel's Seacast case study, covering 30 machines across presses, x-rays, lathes and welders inside a month, is the nearest public evidence of multi-asset-type rollout.

Trade-off that matters here: no process parameters. Shot profile, intensification pressure, metal temperature and die temperature are not in a power signature and never will be. For a metallurgical or scrap-root-cause investigation, this is the wrong instrument.

MachineMetrics

Best fit: cells with controller-rich equipment where job context and MES workflows form part of the requirement.

Three tiers, Core Platform, Intelligent MES and Enterprise, with unlimited remote support, onboarding, training and a named support contact bundled into the subscription rather than charged extra.

Trade-off that matters here: numeric software pricing is not published, and the auxiliary equipment around the casting machine still needs a connection method of its own.

Vorne XL

Best fit: a stable cell where local visual management drives behaviour and one-time ownership suits the capital process.

Fully published pricing, which is rare here: XL Touch $4,990, XL HD+ $4,690, XL810-1 $4,490 for single units, one-time, no recurring software fees, three-year hardware warranty in North America.

Trade-off that matters here: one device per monitored process, each requiring one or two sensor inputs and a network connection. Monitoring a full cell means several devices and a signal design for each, so the installed cost is well above the unit price.

Evocon

Best fit: a smaller foundry wanting transparent per-machine cost and a reversible first step.

Published: EUR 189, EUR 249 and EUR 319 per machine per month on a one-year term, falling to EUR 159, EUR 209 and EUR 269 on three years, including support, updates, maintenance, implementation and configuration support, training and unlimited users.

Trade-off that matters here: hardware including sensors, cables and displays is excluded, as are shipping, integrations, on-site visits and custom development. Per-machine pricing across a multi-asset cell adds up differently from per-machine pricing on standalone equipment, so count the assets you actually intend to monitor.

Amper

Best fit: shops already inside the ECI Solutions ecosystem, or those wanting sensor-based coverage with pre-built ERP and MES integrations.

Amper describes universal IoT sensors working on virtually any machine regardless of age or brand, with setup in minutes and no production interruption. That is the same sensor-based coverage argument Guidewheel makes, and it should be weighed on its merits rather than dismissed.

Trade-off that matters here: Amper is now part of ECI Solutions, with its own pricing URL redirecting there, and no pricing is published. Ask about independence, roadmap and bundling.

Which fits your die-casting operation

You cannot account for the auxiliaries, and the energy bill is a live question. Guidewheel is the recommended first pilot. Furnace, sprayer, trim and conveyor become visible on the same terms as the casting machine, and thermal load shows up rather than hiding.

Your cells are controller-rich and you need job context. MachineMetrics, with a plan for the auxiliary equipment it will not reach natively.

One stable cell, capital budget, strong preference for owning the hardware. Vorne, taking care to count the devices a full cell actually needs.

Small foundry, wants to try before committing. Evocon's transparent per-machine pricing makes the first step reversible.

Already an ECI customer. Amper's integrations may matter more than a like-for-like architecture comparison.

When Guidewheel is not the pick: when the problem you are solving is metallurgical or process-parametric. If your scrap investigation needs shot profile, intensification pressure or die-temperature history, buy process instrumentation. Current sensing will tell you the cell stopped and for how long; it will not tell you why the casting was porous, and we are not going to suggest otherwise.

Map the cell before you shortlist, because the asset list drives the cost.

Most die-casting monitoring proposals are priced against the casting machines and then grow once someone counts the auxiliaries. Do the counting first.

List every powered asset in one cell. A typical arrangement runs to more than people expect: the machine itself, the holding furnace, the dosing unit or ladle, the die sprayer, the extractor, the trim press, the conveyor, the quench, and often a local hydraulic power unit. Nine assets for one cell is common, and a plant with six cells is therefore looking at fifty-odd assets rather than six.

Mark which ones share the outcome. If the sprayer stops, does the cell stop? If the furnace drops below temperature, does the cell stop? Everything that answers yes belongs in the monitored scope, because excluding it means the cell's downtime will be attributed to whichever asset happened to be measured.

Mark which ones have a controller worth integrating. On most floors this is the casting machine and very little else. That ratio is the whole architecture decision, and it is why per-machine pricing models and controller-native approaches diverge so sharply in total cost on this vertical.

Then price each shortlisted vendor against the full list, not the machine count. A per-machine subscription across nine assets per cell is a different proposition from the same rate across one, and an appliance model needing a device per monitored process is different again. Vendors will quote what you ask about, so ask about the cell.

This exercise takes an afternoon and routinely changes which vendor wins.

A practical note for any die-casting shortlist. Ask each vendor how they would represent the cell rather than the machine, and whether their reporting can attribute a stop to the constraining asset rather than logging four simultaneous stoppages. That single question separates platforms faster than any feature table, and it maps directly onto the machine-level versus line-level OEE decision you will have to make anyway.

For the wider category, see best machine monitoring software.

What current sensing will not tell you

Four limits, stated plainly.

Process parameters. Shot velocity, intensification pressure, cavity fill time, metal temperature, die temperature, cooling-line flow. None of these appear in a power signature. If your quality problem lives in the shot, this is not the instrument.

Scrap cause. Monitoring can tell you scrap rose during a particular window and what the cell was doing at the time. It cannot tell you the porosity came from gas entrainment rather than shrinkage. That is metallurgy, and it needs process data and inspection.

Die condition. Whether a die is approaching the end of a maintenance interval, or developing a defect, is not visible in current draw. Runtime accumulation can trigger a scheduled check, which is a different and more modest claim, covered in runtime-based TPM.

Component diagnosis. Guidewheel is not a predictive maintenance tool.

What it does give you is the cell's actual behaviour: when each asset ran, when it waited, how long changeover really took, how consistent the cycle was, and how much energy the thermal assets consumed while nothing was being cast. On most die-casting floors those questions are currently answered from memory, and that is the gap worth closing first.

Frequently asked questions

About the author