Treat a PathPilot conversion as a separate project, not as a feature of the machine purchase.
Screen the machine, its controller, and the cost of keeping it productive as separate items. A clean display or low purchase price cannot establish mechanical accuracy, program throughput, or whether a failed board can be replaced.
Write down the work the machine must do before comparing brands or ages: part envelope, materials, typical tool sizes, required accuracy, cycle length, and whether the programs contain short, dense moves or mostly long cuts. Also identify whether the job is a short lathe cycle, blocky milling, heavy roughing, or detailed 3D surfacing. Each places a different demand on the machine and its control.
| Workload | What to verify | Decision |
|---|---|---|
| Short lathe cycles or blocky parts | Mechanical condition, cycle execution, and whether the program fits or transfers reliably. | Low memory may be manageable; prove it with the actual program. |
| Large, heavy cuts | Spindle condition, rigidity, work envelope, tool interface, and dimensional results under cut. | A heavy machine with box ways can suit roughing, but inspect the particular machine. |
| Small tools or detailed 3D/HSM paths | Control processing response, motion quality, and finish while running the densest representative toolpath. | A file-transfer workaround does not cure pauses caused by control throughput. |
A Fadal 4020 example ran large aluminum and stainless parts, including work described as about 26 by 8 inches, while detailed 3D moves could pause and file-size limits required drip feeding. That makes it a workload-specific example, not a guarantee for every 4020. A separate Fadal 6030 example favored heavy material removal over surfacing; feed had to be reduced to retain accuracy, and excessive cutting force showed up as fretting at the CAT40 taper. Use those distinctions to choose a test, not to assume a model is good or bad.
For the HL-2, evaluate the actual lathe cycle and required accuracy; for each Fadal VMC, run the intended kind of milling program. If the seller cannot demonstrate the workload that matters to your shop, move the machine into the repair-project category rather than treating it as a production replacement.
Demand a live cycle before calling it production-ready
A static inspection shows what is present, not whether the machine can complete work. Ask to see the machine powered and running now. Record the nameplate and control identification, note active alarms, and watch it home, jog through the usable travel, start the spindle, and operate the tool changer if fitted. Then run a representative program through a complete cycle. A machine that last ran years ago has not passed this check.
Watch for axis faults, unexpected pauses, inconsistent motion, spindle noise, failed tool changes, and alarms that reappear after reset. Check whether the operator can recover from a normal stop and restart the program safely. Ask what faults have been repaired and which alarms are being bypassed or cleared to make the demonstration run. A display that looks normal while the machine is idle is not evidence that the electronics, drives, or I/O will hold up in a cycle.
If the machine has not run, ask for a diagnosis and price it as a machine requiring repair, including the possibility that its fault has multiple causes. Do not use “it ran before it was unplugged” as a substitute for a current test. Age-related corrosion and worn electrical contacts can surface after storage or as a circuit is used; a long idle period does not protect the controller.
Keep the demonstration job relevant but controlled. Confirm the workholding and tooling, use an uncomplicated known program first, and progress to the representative cycle only after basic functions pass. If a production-like program cannot finish without faults or unexplained interruptions, stop the evaluation and identify the failure before negotiating a purchase price.
Measure geometry instead of trusting a clean demo
A machine can complete a cycle and still miss the dimensional or repeatability needs of the job. Check axis motion across the usable travel and measure backlash or lost motion, repeatability, and spindle runout with appropriate indicators and methods. Inspect the spindle taper and tool retention, then cut a representative test part and measure it independently. Compare results with the machine builder’s specifications and the tolerances required by your work; do not invent acceptance limits from another machine’s results.
Record measurements by axis and position, not just “looks good.” An error that grows at one end of travel points to a different repair decision from an error that repeats uniformly. A test cut also exposes effects that an indicator check alone misses, including cutting-load deflection, thermal drift, vibration, and how the control executes the path. If the seller will not permit a meaningful test, treat accuracy as unknown and include inspection or repair cost in the offer.
Distinguish the machine’s mechanical capacity from its ability to produce a particular surface finish. Heavy mass and box ways can support roughing cuts, but they do not guarantee high-speed response or accurate small-tool surfacing. Conversely, a modest memory limit does not automatically disqualify a lathe running short cycles. For a Fadal candidate, put both the shop’s roughing operation and its most demanding finish path in the test plan if both matter.
Record mechanical findings separately from control faults. A geometry problem will not be fixed by replacing the user interface, and an old controller cannot explain away worn ways or a damaged spindle taper. If a machine fails the geometry or test-cut requirement, obtain a machine-specific repair assessment before counting on it for production.
Identify the control before valuing its memory
Read the controller’s exact model and options from its identification plate and diagnostic screens; do not infer the hardware from the machine’s build year. Photograph the control, computer, drives, and communication connectors, and obtain the electrical drawings and operator/programming documentation if available. Determine whether the control is a dedicated CNC or PC-based, what storage it uses, how it receives programs, and whether its display and storage components still have a repair path.
Some older PC-based CNC installations have used 386- or 486-era computers, DOS, floppy disks, or other aging storage. Other controls may run programs from internal memory or through a serial connection. These are different failure and maintenance paths.
Memory capacity and machining capability are related, but not interchangeable. Low memory can limit the size of a stored program; a lathe running short, blocky programs may be unaffected. Detailed surfacing programs create a separate challenge: they can generate many short blocks, and the control must process and deliver motion quickly enough to keep the path moving. Determine how the candidate handles both the file size and the block rate in the shop’s actual program.
If the machine has a Fanuc or another established CNC control, identify the exact generation, options, manuals, and spare-part sources. A familiar brand name is not a service plan. Ask whether the specific control can accept the needed file format and motion commands, and whether someone can diagnose its alarms and replace its failed hardware. If the exact control configuration cannot be identified, do not assign it a value based on brand reputation alone.
Prove DNC transfer and dense-path response
When a control cannot store the full program, drip feeding or DNC may let it receive program blocks while machining. That solves a file-size problem only if the control accepts the transfer reliably and can process incoming blocks fast enough. A DNC connection cannot make a slow processor calculate dense 3D motion faster; that is why one machine can run long, simple cuts well yet pause on intricate surfacing.
If the machine depends on RS-232, inspect the port and cable connection, identify the required transfer software, and test the actual communication setup. Verify the control’s handshake and transfer settings from its documentation rather than guessing them. Send a representative program, confirm that the control receives it intact, and run the complete cycle. Check for transfer errors, waiting between blocks, pauses, alarms, or a finish that changes when the path gets dense.
Memory cards, adapters, and storage upgrades may exist for some controls, but compatibility and setup depend on the specific control. Treat them as a documented option, not a universal USB conversion. A storage upgrade may increase how much code the control can hold without improving its processing rate, old-board reliability, or availability of replacement parts.
Run two distinct tests where the shop needs both: a program near the expected file-size limit and the most demanding path by block density. If transfer succeeds but the tool pauses during dense moves, the control’s processing behavior is the limiting factor. If the transfer fails or cannot be repeated consistently, resolve the communication path before approving the machine for production.
Trace spare parts and recovery options
Before purchase, identify the control boards, drives, power supplies, spindle components, relays, encoders, and tool-changer components that a failure could take out of service. Match each critical item to a supplier or repair channel, then get availability, price, and lead time in writing where possible. Ask whether replacement parts are new, repaired, or of unknown condition. A listing on a resale site does not prove that a board is compatible or serviceable.
Check whether the machine has current electrical drawings, alarm information, parameter backups, and a known recovery procedure. Find out who can diagnose it locally and whether that person supports the specific machine and control. If only one hard-to-find board can restore production, calculate the cost of keeping a tested spare or the cost and time to retrofit before the fault occurs. Production risk is the period the shop waits for diagnosis and parts, not just the purchase price of a failed component.
Electronic reliability declines with calendar age as well as operating use. Corrosion and contact wear can affect electronics and relays even when run hours are low; a machine may operate for years and then become difficult to repair when a discontinued component fails. Parts prices and availability can worsen as a control ages. Do not assume that a functioning machine will remain repairable just because it completed the seller’s demonstration.
For a machine intended to replace active capacity, lack of a parts path is a stop condition unless the purchase plan includes a funded repair or conversion and the shop can absorb the downtime. If a fault is already present, get a diagnosis from a qualified machine-tool or control technician before accepting a seller’s guess about the failed component.
Price PathPilot or another retrofit before purchase
Do not count on PathPilot running a 1998 Haas HL-2 or a Fadal until official control support confirms compatibility with that exact machine and its hardware. A familiar interface does not establish that the control can operate the machine’s axes, spindle, tool changer, feedback devices, I/O, or machine-specific functions. Ask for a written compatibility answer using the machine and control details collected during inspection.
Scope a retrofit as a control-system engineering project, not as installing a computer and screen. Confirm what happens to the existing axis motors and drives, encoders, spindle drive, power supply, limit and home inputs, coolant, tool changer, work offsets, and machine interlocks. Some retrofit designs can reuse selected existing components, but reuse depends on their electrical and control compatibility and must be tested. Require a wiring plan, defined responsibility for commissioning, a list of retained and replaced hardware, and a working acceptance test before committing.
Budget time as well as hardware. The retrofit needs integration, electrical troubleshooting, parameter setup, motion tuning, and proof that machine functions operate correctly. A PC-based conversion can become an extended project, and a control that is easier to use is not automatically equivalent to a factory-integrated machine. If production must begin promptly, compare the retrofit schedule and downtime against buying a machine that already passes its acceptance test.
Keep temporary restoration and permanent repair separate. If the factory control passes the inspection and the shop can support it, using that control for an initial production run may be lower-risk than replacing it immediately. A retrofit can remain a separately scoped improvement after compatibility, cost, commissioning time, and safety-related machine functions are resolved.
Calculate delivered cost and downtime exposure
Compare machines using total installed and production-ready cost, not the auction or seller price alone:
Total ready-to-run cost = purchase + inspection + rigging and transport + repair + control or communication work + startup + initial spares + expected downtime cost
Use current local quotes for each term. Include diagnostic labor, DNC software or communication setup if required, retrofit engineering if planned, and the loss of capacity while the machine is unavailable. Compare that result with a newer used machine that can run the shop’s test parts and has a verifiable support path. Auction examples and asking prices from another time or region do not establish what a machine costs now.
Account for the physical move before paying. Confirm the machine’s actual weight and envelope from its documentation, the transport and rigging path, door and ceiling clearance, floor capacity, and the final installation location. A Fadal 4020 is heavy enough that rigging can be a substantial part of the job; confirm whether it can pass through the shop’s entry rather than discover the problem after purchase. Moving from another state can still work financially, but only after the full logistics quote is known.
For a small-batch shop, downtime may cost more than the difference between two purchase prices. Write down how long the shop can wait for a board, technician, or retrofit before the missed work outweighs the savings. If only a specialist can keep a control running, include that dependency in the comparison. A cheap candidate that cannot meet the production start date is not the low-cost option.
Run a written acceptance test before release
Use this sequence to close the evaluation on the resolving branch—the machine is mechanically suitable, its control can run the work, and the parts path is credible:
- Record the machine nameplate, exact control model and options, alarms, installed storage, communication method, and documents supplied.
- Run the machine under power through homing, full usable axis travel, spindle operation, and tool changes. Record faults and confirm they do not recur during the representative cycle.
- Measure motion and spindle condition with appropriate inspection methods, then cut and independently measure a representative part against the shop drawing and the machine’s published tolerances.
- Transfer and execute the shop’s actual program through the intended method. Test both a realistic file size and the densest path; note transfer errors, pauses, or finish problems.
- Confirm in writing the source and lead time for critical spares, the person or company able to service the control, and the cost of any known repair or retrofit before approving the purchase.
If the machine passes, preserve the original control configuration for the production trial, save its documentation and backups, and run the intended job before making any optional retrofit. Verify first-piece dimensions, repeatability, tool changes, cycle continuity, and the ability to recover from a normal stop. Keep any permanent control conversion as a separate decision with its own priced scope and commissioning test.
If any required check fails, stop the acceptance process at that branch: get an independent diagnosis and repair quote, or decline the machine. Do not release a failed demo to production based on a promised storage upgrade, hoped-for board repair, or future controller conversion.
A current cycle proves more than an idle display: it exercises the control, drives, I/O, spindle, axes, and tool changer together. A machine that has been unplugged for years needs a fault diagnosis and repair budget before it can be treated as production-ready.
Why does DNC not fix slow 3D machining?
DNC can supply program blocks when internal memory is too small, but it does not increase the control’s ability to process dense moves. Run the actual detailed path and watch for pauses or finish degradation before deciding that drip feeding is an adequate workaround.
Can PathPilot run a 1998 Haas HL-2 or a Fadal?
Do not assume compatibility from the machine brand or the PathPilot interface. Give official PathPilot support the exact control, motors, drives, encoders, spindle, I/O, and tool-changer details and obtain confirmation before including a retrofit in the purchase plan.
Stop the purchase or conversion if the machine cannot pass its representative cycle, its critical parts have no credible recovery path, or no qualified integrator will own the retrofit and commissioning. Escalate control faults and compatibility questions to official machine-tool or control support; where the legacy product is no longer supported, use a qualified CNC retrofit integrator and require a written scope before work begins.