Where a machine tool spindle gets repaired — and why China is now on the map

Ask ten maintenance managers where to send a failed spindle and you will get ten answers, most of them shaped by one bad experience. That is a poor basis for a decision that can idle a machining centre for six weeks. The question worth answering is narrower: what is actually being bought, and what is the full invoice?

What fails inside a spindle

A spindle is a rotating assembly held to tolerances measured in microns, and it fails in a small number of predictable ways. Bearing raceway fatigue is the most common. Beyond that: worn or fretted taper (the conical seat that locates the tool holder), scored or undersized shaft journals (the ground surfaces the bearings sit on), degraded motor windings in motorised spindles, and encoder or sensor faults.

Each failure maps to a different repair operation, and this is where quotations diverge so widely. Bearing replacement alone is a contained job. Taper regrinding changes the spindle's geometry and needs the machine to be re-zeroed afterwards. Journal repair may mean grinding undersize and then re-chroming (building the surface back up by electroplating, then regrinding to size). Two quotes for "spindle repair" can describe entirely different scopes of work.

Repair or replace?

The conventional inflection point sits around 50–60% of replacement cost. Below it, rebuilding usually wins on both cost and lead time; a new high-speed motorised spindle can carry a delivery time measured in months. Above it — or when the housing itself is damaged — replacement starts to make sense, and a rebuild quote that ignores housing condition is worth questioning.

Three kinds of repair provider

Most of the disagreement in this market comes from comparing unlike things. There are three distinct supplier types, and they are good at different problems.

Comparison of the three supplier archetypes. Vendor-specific figures are as published by the providers and are not independently audited.

Dimension	OEM factory service	Regional independent rebuilder	Asian repair cluster
Parts authenticity	Original, guaranteed	Original or equivalent, varies	Original or equivalent, ask for invoices
Balancing & test gear	Full, to OEM specification	Usually full, shop-dependent	Increasingly full; verify per vendor
Typical lead time	Varies; can be long for older models	Days to a few weeks	Adds international freight and customs
Warranty	Manufacturer warranty, clear venue	Often 12 months or more; local	Adequate on paper; cross-border venue
Cost position	Highest	Middle	Lowest, sometimes by 40%+ on published rates
Best fit	Warranty units, exotic or obsolete designs	Production-critical, short-window failures	Price-sensitive fleets, non-critical machines, planned overhauls

The objections worth taking seriously

US rebuilders have made a direct case against shipping spindles overseas, and it deserves to be read on its merits rather than dismissed as self-interest. Motor City Spindle Repair's argument rests on four points:

• Downtime measured in weeks. International freight plus customs can idle a machine far longer than the repair itself takes.
• Verification difficulty. Without OEM-level balancing and precision standards, vibration, runout or premature failure can follow a repair that "looks fine".
• Communication friction. Time zones and language make it harder to track progress or confirm that a specification was followed.
• Warranty enforcement. If a repair fails, returning the unit or making a claim across a border is materially harder.

Each of these is a real cost, and any honest cross-border comparison has to carry them. Notice, though, what the list is actually describing: lead time, quality assurance, communication and accountability. None of those are inherent to geography. A domestic shop that returns a spindle without a balancing report and without run-in data has the same defect, minus the freight bill.

The counterweight: OEMs repair in China too

That single page changes the nature of the debate. Cross-border spindle repair is not a grey-market workaround whereby cost-cutting buyers accept unknown quality. It is part of a premium manufacturer's own service network. When the feedback is taken as the reference point, "repair in China" is a service tier that a Swiss brand is willing to put its name on, not an exception to be argued about.

There is a straightforward read of this. Motorised spindle repair is concentrated, tooling-intensive work. It needs temperature-controlled assembly space, dynamic balancing machines, taper grinders, run-in stands and a stock of precision angular-contact bearings. Regions that manufacture spindles have those assets already, which is why repair capacity follows manufacturing capacity.

Published price reference points

Concrete numbers are scarce in this market, but Chinese sourcing platforms publish service rates openly, which makes them a useful anchor for the bottom of the range. On 1688, spindle repair services for Japanese DISCO and TSK dicing-saw spindles and Korean RPS air spindles are listed between roughly ¥3,600 and ¥4,700, with vendors claiming 40%+ below original-factory pricing and turnaround inside 72 hours.

Two caveats belong next to those figures. First, they are vendor-stated and platform-mediated; they are not audited results, and the machines in question — dicing saws, air-bearing spindles — are a narrower class than a machining-centre spindle. Second, a low headline price that omits freight, customs, insurance, crating and the downtime on either end is not a price. It is a component of one.

The breadth of what Chinese independents now advertise is arguably more telling than the pricing. Providers such as Shanghai Zhuojue publish service coverage for GMN, FISCHER, IBAG, WEISS, HSD, MAZAK, Okuma, Fanuc, NSK and SKF spindles, while Shanghai Juyan lists roughly thirty European and Japanese brands alongside a temperature-and-humidity-controlled assembly area. Coverage of that range implies accumulated tooling and documentation, which is harder to fake than a price list.

When cross-border repair is the wrong call

Being straight about the limits is more useful than a general recommendation. Cross-border repair tends to be the wrong choice in these situations:

• The spindle is still under machine warranty. Third-party work can void it, and the arithmetic rarely recovers the loss.
• Single machine, no spare, revenue-critical. Four to eight weeks of freight and customs is the deciding cost, not the repair quote.
• An exotic or obsolete design. Where drawings, fixtures and setup knowledge sit with one OEM, that OEM is often the only realistic source regardless of price.
• Regulated production requiring traceable calibration records. Aerospace, medical and automotive PPAP environments may need documentation that a given vendor cannot supply. Establish this before shipping, not after.
• Damage beyond the spindle. Housing distortion, crashed tooling interfaces or damaged machine geometry need a different, often on-site intervention.
• Nobody in-house can inspect the returned unit. Without someone able to check runout and taper contact on arrival, a marginal repair will not be caught until it fails in production.

How to evaluate a repair partner

The differences between a good and a poor rebuild are almost entirely visible in documentation and equipment, not in the workshop's size. A useful set of questions, in roughly the order they matter:

1. Failure analysis before authorisation. A disassembly report with photographs, identifying what failed and why. Without it, a scope of work is guesswork. Reputable shops provide this as a matter of course; some charge for it and credit it against the repair.
2. Bearing sourcing, stated explicitly. Which brand — GMN, FAG, NSK, SKF, or a stated equivalent — and at what precision class. Ask for the bearing invoice if the quotation claims original parts. Budget rebuilds with unstated bearings are where most repeat failures originate.
3. Dynamic balancing to a named standard. Dynamic balancing (correcting mass distribution in two planes so the spindle does not vibrate at speed) is usually quoted against an ISO 1940 balance grade, at a specified RPM. "Balanced" with no grade and no RPM is not a specification.
4. Run-in test data. A run-in (a controlled period of operation before dispatch, letting bearing temperature and vibration stabilise) should produce recorded output: temperature rise over time, and a vibration signature. Ask for the printout, not an assurance.
5. Taper contact pattern. After any taper regrinding, contact should be checked with bluing and reported as a percentage. This is the check that determines whether tool holders locate properly at speed.
6. Journal and housing capability. Can they grind journals, re-chrome undersize surfaces, and confirm housing bore condition? If those operations are subcontracted, that is worth knowing up front — it adds time and handoffs.
7. Itemised written quotation with exclusions. Freight, crating, insurance, customs brokerage and re-testing after repair should each be visible or explicitly excluded. A single lump figure hides where the cost sits.
8. Warranty terms including freight. Duration, what is covered, and specifically who pays return freight if a warranty claim is needed. Cross-border claims hinge on this clause.
9. Two references in your own region. Not testimonials — named shops that will take a call, ideally with machines in the same class as yours.
10. Test evidence at both ends. Photographs or video at teardown and at final test. This is also the practical mitigation for the time-zone and language objection: evidence travels better than conversation.

Terms worth knowing

Term	What it means in practice
Runout / TIR	Total indicated runout — how far the spindle deviates from true rotation, measured at a specified point. The headline accuracy number on a repair report.
Angular contact bearing	The precision bearing type used in most machine tool spindles. Must be replaced with the correct arrangement and preload.
Preload	Controlled internal force applied to the bearing set at assembly. Too little causes vibration; too much causes heat and short life.
ISO 1940 balance grade	Standard defining permitted residual unbalance. Lower grade letter and number means tighter balancing.
Taper contact	Percentage of the tool holder's cone surface touching the spindle taper, checked with bluing. Low contact causes poor finish and chatter.
Ceramic vs steel bearings	Hybrid ceramic bearings run cooler and allow higher speeds at higher cost. Worth agreeing on before the rebuild, not after.

What this means for buyers

The spindle repair map has changed shape, and the change is not really about price. A Swiss manufacturer running a repair facility in Changzhou is the clearest available signal that China now sits inside the legitimate service network for high-end spindles, not outside it. That is a statement about manufacturing depth: the tooling, the bearing stock, the balancing equipment and the temperature-controlled assembly space all follow where spindles are made.

None of which removes the risk on a given job. The remaining variable is process discipline — whether the failure is documented before a price is quoted, whether bearings have a stated origin, whether balancing is referenced to a standard and a speed, and whether run-in data exists as a record rather than a reassurance. Those are questions that travel equally well to a shop twenty kilometres away or nine thousand.

Buyers who treat repair as a rotation problem rather than an emergency tend to come out ahead: a spare spindle on the shelf converts a crisis into a scheduled overhaul, and a scheduled overhaul is exactly the job where cost structure can be chosen rather than dictated.