The first prototype arrives, it fits the wheel exactly, and on the second night it chips at the door track. Nothing about the part is defective; the requirement it was built to simply never mentioned the impact it would meet eleven times an hour.
That is the characteristic failure of a custom development programme, and it is a specification problem rather than a manufacturing one. The process below is arranged so that each stage removes a class of wrong answer before money is spent on tooling.
Defining the application and grip target
The first stage produces a written statement of what the part has to do, and it is the stage that determines whether the programme succeeds.
Four inputs belong in that statement. The grip target, expressed as what the machine has to be able to do on the ice rather than as a target number nobody can measure. The load, taken from the machine weight, the wheel position and the way the machine is driven. The duty cycle, including hours on ice and the number of hard-surface crossings per shift. And the surfaces the part will meet, because a program that includes matting and a concrete apron is a different problem from one that runs only on ice.
Where the machine already runs a studded set, the statement should also record what the current part does well and what it does badly. A development programme that starts without that baseline usually reproduces the existing part with a different price.
What the development programme has to decide
The programme has five decisions to make, and each one is made against the application statement rather than against a preference.
The decisions are the tip form, the carbide grade, the shank and hole fit, the retention method, and the acceptance test. Taking them in that order matters, because the tip form and grade set what the part can do, while the fit and the retention set whether it survives long enough to do it. The acceptance test comes last because it is the only method that can confirm the first four.
Each decision also removes a class of wrong answers. Fixing the tip form eliminates the geometries that cannot engage the ice at the load available; fixing the grade eliminates the compositions that cannot survive the duty cycle; fixing the fit and the retention eliminates the assemblies that would work in a drawing and fail in a wheel. A programme that works in this order rarely needs more than two sampling rounds.
Geometry and grade development
Geometry and grade are developed together, because a tip form that works in one grade may not work in another.
The geometry work starts from the engagement mechanism. The point has to concentrate enough load to penetrate the ice at the load the machine applies, and the shoulder behind it has to transmit the turning load into the rubber without cutting it. Changing the tip angle changes both at once, which is why geometry development needs a measurement of engagement rather than an opinion about sharpness.
The grade work runs in parallel and in the opposite direction. Hardness and grain size trade wear resistance against toughness, and the duty cycle decides where on that line the part belongs. A machine that crosses a hard surface regularly needs toughness even at the cost of wear interval; a machine that runs only on ice can take a harder grade and gain life from it. Standard test methods for hardmetals, including hardness and transverse rupture strength, are published by ASTM International, and the material supply background is described by the International Tungsten Industry Association.
Prototype sampling
Sampling converts a specification into parts, and the number of samples should be set by the test rather than by the budget.
The sample set has to be large enough to test to failure on a sample of parts rather than on one, and it has to include parts at the extremes of the tolerance so that the fit can be assessed across the range. Where the part has a retention feature, the sample should include it in the same condition as production, because a prototype retention feature that is machined rather than formed will not behave the same way in the wheel.
Two practices make the sampling round productive. Record the dimensions actually produced against the drawing, so that the test results can be read against the parts that were tested rather than against the parts that were intended. And keep one sample of each variant as a reference, because the comparison that matters later is against the parts that were in the test.
Custom carbide studs: the five decisions the programme has to lock
Five decisions lock the part: form, grade, fit, fix, test.
Each one removes a class of wrong answer rather than adding a preference. Form decides how the point engages, grade decides how long it lasts, fit decides whether it seats, the fixing decides whether it stays, and the test decides whether any of the other four were right.
Locked in that order, the five give a drawing that a production line can hold and a buyer can inspect against. Locked in a different order, they produce the common failure: a part that is dimensionally correct, made from the right material, and still wrong for the machine, because the requirement was never written down.
Two of the five are cheap to change later and three are not. The grade can be altered between production batches if the drawing allows a range, and the tip form can be modified with a tooling change. The shank fit, the retention method and the acceptance test are difficult to change once tooling exists, which is why they belong in the first sampling round rather than the second.
Bench testing before field testing
Bench testing exists to eliminate variants cheaply, so that the field test measures the machine rather than the part.
Run the bench work in the order below.
- Dimensional inspection of every sample against the drawing, with the results recorded per part.
- Hardness and material verification on a sample, using the standard methods agreed in the specification.
- Seating check in the receiving body, at the extremes of the tolerance rather than at nominal.
- Pull-out test on a sample to the stated acceptance load.
- Wear comparison against the current part, using the same test conditions for both.
A variant that fails at step four does not go to the field, which is the point of the sequence. Reference measurement practice for the dimensional and seating checks is published by the National Institute of Standards and Technology.
Field testing on a machine
The field test measures three things the bench cannot: engagement as the operator feels it, retention under real turning loads, and wear against the actual ice.
Set the test up so that the result is readable. Use two machines or two wheels where possible, one with the current part and one with the prototype, so that the comparison is made in the same conditions. Record ice hours, door crossings and any incident that loads the studs unusually. Inspect at fixed intervals rather than at the end, because the rate of wear is more informative than the total.
The result to look for is not a single number but the shape of the change: whether the prototype moved the limit from tip wear to retention or the other way round, and whether the operator can still steer the machine at the end of the test. Either answer is a usable outcome, provided the test recorded enough to explain it.
Tooling and production ramp
Tooling is committed only after the field result is understood, and the ramp is where the drawing stops being a document and becomes a process.
Work through the ramp in the order below.
- Freeze the drawing, including the tolerances that will be inspected on every lot and those that will be sampled.
- Confirm the tooling produces the tip form and the retention feature within tolerance, on a trial run rather than on a single part.
- Set the process parameters that affect retention, including insertion or cure conditions, and record them with the batch.
- Define the acceptance test and the sample size per lot, and apply it to the first production lots at the full rate.
- Agree the packaging that protects the tips, because a sharp part packed loosely arrives with damaged points.
The parameter that most often changes between prototype and production is the one nobody wrote down, which is why the ramp records process conditions alongside dimensions.
Documentation and approvals
The documentation set is the deliverable that keeps the part reproducible, and it should be agreed before the first production order.
- The controlled drawing, with the inspection dimensions marked.
- The material or grade specification, including the acceptance range.
- The process parameters that affect retention, recorded per lot.
- The acceptance test method and the sample size, with the results recorded per lot.
- The lot marking convention, so that a field part can be traced to its batch.
Machine-level interface conventions that a fitted part has to respect are published by SAE International. Where the part will be exported with a machine, the destination import requirements are summarised by the International Trade Administration, and it is worth confirming the tariff treatment of the finished part rather than assuming it follows the stud.
What ends a development programme early
A programme that stops after the first field test usually stops for one of four reasons, and all four are visible earlier.
- The requirement was written as a preference rather than as a measurable target, so no test can confirm it.
- The baseline was not recorded, so nobody can tell whether the prototype improved on the existing part.
- The retention method was left to production rather than fixed on the drawing.
- The number of variables changed between the first and second round was too large to interpret the result.
The fourth is the most common, and the remedy is simple: change one variable per round, even when that means a third round.
Programme timeline expectations
The timeline is driven by the decisions rather than by the machining, and the stage that consumes the most calendar time is usually the buyer’s own specification work.
A programme with a defined requirement and an existing drawing to base the design on moves through sampling and testing in a small number of rounds. A programme that starts from a general description spends its first weeks establishing what the part has to do, and that work cannot be compressed by pushing the factory. Where a season deadline exists, the sequence that works is to complete the specification and the first sample outside the season and to use the season itself for the field test.
Cost drivers of a custom programme
The cost of a custom programme sits in four places, and only one of them is the part price.
The table below sets out where the money goes and what moves each item.
| Driver | What moves it | What the buyer controls |
|---|---|---|
| Tooling | The complexity of the tip form and the retention feature | Reusing an existing profile where it meets the requirement |
| Sampling rounds | How precisely the requirement was defined at the start | Writing a measurable grip target and recording the baseline |
| Testing | Whether bench and field testing are done once or repeatedly | Fixing the acceptance test before the first sample |
| Engineering time | The internal effort to define and approve the specification | Assigning one owner for the requirement on the buyer’s side |
| Unit price at volume | Grade, tolerance and quantity | Restricting tight tolerances to the dimensions that function |
Written as a formula, the programme cost is tooling + sampling rounds x sampling cost + testing + engineering time + unit price x quantity. Buyers who push hardest on the last term and ignore the first four usually find that the cheapest quotation produces the longest programme.
SENTHAI develops custom carbide wear parts on the same process chain used for its production range, including pressing, sintering and automated induction brazing, and the factory side of that capability is described on the factory page.
FAQ
When is a custom carbide stud worth developing?
When the standard parts available cannot meet the grip or the wear requirement of a defined application, and when the volume justifies tooling. Where an existing profile meets the requirement, adapting the mounting or the pattern is faster and cheaper than developing a new tip.
How many prototypes are usually needed?
Two rounds covers most programmes: one to establish that the geometry and grade work at all, and one to confirm that the changes made after the first round fixed the specific problem that was found. A third round usually means the requirement was not defined precisely enough at the start.
What does a custom stud programme cost to start?
The cost sits in tooling, in the sampling rounds and in the testing, and it is quoted against a drawing rather than given as a general figure. The part that buyers underestimate most often is the internal engineering time their own team has to spend defining the requirement.
If a machine or a product line needs a carbide stud or grip that does not exist yet, send the application, the grip target and the machine data and SENTHAI will set out what the development programme involves.