Appearance
A manufacturer of ceramic roller kilns produced recrystallised SiC kiln rollers (60 mm OD x 3000 mm length, requiring 25 mm axial bore for cooling air at 1250 C) using a diamond-plated core drill (25 mm, metal-bond, 60/80 mesh, Vc = 5 m/min, feed 0.5 mm/min, DI water at 3 bar, 100 mm peck). The abrasive RSiC (hardness > 2000 HV) could not be drilled by any carbide or HSS tool. Bore straightness verified by 24.5 mm test rod. Roller tested at 1250 C for 72 hours with zero measurable deformation.
Diamond Core Drilling Methods for Ceramic Kiln Rollers
Comparison of Drilling Methods for Ceramic Roller Materials
| Drilling Method | Material | Hole Ø Range (mm) | Max Depth (mm) | Surface Finish Ra (µm) | Straightness (mm/m) | Tool Wear Rate | Relative Drilling Time (per metre) | Coolant Required | Recommended Grit/Bond |
|---|---|---|---|---|---|---|---|---|---|
| Diamond core drilling (metal-bond) | Recrystallised SiC (RSiC) | 10–60 | 4000+ | 1.5–4.0 | 0.2–0.5 | Low — metal-bond diamond provides consistent wear | 1× (baseline, 2–4 h/m) | Deionised water, 2–5 bar | 40/60–80/100 mesh, cobalt bond |
| Diamond core drilling (resin-bond) | Mullite (3Al₂O₃·2SiO₂) | 10–50 | 3000+ | 0.8–2.0 | 0.3–0.8 | Moderate — resin bond wears faster but provides smoother finish | 1.2–1.5× | Deionised water, 2–4 bar | 100/120–200/230 mesh, phenolic bond |
| Rotary ultrasonic drilling | RSiC, mullite, fused silica | 5–40 | 500 | 0.3–0.8 | 0.1–0.3 | Very low — ultrasonic vibration reduces tool pressure | 2–3× | Water or water-based emulsion, 3–5 bar | 80/100–140/170 mesh, metal bond |
| Conventional carbide twist drilling | NOT suitable for RSiC or mullite | N/A | N/A | N/A | N/A | Extreme — carbide drill fails within 1–5 mm in RSiC | N/A | N/A | N/A |
| Laser drilling (CO₂ or Nd:YAG) | RSiC, mullite (thin sections) | 0.5–5 | 50 | 3.0–8.0 | N/A | N/A (non-contact) | 0.1–0.3× (per mm for thin sections) | None (dry) | N/A |
Recommended Parameters for Diamond Core Drilling of Ceramic Kiln Rollers
| Material | Hardness (HV) | Diamond Grit (mesh) | Bond Type | Concentration | Drill OD (mm) | Wall Thickness (mm) | Spindle Speed (rpm) | Cutting Speed Vc (m/min) | Feed Rate (mm/min) | Peck Depth (mm) | Coolant Type | Coolant Pressure (bar) | Expected Drill Life (cumulative mm) |
|---|---|---|---|---|---|---|---|---|---|---|---|---|---|
| Recrystallised SiC (RSiC) | 2200–2800 | 60/80 (coarse) | Metal bond (cobalt) | 100–125 | 20–40 | 2–4 | 100–300 | 4–8 | 0.3–1.0 | 50–150 | Deionised water | 3–5 | 1000–5000 |
| Mullite (3Al₂O₃·2SiO₂) | 1200–1600 | 100/120 (medium) | Metal bond (bronze) | 100 | 20–40 | 2–5 | 200–500 | 6–12 | 0.5–2.0 | 100–200 | Deionised water | 2–4 | 2000–8000 |
| Fused silica (SiO₂) | 600–700 | 200/230 (fine) | Resin bond | 75–100 | 10–30 | 2–4 | 500–1500 | 10–20 | 1.0–3.0 | 50–100 | Deionised water | 2–4 | 5000–20000 |
| Siliconised SiC (SiSiC) | 2500–3000 | 40/60 (coarse) | Metal bond (cobalt) | 125 | 20–50 | 3–5 | 80–200 | 3–6 | 0.2–0.6 | 50–100 | Deionised water | 4–6 | 500–2000 |
| Alumina (Al₂O₃, 99%) | 1500–1700 | 80/100 (medium) | Metal bond (cobalt) | 100–125 | 10–40 | 2–4 | 300–600 | 8–15 | 1.0–3.0 | 100–150 | Deionised water | 3–5 | 3000–10000 |
Glass Tempering Furnace Components
Comparison of Drilling Methods for Glass Tempering Furnace Roller Shafts
| Shaft Material | Max Operating Temp (°C) | Shaft Ø (mm) | Bore Ø (mm) | Bore Depth (mm) | Drilling Method | Vc (m/min) | f (mm/rev) | Coolant | Tool Material | Straightness (mm/m) | Expected Tool Life (cumulative metres) |
|---|---|---|---|---|---|---|---|---|---|---|---|
| AISI 310S (25Cr-20Ni) | 700 | 50–100 | 15–30 | 2000–4000 | Gun drilling | 40–60 | 0.03–0.06 | Sulphurised oil, 40–60 bar | Carbide K10 with TiAlN | < 0.1 | 30–80 |
| Inconel 600 | 700 | 50–80 | 15–25 | 2000–3000 | Gun drilling | 20–30 | 0.02–0.04 | High-EP oil, 60–80 bar | Carbide K20 with AlCrN | < 0.1 | 15–40 |
| AISI 321 (Ti-stabilised 18/8) | 600 | 60–100 | 20–30 | 2000–4000 | Gun drilling | 50–70 | 0.04–0.07 | Sulphurised oil, 40–60 bar | Carbide K10 with TiAlN | < 0.1 | 40–100 |
| Cast heat-resistant alloy (HK-40) | 700 | 80–150 | 30–50 | 2000–4000 | BTA drilling (2-pass) | 30–50 | 0.08–0.15 | Sulphurised oil, 30–50 bar | Carbide BTA head, TiAlN | < 0.15 | 20–60 |
Comparison of Ceramic and Glass Industry Components Requiring Deep Hole Drilling
| Component | Typical Material | Bore Function | Bore Ø (mm) | Bore Length (mm) | Depth-to-Diameter Ratio | Surface Finish Ra (µm) | Tolerances | Industry Standard | Typical Production Volume |
|---|---|---|---|---|---|---|---|---|---|
| Roller kiln roller | RSiC, mullite | Cooling air circulation | 20–40 | 2000–4000 | 50:1–150:1 | 1.5–4.0 (as-drilled) | Straightness < 0.5 mm/m | ISO 13006 (ceramic tile) | 500–5000 rollers/year per kiln line |
| Glass tempering furnace roller shaft | 310S, Inconel 600 | Cooling air passage | 15–50 | 2000–4000 | 50:1–200:1 | 0.8–1.6 | Straightness < 0.1 mm/m | EN 12150 (heat-soaked tempered glass) | 100–500 shafts/year per furnace line |
| Glass lehr conveyor roller | AISI 321, 304L | Even heat distribution | 15–30 | 3000–6000 | 100:1–300:1 | 0.8–1.6 | Runout < 0.2 mm | ANSI Z97.1 (safety glazing) | 200–1000 rollers/year |
| Ceramic tile press die | P20, H13 (32–48 HRC) | Heating/cooling fluid | 8–20 | 200–1000 | 15:1–80:1 | 0.4–0.8 (after polishing) | Position +/-0.2 mm | ISO 13006 | 10–50 dies/year per press line |
| Glass grinding machine spindle | 4140 (32–38 HRC), 17-4PH H900 | Coolant delivery | 6–15 | 300–800 | 30:1–100:1 | 0.4–0.8 | Runout < 0.01 mm | ISO 4840 (spindle accuracy) | 100–500 spindles/year |
| Glass bending furnace mold support shaft | 310S, Nimonic 80A | Cooling channel | 10–30 | 1000–3000 | 50:1–200:1 | 0.8–1.6 | Straightness < 0.1 mm/m | Automotive glass specs (ECE R43) | 50–200 shafts/year |
FAQ
What makes recrystallised silicon carbide (RSiC) so difficult to deep hole drill, and why is diamond tooling essential?
Recrystallised silicon carbide (RSiC) is one of the most difficult ceramic materials to drill by any method. Its hardness (2200–2800 HV, exceeded only by diamond, cubic boron nitride, and a few superhard materials) means that conventional carbide or HSS tools cannot cut it — a carbide drill (1500–1700 HV) is softer than the RSiC workpiece and will wear to failure within 1–5 mm of cumulative drilling. The second challenge is the material's extreme abrasiveness: even diamond core drills wear measurably when drilling RSiC (a metal-bond diamond core drill loses 0.01–0.05 mm of the bond matrix per metre of drilling, requiring periodic redressing of the diamond surface). The third challenge is the material's low fracture toughness (2–4 MPa·m¹/², similar to glass) — the combination of the drilling pressure and the brittle material causes microcracks at the bore exit if the drill is not supported by a backup plate. Diamond tooling with a metal bond is essential for drilling RSiC because: the diamond grit (6000–8000 HV) is 2.5–3× harder than the RSiC, enabling true cutting rather than rubbing; the metal bond (cobalt or bronze matrix) holds the diamond grit firmly and conducts heat away from the cutting interface; and the coarse grit (40/60 to 80/100 mesh) provides the aggressive cutting action needed to fracture and remove the hard SiC particles. Resin-bond diamond tools are not recommended for RSiC because the resin bond wears too rapidly (the abrasive SiC particles erode the resin matrix faster than the diamond grit can cut, causing the diamond to fall out unused).
The drilling parameters for RSiC are significantly different from those for metals. The cutting speed must be low (Vc = 4–8 m/min, compared to 60–120 m/min for steel drilling) because the diamond grit must fracture the SiC particles by impact rather than by continuous abrasion — higher speeds cause the diamond to slide over the RSiC surface without penetrating, generating heat that softens the metal bond and accelerates tool wear. The feed rate must be low (0.3–1.0 mm/min) and controlled by the drilling force rather than by a constant feed rate — the drill should be advanced at a force of 50–200 N (measured by a load cell), and the feed rate automatically adjusts as the drill encounters harder or softer sections of the RSiC (the material density can vary by 5–10% due to the recrystallisation process). The coolant must be deionised water (not oil) because oil-based coolants contaminate the porous RSiC (the green RSiC tube is 15–20% porous before firing, and any absorbed oil will carbonise during firing at 1250 C, causing internal defects). The peck depth should be limited to 50–150 mm because the abrasive SiC chips fill the annulus between the core drill OD and the bore wall, and the annulus must be flushed clean by the water flow before the drill can advance further.
How does the operating temperature affect the material selection for glass tempering furnace roller shafts, and how does this influence the gun drilling parameters?
The operating temperature of a glass tempering furnace (600–700°C at the roller surface, with peak transient temperatures of up to 800°C during furnace startup) requires the roller shafts to be made from heat-resistant alloys that maintain their strength and dimensional stability at these temperatures. The most common roller shaft materials are AISI 310S (25% chromium, 20% nickel, balance iron), which provides oxidation resistance up to 1050°C and creep strength adequate for the 600–700°C operating range; Inconel 600 (72% nickel, 15% chromium, 8% iron), which provides higher creep strength than 310S and better resistance to thermal fatigue from the furnace heating/cooling cycles; and AISI 321 (18% chromium, 10% nickel, titanium-stabilised), which is a lower-cost alternative for furnaces operating below 600°C. The material selection directly affects the gun drilling parameters: 310S stainless steel has a thermal conductivity of 14 W/m·K (compared to 50 W/m·K for carbon steel), which means the heat generated at the cutting edge accumulates locally and requires higher coolant flow to prevent overheating. The specific cutting energy for 310S is approximately 3.2 J/mm³ (compared to 2.0 J/mm³ for carbon steel), requiring 60% more spindle power for the same material removal rate. The recommended gun drilling parameters for 310S are: cutting speed Vc = 40–60 m/min (significantly lower than the 80–120 m/min for carbon steel), feed f = 0.03–0.06 mm/rev, and coolant pressure of 40–60 bar with a sulphurised oil coolant (EP additive concentration 1.5–2.0%). For Inconel 600 (which has an even lower thermal conductivity of 13 W/m·K and a higher work-hardening rate than 310S), the cutting speed must be further reduced to Vc = 20–30 m/min, the tool material must be upgraded to a coated grade with better thermal stability (AlCrN-coated carbide, which maintains hardness to 1100°C versus 800°C for TiAlN), and the coolant pressure must be increased to 60–80 bar to ensure chip evacuation at the lower feed rate.
The bore straightness requirement for glass tempering furnace roller shafts (typically < 0.1 mm/m) is driven by the roller's function: the roller must rotate smoothly at 2–10 rpm while supporting a 3–6 mm thick heated glass sheet. Any wobble (from a non-straight bore) causes the glass sheet to deviate from the horizontal plane as it passes through the furnace, resulting in optical distortion of the tempered glass (a defect called "roller wave" or "washboard" that is visible as a periodic variation in the glass's reflected image). The straightness is verified by a laser straightness gauge that measures the bore axis deviation relative to the shaft OD at 500 mm intervals along the full shaft length. For a 4000 mm shaft, the maximum allowable deviation is 0.4 mm (0.1 mm/m × 4 m). If the measured deviation exceeds this limit, the shaft can be straightened by a press (for steel shafts, not for ceramic rollers), or the bore must be corrected by a finish BTA reaming pass with a floating reamer head that self-centres in the bore and removes 0.1–0.3 mm per side to correct the axis deviation.
What is the typical layout of fluid channels in a ceramic tile press die, and how are the intersecting bores drilled and sealed?
The fluid channel network in a ceramic tile press die consists of three layers of drilled channels that circulate heating oil (to maintain the die at 120–180°C during the pressing cycle) and hydraulic oil (for the pressing cylinders embedded in the die). The channels are arranged in a three-dimensional grid: the primary channels (15–20 mm diameter, 500–1500 mm length) run in the X-direction (parallel to the die width), the secondary channels (10–15 mm diameter, 500–1500 mm length) run in the Y-direction (perpendicular to the primary channels), and the distribution channels (8–12 mm diameter, 100–500 mm length) connect the primary and secondary channel networks to the heating zones and to the press cylinder ports. The primary channels are gun-drilled from the side faces of the die plate (drilling through the full width or length of the plate). The secondary channels are gun-drilled from the opposite faces, intersecting the primary channels at 90-degree junctions. The distribution channels are drilled from the die face (the surface that contacts the tile) to within 3–5 mm of the primary or secondary channel, then punctured by a small-diameter drill that intersects the main channel.
The intersecting bores at the primary-secondary junctions must be deburred to prevent the burrs from blocking the fluid flow or breaking off and contaminating the oil system. The deburring is performed by one of three methods: electrochemical deburring (ECD) — a shaped cathode is inserted into the primary channel and positioned at the intersection, with a current pulse dissolving the burr in 10–30 seconds; abrasive flow machining (AFM) — a silicone-based abrasive media is forced through the entire channel network at 20–50 bar, abrading the burrs and polishing the channel walls; or manual deburring with a flexible carbide scraper inserted through the distribution channel. The outer ends of the primary and secondary channels are sealed by threaded plugs (typically 3/8" or 1/2" BSPT taper thread, sealed with PTFE tape or a thread-locking compound) or by welded plugs (a steel disc welded over the channel end and ground flush). The channel network is leak-tested after sealing by pressurising the system with hydraulic oil at 1.5× the working pressure (typically 300 bar for the hydraulic channels and 20 bar for the heating channels) and holding the pressure for 30 minutes with a maximum allowable pressure drop of 1 bar. The drilling of the channel network for a 1000 mm × 1000 mm × 400 mm thick die plate requires 15–40 channel bores with a total drilling length of 10–40 m, and the complete drilling and deburring process takes 2–5 days for a single die plate.
How is the bore straightness of glass lehr conveyor rollers verified, and what tolerance is acceptable?
The bore straightness of glass lehr conveyor rollers is verified by one of two methods, depending on the roller length and the accuracy requirement. For rollers up to 3000 mm length, a straightness gauge bar is used: a precision-ground steel bar (25 mm diameter, with a 0.01 mm/m straightness tolerance) is inserted through the full bore length. If the bar passes through without resistance, the bore straightness is considered acceptable. For rollers longer than 3000 mm, or where quantitative straightness data is required (for warranty or process control), a laser straightness gauge is used: a laser transmitter is mounted at one end of the bore, and a position-sensitive detector (PSD) is mounted at the other end. The laser beam establishes a reference line, and the detector is moved along the bore on a centring fixture, measuring the deviation at 500 mm intervals. The straightness is reported as the maximum deviation (in mm) per metre of bore length. The acceptable straightness tolerance for glass lehr conveyor rollers is < 0.2 mm/m (compared to < 0.1 mm/m for tempering furnace rollers) because the lehr conveyor operates at lower temperatures (300–400°C versus 600–700°C) and at lower speeds (1–5 rpm versus 2–10 rpm), so the roller wobble that causes glass distortion is less critical. The roller runout (the out-of-roundness of the roller OD) is also measured by rotating the roller on V-blocks and measuring the radial deviation with a dial indicator. The acceptance criterion for lehr roller runout is < 0.2 mm TIR (total indicated runout). If the runout exceeds this limit, the roller OD is ground to correct the runout after the bore is completed.
What are the main failure modes of gun-drilled cooling passages in glass tempering furnace shafts, and how is the shaft life maximised?
The main failure mode of gun-drilled cooling passages in glass tempering furnace shafts is not mechanical wear but thermal fatigue cracking of the shaft around the cooling passage. The shaft surface is heated to 600–700°C by contact with the glass and the furnace atmosphere, while the cooling air passing through the gun-drilled central bore is at 20–60°C (depending on the compressed air system). The temperature gradient across the shaft wall (from 600°C at the OD to 60°C at the bore surface) creates a steep thermal stress gradient: the OD is in compression (it tries to expand but is constrained by the cooler interior), and the bore surface is in tension (it tries to contract but is constrained by the hotter exterior). The tensile stress at the bore surface, calculated from the thermal gradient and the material's coefficient of thermal expansion, can reach 100–300 MPa for 310S stainless steel at a 600°C temperature difference. If the bore surface has any roughness, tool marks, or microcracks from the gun drilling process, the tensile stress at these stress raisers can exceed the material's fatigue strength, initiating a thermal fatigue crack that grows from the bore surface radially outward. Over 5000–20000 thermal cycles (heating up and cooling down), the crack can penetrate the full shaft wall thickness, causing the shaft to fail by leaking cooling air into the furnace or, in severe cases, by catastrophic fracture.
Shaft life is maximised by three strategies. First, the gun-drilled bore surface finish must be Ra < 1.6 µm (achieved by a final burnishing pass with a roller burnishing head that cold-works the bore surface, creating a compressive residual stress of −100 to −300 MPa that counteracts the tensile thermal stress). Second, the shaft must be stress-relieved after gun drilling by heat-treating at 750–800°C for 2–4 hours in a vacuum furnace, which reduces the residual tensile stresses from the drilling process and stabilises the microstructure for the 600–700°C operating temperature. Third, the cooling air flow must be maintained at a minimum of 500 L/min per shaft (for a 25 mm bore) to ensure that the bore surface temperature does not exceed 150°C — if the cooling air flow is interrupted for more than 5 minutes during operation, the shaft bore temperature rises rapidly, and the thermal stress can cause immediate cracking. A redundant cooling air supply with automatic switchover should be installed on any tempering furnace line. With these measures, the typical service life of a glass tempering furnace roller shaft with a gun-drilled cooling passage is 3–5 years (20 000–30 000 thermal cycles), compared to 1–2 years for a shaft with no stress relief or burnishing.
The information provided in this article is for general informational purposes only and does not constitute professional engineering advice. Always consult qualified kiln engineers, glass manufacturing specialists, and equipment manufacturers for specific applications. Data and parameter recommendations are based on published research and industry experience as of 2026.