Slide 003: Annealing Cycles and Troubleshooting Hot Glass Failures

Slide Visual

Slide Overview

This slide provides deep coverage of annealing cycle design for different glass thicknesses and types, and systematically addresses the most common hot glass failures: devitrification, thermal shock, incomplete fusing, bubble entrapment, and kiln shelf adhesion. Students learn to diagnose failures by visual evidence and adjust their firing schedules to prevent recurrence.

Instruction Notes

Annealing is the difference between a finished piece that lasts indefinitely and one that self-destructs without warning. Every hot glass piece -- whether flameworked, kiln-fused, or slumped -- must be annealed. The physics is straightforward: during heating and cooling, the surface and interior of the glass reach different temperatures because glass is a poor thermal conductor (approximately 1 W/m*K). This temperature differential creates internal stress. If the stress exceeds the glass's tensile strength (approximately 7,000 psi for soda-lime glass), it cracks. Annealing holds the glass at a temperature where atoms are mobile enough to rearrange and equalize stress, then cools slowly enough that new stress does not form.

Annealing Schedules by Thickness

The annealing schedule must be adjusted for piece thickness. The critical variables are the soak time at the annealing point and the cooling rate through the critical range (annealing point to strain point).

Soda-Lime Glass (COE 90, Bullseye-type):

Thickness	Annealing Soak (960F)	Critical Cool Rate	Critical Cool Time (960F to 800F)	Total Anneal Time
6mm (1/4")	30 min	100F/hr	1.6 hr	~2 hr
12mm (1/2")	60 min	50F/hr	3.2 hr	~4.5 hr
19mm (3/4")	120 min	30F/hr	5.3 hr	~8 hr
25mm (1")	180 min	25F/hr	6.4 hr	~10 hr
50mm (2")	600 min (10 hr)	10F/hr	16 hr	~28 hr

The relationship is roughly quadratic -- doubling the thickness quadruples the required time. This is because heat must conduct from the surface to the center, and the time for heat conduction scales with the square of the distance. A 50mm casting requires nearly 28 hours of annealing -- this is not an error; it is physics.

Borosilicate Glass (COE 33, Pyrex-type / flamework):

Thickness	Annealing Soak (1050F)	Critical Cool Rate	Notes
Small beads (<12mm)	20 min	150F/hr	Standard bead anneal
Medium beads (12-25mm)	45 min	100F/hr	Marble-size
Sculpture (25-50mm)	120 min	50F/hr	Requires careful scheduling

Borosilicate requires less annealing time relative to soda-lime because its lower CTE (3.3 vs. 9.0) generates less stress for the same temperature differential. However, the annealing point is higher (1050F vs. 960F), so the kiln must reach a higher temperature.

Verifying Annealing Success

Polarized light examination is the only reliable method to verify that annealing was successful. A polariscope (two crossed polarizing filters with a light source) reveals stress birefringence -- internal stress causes the glass to rotate the plane of polarized light, producing colored fringes visible between crossed polarizers.

• No fringes: stress-free, properly annealed
• Faint, uniform color: minor residual stress, acceptable for most applications
• Bright, multi-colored fringes: significant stress, piece should be re-annealed
• Concentrated fringes at edges or joints: localized stress, high fracture risk

Students should examine every piece under polarized light before considering it finished. A simple polariscope can be built from two polarizing films and a flashlight for under $20.

Failure Mode Diagnosis

Devitrification: - Appearance: White, hazy, or scummy surface coating, sometimes with a crystalline texture - Cause: Glass spent too much time in the devitrification temperature range (approximately 1100-1300F for soda-lime). Also caused by contamination on the glass surface (fingerprints, dust, old kiln wash, fabric fibers) - Prevention: Ramp quickly through the devitrification zone during heating (AFAP from 1350F to peak); crash cool through it after peak temperature. Clean all glass with isopropyl alcohol before loading. Apply devitrification spray (a boron-based overglaze like Bullseye Spray Shelf Primer or commercial "devit spray") for additional protection. - Recovery: Light devitrification can sometimes be removed by re-firing to full fuse temperature. Heavy devitrification is usually permanent.

Thermal Shock Fractures: - Appearance: Clean, often curved cracks, frequently running from an edge or corner. Crack surfaces are smooth and glassy (as opposed to mechanical break surfaces which are rough). - Cause: Ramp rate too fast for the piece thickness, or kiln opened before the piece was cool enough. Most common during the initial heating ramp (segment 1) or during final cooling (kiln opened above 200F). - Prevention: Slower initial ramp rates for thick pieces (reduce from 300F/hr to 150F/hr for pieces over 12mm). Never open the kiln above 200F. For large or thick pieces, do not open until room temperature. - Diagnostic clue: If the crack runs from an edge, the edge heated or cooled faster than the center. If from a corner, the corner concentrated stress.

Incomplete Fusing: - Appearance: Visible seams between pieces, unfused edges, or pieces that remain distinct when full fuse was intended. Surface may show partial glossing with matte areas. - Cause: Insufficient peak temperature, insufficient hold time at peak, or CTE-incompatible glass (pieces pull apart during cooling as one contracts more). - Prevention: Verify temperature with a shelf-level thermocouple. Extend hold time at peak by 5-10 minutes. Test all glass for CTE compatibility before committing to a full project. - Diagnostic clue: If pieces separated cleanly, suspect CTE mismatch. If partially fused with matte areas, suspect low temperature or short hold.

Bubble Entrapment: - Appearance: Visible air pockets between layers, ranging from tiny seed bubbles to large voids - Cause: Air trapped between glass layers could not escape before the glass sealed at the edges during heating - Prevention: Include a bubble squeeze hold at 1350F for 20-30 minutes. Also consider stacking orientation: avoid trapping air in concave surfaces (place concave pieces cavity-down so air can escape upward). Avoid large flat contact areas between layers -- small air channels at the edges help air escape. - Diagnostic clue: Bubbles concentrated at center = air trapped by edge-first sealing. Bubbles along edges = kiln wash contamination or moisture.

Kiln Shelf Adhesion: - Appearance: Glass permanently bonded to the kiln shelf. Attempting to remove it breaks the piece and damages the shelf surface. - Cause: Missing or inadequate kiln wash; kiln wash not fully dried; glass flowed beyond the expected footprint and onto bare shelf. - Prevention: Always apply kiln wash (three crossed layers, fully dried). Use fiber paper as belt-and-suspenders backup for critical pieces. Leave 1" margin between glass edge and shelf edge. If glass is expected to spread (full fuse of a single layer), position it with extra margin.

Key Talking Points

1. Annealing time scales quadratically with thickness -- 2x thickness = 4x the time
2. Soda-lime annealing point: 960F, strain point: ~800F -- critical cooling between them
3. Borosilicate annealing point: 1050F, strain point: 950F -- same principles, different temps
4. Polarized light examination is the ONLY way to verify annealing success -- visual inspection is insufficient
5. Devitrification: white haze from too long in 1100-1300F range -- crash cool through it; clean glass before loading
6. Thermal shock: clean cracks from too-fast ramp or opening kiln too early -- slow down, be patient
7. Incomplete fuse: insufficient temperature, time, or CTE incompatibility -- verify with thermocouple and compatibility test
8. Bubble entrapment: add bubble squeeze hold at 1350F; check stacking orientation
9. Kiln shelf adhesion: kiln wash is non-negotiable; allow for glass spread at full fuse

Learning Objectives (Concept Check)

• [ ] Can the student calculate appropriate annealing soak time and cooling rate for a given piece thickness?
• [ ] Can the student diagnose a hot glass failure from visual evidence and identify the firing schedule error?
• [ ] Can the student modify a firing schedule to prevent devitrification while maintaining proper annealing?
• [ ] Can the student explain the quadratic relationship between thickness and annealing time?
• [ ] Can the student use polarized light examination to assess residual stress in a finished piece?

Adaptations for Different Learning Styles

Visual Learners

• Failure mode photo gallery: 8-10 photographs showing each failure type with diagnostic labels
• Annealing time vs. thickness graph: plot the quadratic curve so students can visualize why thick pieces take so much longer
• Polarized light demonstration: show a stressed piece and a properly annealed piece side by side under crossed polarizers
• Diagnostic flowchart: "Piece has white haze?" -> "Check devitrification zone ramp rate" / "Piece has clean cracks?" -> "Check initial ramp rate and kiln opening temperature"

Kinesthetic Learners

• Polariscope construction exercise: students build a simple polariscope from two polarizing films and a flashlight, then examine sample pieces
• Firing schedule modification exercise: given a failed piece and its schedule, students identify the error and write a corrected schedule
• Hands-on failure examination: provide sample failed pieces (devitrified, cracked, bubbled) for students to examine and diagnose
• Thickness measurement exercise: students measure their project pieces with calipers and calculate required annealing parameters

Auditory Learners

• "Failure story" format: present each failure mode as a narrative. "A student made a beautiful fused platter. Three days later, it cracked in half on the shelf. Let's figure out why..."
• Group troubleshooting discussion: show a failed piece, let students discuss possible causes before revealing the answer
• Verbal walkthrough of the diagnostic process: "First I look at the crack pattern. Clean curves = thermal. Rough surfaces = mechanical. Now I check the schedule..."

Reading/Writing Learners

• Failure diagnosis worksheet: given a photo and description of a failure, students write a diagnosis and corrected schedule
• Annealing schedule reference card with formulas for calculating soak time and cool rate based on thickness
• Written comparison: "Describe the difference between thermal shock and delayed stress failure. How do you distinguish them visually?"

Standards and References

ASTM C336 - Standard Test Method for Annealing Point and Strain Point: - Defines the precise viscosity values at which annealing and strain points occur - Annealing point: viscosity = 10^13.4 poise; Strain point: viscosity = 10^14.5 poise

ASTM C148 - Polariscopic Examination of Glass: - Standard methodology for using polarized light to detect and quantify residual stress - Provides the basis for the polariscope examination technique taught in this slide

ASTM C1279 - Standard Test Method for Non-Contact Stress Measurement: - Advanced methodology for quantitative stress measurement in flat glass - Referenced for understanding that stress measurement is a established industrial practice, not just an academic exercise

Bullseye Glass TipSheet 1 - Annealing Thick Slabs: - Industry-standard reference for annealing thick glass (over 6mm) - Provides specific schedules for thicknesses up to 75mm - Recommended as a student reference handout

Session Details

• Time Allocation: 35 minutes (15 min lecture + 10 min polariscope demo + 10 min failure diagnosis exercise)
• Breakpoints for Discussion:
• After thickness table: "Your project is a 25mm thick cast bowl. How many hours of annealing does it need?" (Answer: approximately 10 hours -- students are always surprised by this number)
• After polariscope demo: "This piece looks perfect to the naked eye. Now look through the polarizer... what do you see?" (Answer: bright colored fringes indicating significant residual stress)
• After devitrification: "Your piece has white haze on top but the bottom is clear. What does this tell you about the cause?" (Answer: top surface was contaminated -- fingerprints, dust, or the piece was placed face-up where debris could settle)
• After failure diagnosis: "You have two failures: one cracked, one devitrified. Both used the same schedule. What's different?" (Answer: the cracked piece was thicker or positioned closer to the kiln wall; the devitrified piece had contamination or spent too long in the devit zone)

Discussion Prompts

1. Engineering Trade-off: "A production glass studio fires 200 pieces per week. Each piece needs 4 hours of annealing. How do they manage kiln time? What shortcuts are tempting, and which ones are acceptable?"
2. Failure Analysis: "Examine these three failed pieces. For each one: identify the failure mode, suggest the likely cause, and write a corrected firing schedule."
3. Quality Control: "Should you polariscope-check every piece, or only random samples? What's the cost-benefit of each approach?"
4. Material Science: "Why does the annealing time scale quadratically with thickness? Connect this to what you know about heat conduction."

Instructor Notes

• The polariscope demonstration is the highest-impact teaching moment in this slide. Students are universally surprised that a "perfect-looking" piece can have dangerous internal stress. Build a simple polariscope and have it available for every class session from this point forward.
• Keep a collection of failed pieces organized by failure mode -- these are more valuable teaching tools than any slide deck. Label each with the failure type, probable cause, and the schedule that produced it.
• Common student reaction to the annealing time table: disbelief. "Ten HOURS for a 1-inch piece?" Yes. Glass is a poor thermal conductor. The physics is non-negotiable.
• Emphasize that re-annealing is always an option. If a piece has residual stress, it can be placed back in the kiln and annealed again. The piece is not ruined -- it just needs more time.
• SAFETY CALLOUT: A piece with significant residual stress can fracture without warning. Pieces that have not been polariscope-checked should be handled with cut-resistant gloves and eye protection. When a stressed piece breaks, fragments can fly several feet.
• The diagnostic flowchart should be posted in the kiln area as a permanent reference

Common Misconceptions Corrected

• Myth: "If it survived the first 24 hours, it's properly annealed." Reality: Delayed stress failures can occur weeks or months after firing. The residual stress does not resolve on its own -- it persists until either relieved by re-annealing or released by fracture.
• Myth: "Thicker pieces just need a longer soak -- the cooling rate can stay the same." Reality: Both the soak time AND the cooling rate must be adjusted for thickness. A thick piece cooled at a fast rate will develop stress even after a proper soak, because the fast cooling re-introduces temperature differentials that create new stress.
• Myth: "Devitrification is caused by too much heat." Reality: Devitrification is caused by too much TIME at moderate temperatures (1100-1300F). It is actually prevented by higher peak temperatures (which dissolve surface crystals) and rapid transit through the devitrification zone.
• Myth: "Bubbles are caused by the glass being too hot." Reality: Bubbles are caused by trapped air that cannot escape. They form at moderate temperatures (when the glass seals before air escapes) and are prevented by the bubble squeeze hold, not by reducing peak temperature.
• Myth: "My kiln controller reads 960F, so the glass is at 960F." Reality: The thermocouple measures air temperature near the kiln wall. Glass at the center of the kiln can be 20-50F cooler. Use witness cones or shelf-level thermocouples for verification.

Accommodations for Neurodiversity

ADHD Support

• The failure diagnosis exercise is inherently engaging -- it's detective work with physical evidence
• Provide a diagnostic decision tree card (laminated, at each station) rather than expecting students to remember the flowchart
• Break the lecture into two segments: annealing theory (8 min) then failure modes (7 min) with the polariscope demo as the transition activity

Autism Spectrum Support

• The quadratic relationship between thickness and time is mathematically precise and predictable -- present the formula explicitly: soak_time = base_time * (thickness / base_thickness)^2
• Diagnostic process is rule-based: "If crack is clean and curved = thermal shock. If surface is hazy = devitrification. If pieces separated = CTE mismatch."
• Provide the annealing schedule table as a permanent reference -- it eliminates ambiguity in schedule design

Dyslexia Support

• Failure mode identification uses photographs with color-coded labels (red = thermal shock, yellow = devitrification, blue = bubbles, green = incomplete fuse)
• Annealing schedule table uses large, clear formatting with high contrast
• Diagnostic flowchart uses arrows and icons rather than blocks of text

Sensory Processing Support

• Polariscope produces colored light patterns -- some students may find the visual patterns disorienting. Allow them to look briefly and then look away
• Failed pieces with sharp fracture edges should be handled carefully and with warning
• If demonstrating stress fracture (breaking a stressed piece intentionally), warn about the sound and flying fragments

---

Last Updated: 2026-03-19 Content Review: Q1 2026