I Want to Use Frosted Glass Candle Jars — How Do I Test Whether the Coating Can Tolerate Repeated Heating?

Etched or coated: two finishes sold under one word

There are two broad ways to make clear glass look frosted. The first removes or roughens the surface of the glass itself, commonly with acid etching or sandblasting. Nothing is added: the frost is the glass. The second adds a layer on top: a sprayed lacquer or paint, or a ceramic-type enamel fired onto the glass. Coloured frosts can be either coloured glass that has been etched, or clear glass with a coloured coating. Etching roughens glass but adds no colour, so the colour has to be in the glass or in a layer on it.

Why it matters for candles: glass and a coating are different materials. As a generalisation, a sprayed organic coating is more likely than etched glass to change with heat. It can yellow or brown, soften and go tacky, give off a smell of its own while warm, or lose its grip and lift at an edge after repeated warming and cooling. Fired enamels are generally tougher than sprayed lacquers, but "generally" is not a test result. An etched surface has no layer to lift. It can still stain with soot, and a rough surface can hold soot more stubbornly than smooth glass.

A candle does not heat the jar once. It heats it every session, lets it cool, and heats it again, and the hot zone moves down the glass as the wax level drops. The upper walls sit in the path of rising warm air every time. That is why a finish can look perfect after the first burn and show a faint yellow band by the fifth, and why this guide is built on cycles rather than a single burn.

This guide covers frosted finishes and repeated heat only. For painted, printed and other decorative coatings in general, see the painted and coated glass guide; for choosing frosted jars for their looks and hiding wax imperfections, see the frosted-jar buying guide. Whether the glass itself can take a burning candle is the job of the glass-suitability hub.

What CSI's frosted and matte pages actually say

We would rather you read our own listings critically than assume more than they say. Here is every CSI listing with a frosted, matte-coated or velvet finish that is relevant to this test, with its heat and finish wording quoted as it appears on the page on 26 September 2026.

Three things follow. First, none of these pages states a temperature. The seven CSI listings that do say "Safe for hot-pour up to 85°C" are the 150 g apothecary-style lidded jars and the Apothecary Glass Candle Jar with Dome Lid, all clear or coloured glass, not frosted. That figure belongs to those jars and does not carry over to any frosted one. Second, "withstand high heat" is the page's claim about the jar; it is not a statement about a finish, and CSI publishes no thermal-shock, burn-down or finish-durability test data for these jars. Third, the only CSI jar pages that name a coating are the six 70 g votive listings, which say "external coating". The 200 ml frosted pages are silent on it.

Six bench clues to etched vs coated, none conclusive

Before you pour anything, spend ten minutes with jar 2, your bench-test jar, a torch and a phone camera. If the clues point to a coating, you test exactly as described below. If they point to etching, you still run the full test; you simply expect the adhesion checks to be uneventful and watch soot staining more closely.

1. Inside, outside or both

Run a clean, dry fingertip around the inside wall, then the outside. Many sprayed finishes are applied to the outside only; etching can be on either face. On a coloured frost, look down through the mouth at the inside wall: if the inside is glossy and the colour sits behind it, the colour is a surface layer on the outside.

2. The edges: rim, base ring and any cutout

Hold the torch low across the rim and the base. A coating often shows a step, a slightly thicker edge, overspray, or a clear patch where the jar was held during spraying. An etched surface usually runs to the edge with no step. On the velvet heart jar, the cutout gives you a cut edge to examine.

3. Texture under magnification

Photograph the surface with your phone's macro or zoom. Sprayed coatings can show a faint orange-peel texture, runs or dust specks trapped in the layer; etched glass tends to look uniformly granular.

4. Tape on the base

On the underside of jar 2 only, press a strip of strong clear tape firmly onto the frost, wait a minute and rip it off sharply. Flecks of finish on the tape mean a layer that is already weak before any heat. Etching cannot come off on tape.

5. Fingernail and coin, on the base

Still on jar 2's base, drag a fingernail and then a coin edge with moderate pressure. A coating may score and show glossy glass underneath; etched frost does not reveal a different layer. Do this only on the sacrificial jar, never on a jar you will sell or burn.

6. A drop of fragrance oil

Put one drop of the oil you will use on the frosted base of jar 2, leave it for 24 hours and wipe it off. Some coatings mark, dull or soften under fragrance oil. This matters because oil drips down the outside of jars during pouring more often than makers like to admit.

The grading sheet: five things to score after every session

Coating changes are gradual, and memory is a poor reference. That is why jar 1 exists. After each graded session, let the candle go out, let the glass cool fully to room temperature, then put the jar beside the control under the same light and score five things. Never touch or scratch a hot jar to "check the finish"; all feel and adhesion checks happen cold.

The pass line is simple: no 2 on any jar for a reason connected to the finish, and no 1 that grows from one graded session to the next. A single 1 that stays a 1 is a judgement call you make with the customer in mind; a 1 that becomes a slightly bigger 1 at every check is heading for a 2 in someone's living room.

Separate finish problems from candle problems before you score. Black soot on the upper inside wall usually means a flame that is too big or a wick that needs trimming, not a coating fault; so does a hot spot next to a wick that has drifted to one side. Those are wick issues for the diameter-and-wick hub, and they are exactly why jars 8 to 10 are reserves: re-run the jar with the wick fixed and keep grading the finish.

The 10-jar repeated-heating test, step by step

Before you start, have three things settled: the wax you will sell, the fragrance load you will sell at, and a wick you have already approved for this jar. The coating test is not the place to find a wick, because a wick that is too large puts more heat into the glass and would fail a finish that is fine in normal use. If you have no approved wick yet, bracket one first; the reserves in this test are not enough for both jobs.

Count the materials before you start so the test does not stall. Six pours: one unscented 150 g candle and five scented 150 g candles at 8% need 840 g of wax and 60 g of oil. A 1 kg bag of CSI 464 leaves 160 g, enough for one reserve re-run; a 100 g bottle of oil leaves about 38 g after the pours and the drips, enough for three re-runs of oil but not of wax. The number of sessions to reach the stop line is whatever your candle needs; do not shorten the test to hit a number.

Four heat events besides the flame

The burn is the main event, but a frosted finish meets heat four other times in its life. Your test should include each one you actually use.

1. The pour

Hot wax touches the inside wall, and on a jar frosted inside that is the finish itself. None of CSI's frosted pages gives a temperature, so there is no jar figure to respect; stay at the bottom of your wax page's range. For CSI 464 that is 80°C of its stated 80°C – 90°C. CSI 468 states 58°C – 60°C, which puts less heat against the glass at pouring; that is a general tendency, not a promise about any finish. Warm clean jars evenly before pouring rather than pouring into cold glass (the cooling side of this is covered in why jars crack during cooling).

2. Drips down the outside

Fragrance oil and hot wax run down the outside of jars in every workshop. Jar 7 exists to test exactly this: a finish that marks where oil ran, or dulls where you wiped hot wax off, will show it on every candle you make at production speed.

3. A heat gun on the tops

Many makers smooth candle tops with a heat gun. CSI's Hot Air Gun page says its temperature "can not be adjusted" and reaches "about 200°c/390°f". Point it at the wax surface, keep it moving, and keep it off the frosted rim and walls; concentrated heat on one spot of glass is a thermal-stress risk as well as a finish risk. If you use a gun in production, use it on jars 4 to 7 in the test the same way.

4. Cleaning for reuse

Customers who reuse jars soak them in hot water, scrape out wax and scrub. Jar 4's second life covers your own refill cleaning. If you tell customers the jar can be reused, clean jar 4 the way you would tell them to, and grade the finish afterwards before you refill it.

Frosted and coated jars to test, and the test wax

Ranked by how directly each one answers the frosted-finish question, then by price and stock depth on 26 September 2026. We have no finish-durability data for any of them; each card says what its page states and what the test must find out.

What the test costs, and fitting it in before Diwali

Jasmine Fragrance Oil stands in here for whatever oil you sell; its page states "Recommended load up to 10% soy", so 8% sits inside both its figure and CSI 464's 10% ceiling. The six planned pours use ₹426.22 of wax and ₹258.00 of oil at those pack prices; one test candle costs about ₹201.86 in materials (₹70.80 jar, ₹70.02 wax, ₹51.60 oil, ₹9.44 wick). All figures exclude labour, freight, packaging and GST.

Now the other side. Worked example: pour 100 candles in untested frosted jars at 8% with CSI 464 at the 10 kg price, Jasmine at the 1 kg price and C2 wicks, and the materials come to ₹19,733.72. If the finish yellows by the fifth burn, every one of those candles carries the fault, and the cost of finding out that way is the candles plus your customers' trust.

Frequently asked questions