Should Candle Jar Supplier Changes Trigger a Small Production Test Even if Dimensions Remain Identical?

Should a supplier change trigger a test when the dimensions are identical? Yes — a small one, every time

A jar specification usually covers outside height and diameter, neck finish, capacity and perhaps weight. It rarely covers what matters to adhesion: wall and base thickness distribution, the inner surface, the type of glass, residues from the production line, and how the jars were packed and handled before they reached you. Two suppliers — or one supplier's two factories — can meet the same drawing and still give your wax a different surface to grip and a different mass of glass to warm. Commonly cited contributors to wet spots include exactly those differences. So a change of supplier is a change of material, even when the drawing is unchanged.

That does not mean every delivery needs a full validation. It means you need a standing rule: which kinds of change trigger which size of test, written once and followed without debate. A cheap measurement check on every delivery. A small production test — a dozen or so jars, poured beside your old jars from one melt — whenever the supplier, factory, mould or glass changes. A full validation only when the small test fails or the shape or capacity changes. The rule's value is that nobody has to decide, under festive deadline pressure, whether this particular "identical" jar is worth a test.

The case for the rule is arithmetic. In this page's example, a 1,000-jar order of CSI's clear 200 ml Salsa Glass Jar filled with Luxury Soy Wax Chunks and Orchard Pear at 8% (an illustrative 160 g fill — 200 ml is the brim volume and the fill weight is not published) is about ₹1,71,419.20 of materials. If an example 20% of those candles spotted, ₹34,283.84 of materials would sit in jars you must hide, discount or re-pour. The small production test costs ₹884.20 in CSI materials with your own wax and oil. It pays for itself if there is even a 2.6% chance it catches a problem of that size.

Tier 1: the delivery check every carton gets

Tier 1 costs nothing but ten minutes. Take five jars from different cartons in the delivery — not five from the top of one box — and weigh each empty on your candle scale. Measure the outside diameter and height, and the inside diameter at the rim with calipers or a ruler. Look inside each jar against the light for dust, film, smears or packing fibres, and note how the jars were packed: loose, in dividers, shrink-wrapped, with paper. Compare every figure with the first approved delivery's figures, which you recorded when the jar was approved.

What counts as a meaningful difference is your house choice. A working rule you might adopt: if the average empty weight differs from the approved delivery by more than a few per cent, or any jar shows visible residue, the delivery moves up to tier 2 before use. Neither CSI nor any industry body publishes a tolerance for this; the point is that your own numbers are recorded, so a drift is visible. The jar measurement sheet lists every measurement worth taking, and whether new jars should be cleaned covers what to do about residue.

Tier 2: the small production test

Tier 2 asks two questions with the smallest sensible number of jars: does your wax grip the new glass as well as the old, and does the candle burn the same in it? Pour 15 jars from one melt of your production wax lot: three new jars as received, three new jars after a controlled isopropyl-alcohol clean, three old jars (or gauges) for adhesion; then three new and three old for a burn comparison. Draw the new jars from at least three different cartons, so you are testing the delivery, not one lucky box. Use one measured pour temperature that every jar allows: for Chunks, CSI's product page states fragrance addition at 80–90°C and pour at 75–80°C in its spec table (its step list also shows 70–80°C); the Salsa jar's page publishes no hot-pour limit, so check the supplier's guidance.

If the cleaned new jars pass and the as-received ones do not, release the jar with a fixed cleaning step — cleaning commercially without residue sets one up. If the new jars spot even when cleaned, or burn differently, the delivery goes to tier 3: the full container validation, with its retest matrix and five gates. The temperature-cycle checkpoint should use your own logged route; an example cycle is 8 hours warm and 16 hours at room temperature, three to five times, kept well below CSI 468's stated 50–53°C melting point. The temperature-cycle test covers it.

The money: a test against a lot

Here is the arithmetic in full. At production rates — Chunks at the 5 kg rate, Orchard Pear at the 1 kg rate, the Salsa jar at its pack-of-100 rate of ₹29.50 and an Eco Thin C1 at ₹5.90 as the example wick line — one 160 g candle costs ₹171.42 in materials. The spotted share and the order size are examples; your own figures replace them.

The break-even line is the heart of the policy. You do not need to know how likely a new supplier is to cause spots — nobody publishes that figure, and we will not invent one. You only need to believe the chance is more than about 2.6%. For most makers, a new supplier, factory or mould clears that bar easily, which is why the rule says "test every time". The calculation also shows when a test is not worth its cost: a repeat order of a jar you have used for a year from the same supplier is tier 1, not tier 2. The cost of cosmetic rejects and monitoring future wax and jar lots carry this thinking into a full QC system.

The priced tier 2 kit

The example assumes the new jar is CSI's clear Salsa jar and the production candle is Chunks with Orchard Pear at 8%. If your new jar is from another supplier, drop the jar line and use their jars. Grams follow CSI's convention (oil = W × L) with a 5% pour-loss working allowance on an illustrative 160 g fill.

In this example, two packs of 5 Salsa jars cost ₹40.12 a jar; the pack of 100 is ₹29.50 a jar. That is the honest trade: you pay a little more per jar for the test quantity than for production, which is a good reason to run the test before the big order, not after. CSI publishes no volume discount beyond the listed packs; larger quantities are quoted on WhatsApp.

Release, and the first production batch

A tier 2 pass is not the end. The first production batch in the new jar carries 10 tagged jars, drawn from across the batch — first poured, last poured, outer and centre positions on the cooling table — photographed on four sides at 7 days before the rest of the batch ships. This catches the things a fifteen-jar test cannot: a batch-scale cooling pattern, a carton-to-carton difference, a team handling the new jars differently. Outer versus centre jars in a 100-candle batch explains why position matters at scale.

Then record the approval where the next person will find it: the jar supplier and factory, the delivery date and your tier 1 figures, the tier 2 sheet, the pour temperature and any cleaning step now in use, the tagged-jar result, and the date and signature. When the supplier changes anything again — even packing — the rule tells you which tier applies. Locking jar preparation, load, pour temperature and cooling shows how to fold this into a full production standard.

The decision rule

Apply after tier 1, after tier 2 and after the tagged jars. Stop at the first line that is true.

Nowhere in the rule does a jar change lead straight to a wax change. A wax change re-opens wick, load, cure, throw and adhesion for every SKU; a jar change re-opens adhesion, wick, pour routine and route behaviour for the SKUs in that jar. Keep the response the size of the change. Fixing one SKU versus the whole collection explains why scale matters so much.

The CSI products in this example

Where to go next

This post owns the standing rule for jar changes. The validation post owns tier 3; the glass-comparison post owns diagnosing two jars that differ; thicker glass and same capacity, different shape own the two most common physical differences. The section hub is the place to start if your spots began with a fragrance or dye change instead.

Frequently asked questions