My Candle Looks Patchy Through Clear Glass After Cooling — How Do I Determine Whether the Problem Is Wax Shrinkage or Jar Preparation?
Aktie
Can I tell shrinkage from dirty glass by looking? Not reliably from one candle. Contamination tends to give small, sharply outlined patches that repeat in the same places and preparation fixes them; contraction tends to give larger, softer-edged areas that vary jar to jar. Both overlap enough that a single photograph proves nothing.
So what does settle it? A paired split test. Twelve jars, one pot, one load, one pour temperature, one cooling spot. Six jars washed and fully dried, six poured exactly as they arrived. Score both arms at the same three checkpoints.
Does CSI publish a jar-cleaning instruction? No. No CSI container page gives a washing, wiping, drying or preheating procedure at all, and no CSI wax page publishes an adhesion figure or test. That gap is the honest reason you have to run the test yourself rather than follow a spec.
What each cause tends to look like — and why that is not enough
There are real tendencies, and they are worth knowing — as a reason to suspect something, never as a verdict. Contamination on the inner wall is a place the wax never bonded in the first place, so it tends to produce patches with definite outlines that sit where hands, dust or residue were. Contraction is a bulk behaviour of the wax as it solidifies and keeps responding to temperature, so it tends to produce broader, softer areas that differ from jar to jar even within one batch. The trouble is that a jar with a thin invisible film across most of its wall will give you broad soft patches too, and a jar cooling against a cold edge will give you a sharply bounded one. The tendencies overlap. That is why this post is about a test and not about a diagnosis.
| Feature | Could suggest preparation | Could suggest contraction | Why it is not conclusive on its own |
|---|---|---|---|
| Patch outline | Sharp, defined, sometimes with a visible boundary | Soft, map-like, fading at the edge | A cold spot in the glass gives a sharp boundary too |
| Repeatability across jars | Same places on many jars, if the handling is the same | Different places on every jar | Consistent handling and consistent cooling both repeat |
| Size distribution | Many small marks, sometimes specks | Fewer, larger areas | A thin uniform film produces large areas as well |
| Height on the wall | Wherever the contaminant was — often where fingers grip | Often lower or in a band, following how heat left | People grip jars at the same height heat leaves them |
| Timing | Usually present as soon as the wax sets | Can appear or grow over days | A delayed appearance can still start at a weak bond |
The second reason to test rather than diagnose is that there is nothing to look up. Not one CSI container page states a washing, rinsing, wiping, drying or preheating instruction — across 74 listed containers, the pages give capacities, some dimensions, a hot-pour limit on two of them, and nothing about how to prepare the glass. And on the wax side, only two pages in the entire range mention glass adhesion at all: CSI 464 ("excellent glass adhesion", "reduced frosting compared to conventional soy waxes") and CSI 468 ("formulated for excellent glass adhesion and a one-pour finish"), both as wording with no figure, no method and no threshold behind it. Luxury Soy Wax Chunks, Natural Soy Pearl Wax and Coconut Wax say nothing about it. In the absence of a published standard, the only authority is your jar.
Designing the split so the answer means something
A split test is only as good as the things you held still. The design below has one variable and nine controls, and every one of those nine is a way people accidentally ruin this experiment.
| Element | In this test | How it goes wrong |
|---|---|---|
| Jar preparation | The variable. Arm A washed and fully dried; Arm B exactly as received | Half-drying Arm A, so you are testing residual water instead of cleanliness |
| Wax batch | One bag, one pot, one melt | Topping the pot up mid-pour from a second bag |
| Fragrance oil and load | One oil, one percentage, weighed not measured | Eyeballing the oil for the last few jars |
| Pour temperature | One reading, logged at the first and last jar | A long pour where the pot drifts 10°C between jar 1 and jar 12 |
| Jar type and size | Twelve identical vessels | Using whatever is in the cupboard to make up the twelve |
| Wick | One size, centred the same way, same stickers | Mixing wick sizes "because it won't matter for adhesion" |
| Cooling location | One bench, one spacing, nothing moved for 48 hours | Overflow jars on a window ledge |
| Arm layout on the bench | Arms interleaved, not grouped — A B A B across the tray | All of Arm A on the warm side of the bench |
| Scoring | One lamp, one distance, one printed grid, one scale 0–3 | Grading Arm A in daylight and Arm B at night |
| Who scores | Ideally someone who does not know which arm is which | Scoring your own cleaned jars generously |
Three checkpoints and what each one tells you
The checkpoint is this post's unit of analysis. One look at a candle tells you how it looks today; three looks at fixed intervals tell you what is driving it. Every temperature and duration below is your choice, recorded — there is no industry protocol here and this guide does not present one.
| Checkpoint | When | What to record | Arm A better here would suggest | Both arms equal here would suggest |
|---|---|---|---|---|
| 1 · At set | 24–48 hours after the pour, lids off, nothing moved | Face-by-face score 0–3, photograph against the grid, room temperature | A genuine preparation effect — the wax never bonded where the glass was not clean | Preparation is not your lever; look at glass temperature, pour temperature and cooling |
| 2 · At one week | Seven days, lids on, same room, undisturbed | Same score, same lamp, same distance; note whether patches grew or appeared | A weak bond that has continued to let go | Ordinary contraction continuing in both arms equally |
| 3 · After one cycle | After a warm-and-cool movement you define and log yourself | Same score, plus the two temperatures and the hold times you used | Cleaned glass holding on better through movement | Temperature movement is the dominant factor, not the glass |
On checkpoint three, be deliberate and be honest about what it is. Pick a warm place and a cool place you can actually reach — a closed room in the afternoon and a cupboard in the morning, say — hold the jars in each for a period you choose, photograph at the end of each hold, and write down every temperature and duration. That gives you a repeatable comparison between two arms. What it does not give you is any way to convert the result into shelf life or into months of real time, and anyone who offers you that conversion is guessing. Treat it as "these twelve candles, this cycle, this result".
The 12-jar split test, step by step
What it costs, and what it saves
Here is the complete bill at live September 2026 prices, with every figure's working shown. The point of the table is the comparison at the bottom: this test costs a quarter of the wax bag it is designed to stop you buying.
| Line | Packs bought | Pack price | Used | Cost of what is used |
|---|---|---|---|---|
| Luxury Soy Wax Chunks | 1 kg | ₹599 | 772.8 g | ₹462.91 |
| Sandalwood Fragrance Oil | 100 gm | ₹502.68 | 67.2 g | ₹337.80 |
| Clear Glass Votive Jar 70 gram | Pack of 10 + pack of 5 | ₹477.90 | 12 jars | ₹382.32 |
| Eco Candle Wicks · Thin (C1) | 20 wicks | ₹118 | 12 wicks | ₹70.80 |
| Candle Wick Stickers | 1 sheet · 40 | ₹100 | 12 stickers | ₹30 |
| Materials consumed | 12 candles | ₹1,283.83 | ||
| Cart total | ₹1,797.58 | 3 spare jars, 8 spare wicks, leftover wax and oil | ₹106.99 per test candle | |
| The alternative: a new wax on a hunch | 10 kg | ₹5,074 to ₹6,370 | plus re-testing wick and load | 4 to 5 times the test |
There is a second saving that is easy to miss. If the test says preparation is not your lever, you have bought yourself permission not to add a washing step — and at any volume that is worth real money. Ten seconds a jar across a 1,000-candle festive run is nearly three hours of labour you now do not spend, every run, on something that does nothing.
What to buy for the split test
Cheap clear glass, a wax whose page publishes the temperatures you need to hold still, and a stated-capacity vessel so your arithmetic is not a guess. Each card lists every pack and the per-unit maths.
| Size | Price | Per piece |
|---|---|---|
| Pack of 5 | ₹159.30 | ₹31.86 |
| Pack of 10 | ₹318.60 | ₹31.86 |
| Size | Price | Per kg | 70g votives @ 8% |
|---|---|---|---|
| 500 g | ₹299 | ₹598 | ≈7 |
| 1 kg | ₹599 | ₹599 | ≈15 |
| 5 kg | ₹2,995 | ₹599 | ≈77 |
| 10 kg | ₹5,999 | ₹599.90 | ≈155 |
| Size | Price | Per 100g | 70g votives @ 8% |
|---|---|---|---|
| 15 gm | ₹93.22 | ₹621.47 | ≈2 |
| 50 gm | ₹271.40 | ₹542.80 | ≈8 |
| 100 gm | ₹502.68 | ₹502.68 | ≈17 |
| 500 gm | ₹2,513.40 | ₹502.68 | ≈89 |
| 1 kg | ₹5,026.80 | ₹502.68 | ≈178 |
Reading the result: four outcomes and what each one means
Add the six scores in each arm at each checkpoint. Four patterns cover almost everything you will see, and each one tells you exactly where your next afternoon goes.
| Pattern | What it points to | What it costs to act on | Your next test |
|---|---|---|---|
| Arm A clearly better at checkpoint 1 and stays better | A real preparation effect | Seconds per jar, forever — so cost it per candle before you commit | Narrow it down: washed versus wiped versus untouched, three arms, one pot |
| Both arms equally patchy at every checkpoint | Preparation is not your lever; conditions or contraction are | ₹0 — you have saved yourself a labour step | Glass temperature at pour: warmed against room-temperature jars, same minute |
| Both arms clean at checkpoint 1, both patchy by checkpoint 3 | Temperature movement after the pour, not the glass | Storage and packing decisions, not raw materials | A recorded hold-and-move cycle with photographs at fixed points |
| Scores scattered with no arm pattern but a position pattern | Cooling position — outer jars, a draught, a cold surface | ₹0 to rearrange a bench | Mark four faces on each jar and run the orientation test |
Two of those four outcomes cost you nothing to act on, and all four leave the wax alone. That is the honest arithmetic of this subject: the levers are ordered by cost, the cheap ones are numerous, and the expensive one — a different wax — belongs at the end of a documented process, not at the start of a panic. If you do reach that point, the way to do it properly is candidate waxes in your actual jar, on the same day, scored on the same sheet, with a pilot before a bulk buy. The one-sided and middle-band patterns have their own posts: one side only and a band around the middle.
CANDLEMAKINGSUPPLIESINDIA supplies raw materials, not finished candles. The sale we are declining on this page is obvious: a 10 kg bag of a different wax, which is four to five times what the split test costs and far more profitable for us. We are telling you to spend ₹1,283.83 finding out which lever matters, because a maker who fixes the real cause carries on pouring and a maker who replaces wax on a hunch usually stops.
Every price and pack size on this page comes from live CSI stock in September 2026 and links to the product page, where current pricing always takes precedence. Specifications are quoted only where the page states them: the Luxury Soy Wax Chunks load, add temperature, two pour statements and cure guidance; the Clear Glass Votive Jar's stated "Capacity: ~70g wax fill". Two facts here are absences and they are stated as absences, because they are the reason this guide is a test rather than a procedure: no CSI container page gives a washing, wiping, drying or preheating instruction, and no CSI wax page publishes an adhesion measurement — only CSI 464 and CSI 468 mention glass adhesion at all, and both do so as wording with no figure, method or threshold. Luxury Soy Wax Chunks, Natural Soy Pearl Wax and Coconut Wax are silent on appearance and adhesion. No contraction rate, coefficient, adhesion score, threshold temperature or accelerated-to-real-time conversion is published by CSI and none is invented here.
For help designing a test for a specific jar, bulk pricing, GST invoicing, MSDS or IFRA documentation, message us on WhatsApp at +91 7397976926.
Frequently asked questions
Customer reviews: The five reviews at the top of this page are genuine, published Judge.me reviews left by CSI customers on their product pages. Names, star ratings, dates and products are as recorded by Judge.me; one longer review is shortened with an ellipsis and otherwise unedited.
Figures verified September 2026: Luxury Soy Wax Chunks for Jar / Container Candles — 500 g ₹299, 1 kg ₹599, 5 kg ₹2,995, 10 kg ₹5,999; stated "up to 13% fragrance load capacity", "Recommended fragrance load: 8% to 13% of total wax weight", "10% Recommended", "Recommended fragrance add temp 80-90°C", pour temperature stated two ways on the same page as "Recommended pour temp 75-80°C" and "Pour at 70-80°C", cure "a minimum of 48 hours (ideally 1-2 weeks) before performing burn tests"; melt point not stated; silent on adhesion and appearance. Clear Glass Votive Jar 70 gram — pack of 5 ₹159.30, pack of 10 ₹318.60 (₹31.86 per jar); stated "Capacity: ~70g wax fill"; no dimensions stated, and the page body also refers to matte finishes. Sandalwood Fragrance Oil — 15 gm ₹93.22 (₹6.21/g), 50 gm ₹271.40, 100 gm ₹502.68 (₹5.0268/g), 500 gm ₹2,513.40, 1 kg ₹5,026.80. Eco Candle Wicks Thin (C1) — 10 wicks ₹59, 20 wicks ₹118 (₹5.90 each). Candle Wick Stickers 1 sheet of 40 ₹100 (₹2.50 each). Candle Making Weighing Scale ₹354. Luxury Soy Wax CSI 464 500 g ₹260, 10 kg ₹5,074 and Premium Coconut Soy CSI 468 10 kg ₹6,370 are quoted as the cost of the wax change this test is designed to defer; they are also the only two CSI wax pages that mention glass adhesion, as wording rather than measurement. Stated absences: none of CSI's 74 listed containers states a washing, rinsing, wiping, drying or preheating instruction; no CSI wax page publishes an adhesion figure, test method or threshold temperature; Luxury Soy Wax Chunks, Natural Soy Pearl Wax and Coconut Wax state nothing about appearance or adhesion. Planning assumptions, labelled: fragrance load is calculated as a share of the finished candle, so a 70 g votive at 8% is 64.4 g of wax and 5.6 g of oil (the Luxury Soy Wax Chunks page states its own load as a percentage of total wax weight; both conventions exist and this page states which it uses). Wax cost per votive = 1 kg price ÷ 1,000 × 64.4 g = ₹0.599 × 64.4 = ₹38.58; oil cost per votive = ₹5.0268 × 5.6 = ₹28.15. Test consumption: 772.8 g wax at ₹462.91, 67.2 g oil at ₹337.80, 12 jars at ₹382.32, 12 wicks at ₹70.80, 12 stickers at ₹30, total ₹1,283.83, or ₹106.99 per test candle; cart ₹599 + ₹502.68 + ₹477.90 + ₹118 + ₹100 = ₹1,797.58. One wick per single-wick candle. Checkpoint timings and the temperature-cycle hold are the maker's own choices, recorded, and no accelerated-to-real-time conversion is offered. Prices and availability change — the linked product pages are always authoritative.