Why Are There Air Pockets Inside My Candle?
Share
Do they matter? Mostly no. Surface bubbles and fine haze are cosmetic. A void wider than 5 mm sitting within about 10 mm of the wick is not cosmetic — it interrupts the fuel column. That is the one you re-pour for.
How do I stir without aerating? Slow, submerged, figure-of-eight, for the full two minutes the wax specifies. Do not whisk. Do not lift the stirrer out and drop it back in. Every time it breaks the surface it drags air under.
Can I buy my way out of this? No. This is a technique fault, not a materials fault. No wax, additive or tool in the catalogue prevents air pockets — a ₹349 mixer run too fast will make them worse.
The three sources of air, and how to tell them apart
Worth stating plainly, because it changes what you go looking for. A sinkhole is shrinkage as the wax crystallises. A crack is stress released when a set candle cools unevenly. An air pocket is neither — it is a physical bubble of atmosphere that was in the liquid wax at the moment the wax stopped being liquid. The cause is therefore always upstream of the pour, in something you or your equipment did, and the fix is always technique rather than raw material. If you are not yet sure which of the three faults you have, work through the sinkholes vs cracks vs air pockets diagnosis first; this page assumes you have landed on air pockets.
Air enters liquid wax by three routes and no others: whipped in during stirring, splashed in during pouring, or trapped against a cold vessel wall as the wax skins on contact. Each leaves a different pattern, because the physics differs — whipped air is finely divided and evenly spread, splashed air arrives in a few larger slugs near the end of the fill, and wall-trapped air is held by the surface it is stuck to. Read the pattern and you have named the step that is going wrong.
| Source | What you see | Where in the jar | The fix |
|---|---|---|---|
| Stirring | Large numbers of fine bubbles, mostly under 2 mm; a cloudy, "fizzy" look in clear glass | Evenly through the full depth, wall to wall | Slow, submerged, figure-of-eight stirring for the full two minutes |
| Pouring | A handful of larger irregular voids, 3–10 mm, rounded or teardrop | Upper half, often near the centre around the wick | One unbroken stream down the side, spout close to the surface |
| Cold vessel wall | Small round bubbles in a line or curtain, seen only from outside the glass | Flat against the inside of the vessel; the core is clean | Warm the vessel to roughly 40–50°C before it takes wax |
| Not air at all | A single conical cavity, or a straight fracture line | Under the wick, or running through the body | See sinkholes and cracking after cooling |
Stirring: two full minutes without aerating a single gram
The instruction on the wax is unambiguous. Eco Soy CSI 400 at ₹399 per kilogram specifies adding fragrance at 80–90°C with two full minutes of stirring, and Luxury Soy Wax CSI 464 melts and pours in the same 80–90°C band. Two minutes is not padding — it is how long fragrance oil takes to dissolve into wax rather than sit dispersed as fine droplets, which is why undermixing shows up later as separation and weak throw.
Here is the conflict nobody warns you about. Two minutes feels absurdly long with a stick in your hand, so the instinct is to stir faster to make the time pass. Faster stirring is exactly how you aerate a batch. Wax at 80–90°C is thin — closer to warm cooking oil than to syrup — and thin liquids entrain air readily. Every fast circular sweep pulls a vortex down from the surface; every time the stirrer breaks the surface and goes back in it carries a film of air under. Do that for 120 seconds and you have made a very mild wax foam, which then sets in eight to twelve minutes with the bubbles still in it.
Then the mixer question. The CSI Pro Mixer, a cordless handheld mixer, is ₹349, and it does something a stick cannot: an even, unarguable two-minute incorporation on a large batch without your forearm giving out on the eighth jug of the morning. It is also the fastest way to aerate a batch that exists in this catalogue. A whisk-style head spun at full speed in thin 85°C wax folds air in faster than any hand could, and the result is fine bubbles from base to rim.
Pouring: one steady stream, close to the surface
The pour lasts about six seconds and creates most avoidable voids. The physics is the same as filling a glass from a tap: a stream that falls a long way, or breaks up on the way down, entrains air along its whole length and drives it under on impact. A stream arriving short, smooth and at an angle slides in and displaces almost nothing.
Four rules cover it, and none cost anything. Pour down the side, not into the middle. Angle a wide jar slightly, put the stream onto the glass a couple of centimetres above the rising wax level, and let it run down the wall rather than plunge into the pool. Pouring into the centre also drives the stream onto the wick, pushing it off vertical and sometimes washing the sticker loose. Keep the jug close to the surface and follow the level up — a 20 cm drop is a 20 cm column of stream picking up air; two centimetres is nothing. Pour one continuous stream: set the flow and hold it. Do not dribble. A hesitant stop-start trickle is worse than a confident pour, because every restart slaps a fresh pocket of air into the surface and a thin broken stream exposes more surface to air per gram of wax.
Pour temperature interacts with all of this, and it is a genuine confounder. Wax poured hot is thinner, entrains air more readily on impact, but also lets that air rise out during a longer cooling window. Wax poured cool is thicker, splashes less, but sets fast enough to freeze in whatever it caught — and thick wax poured cool is the classic recipe for both trapped air and a lumpy top. The relationship between pour temperature and cavities is worked through properly in does pouring temperature cause candle sinkholes, and the same temperature discipline is what produces a clean top surface — see why is the top of my soy candle rough. Whatever band your wax specifies, the practical instruction is the same: know the number, read it off a thermometer rather than guessing, and pour the whole batch inside it.
Cold vessel walls: bubbles cling where wax chills instantly
This is the source most makers never consider, and the one that produces the bubbles customers complain about — wall bubbles are the only kind visible from outside a clear or frosted jar.
The mechanism is straightforward. Glass conducts far better than wax and holds the temperature of the room it has been sitting in. Pour 85°C wax into a jar off a shelf in a 20°C air-conditioned workshop and the layer touching the glass drops below its congealing point in a fraction of a second. That first skin is what everything else sets against. Any bubble sitting at the glass at that moment is no longer floating free in liquid wax; it is embedded in solid wax, and it will still be there in three weeks. The same instant chill causes wet spots — the patchy grey areas where wax pulled away from the glass instead of bonding to it.
Two practical notes. Warm the whole tray at once — twelve White Matte Glass Jars (₹354 per pack of 5, ₹70.80 each) in a low oven while you do the fragrance addition costs no extra time, because it runs in parallel. And do not overheat: a jar too hot to hold slows setting so much that the candle stays soft for hours, inviting frosting and a greasy top.
Season matters more in India than most guides admit. A workshop in Chennai in May and the same workshop in Delhi in January are two different manufacturing environments, and a recipe that pours clean in one will bubble in the other. If air pockets appeared suddenly and nothing in your process changed, check the ambient temperature before you blame your wax. Cold rooms also compress the working window between stirring and pouring, which pushes you into rushing — bringing back source number two.
Which pockets actually matter, and how to fix the ones that do
Being honest here saves money. Surface bubbles are cosmetic — a scatter of small craters on top of a soy candle affects nothing about how it burns, and nobody sees them under a lid. Fine bubbles through the body are cosmetic too: invisible in a matte, frosted or tin container, and in clear glass they read as a slight cloudiness many buyers cannot tell from normal soy crystallisation. If you pour into opaque containers, a light stirring haze is not a reason to bin a batch.
A large void near the wick is a different matter, and it is not cosmetic. The wick draws liquid wax up from the melt pool continuously. When the pool burns down to a cavity the flame drops as the fuel column is interrupted, or the pool floods into the void and drowns it, or it flares as it briefly gets more air and less wax. A void touching the wick also means the wick is not bedded in wax over that stretch, which is where a wick leans or a hot spot builds against the glass. This is the failure mode that produces a customer message.
| Test | How to run it | Pass / fail rule |
|---|---|---|
| Backlight (clear glass) | Hold the cured candle between your eye and a bright lamp, rotating slowly through 360° | Fail if a void wider than about 5 mm sits within roughly 10 mm of the wick. Fine haze anywhere: pass |
| Fill weight (opaque jars) | Weigh the finished candle on a weighing scale (₹354) against your recorded target for that vessel | Fail at more than about 3% under target with a correct fill line — that missing mass is air |
| First burn | Test-burn one jar per batch to a full melt pool, watching the flame at the void depth | Fail on any flame drop, flare or drowning as the pool reaches that depth |
Your options narrow sharply as the wax cools, so the response depends on how long ago you poured.
Seven lines, in the order you do them. Stick it above the bench; it is worth more than any product on this page.
If air pockets come and go across batches you have an uncontrolled variable, and you will not find it from memory. Record these eight fields at the bench, during the pour, for every batch until the fault stops appearing. It takes under a minute and converts an argument into a comparison.
| Field | What to record | Why it matters |
|---|---|---|
| Room temperature | The actual reading at the bench, in °C | The strongest predictor of wall bubbles; a January pour and a May pour are different processes |
| Vessel prep | Warmed or not, method, approximate jar temperature | Isolates source three; half a tray warmed and half not is a one-batch experiment |
| Add temperature | The thermometer reading when fragrance went in | 80–90°C is the stated band for Eco Soy CSI 400; cool wax needs more agitation |
| Stir method and time | Stick or mixer, speed setting, seconds on a clock | Isolates source one; "two minutes" is usually 40 seconds in practice |
| Rest before pour | Seconds the jug stood still | The cheapest variable on the sheet, and the one most often skipped |
| Pour temperature | Reading at the moment the first jar was filled | Drives viscosity, splash and how long air has to escape |
| Pour technique | Side or centre, drop height, stream broken at any point | Isolates source two; drop height is the number people guess wrongly |
| Result at 24 hours | Bubble pattern, size and location; photograph against a lamp | Turns "it looked bad" into evidence you can line up against the other seven |
One last commercial point. Air pockets are a consistency failure rather than a quality failure — the candle usually still burns, still smells and still sells. What they cost you is the ability to promise the same jar twice. A wholesale or corporate gifting client is not buying one good candle, they are buying two hundred identical ones, and a tray where six jars are cloudy and fourteen are clear cannot be photographed as a set. Fixing the stir and the pour is about quoting a repeat order with confidence.
CANDLEMAKINGSUPPLIESINDIA supplies raw materials, not finished candles. On a page like this one that is worth saying out loud, because the commercially useful version of this article would blame the wax and sell you a different one. It is not the wax. Air pockets are caused by handling, and the two products named here — ₹118 stirring sticks and a ₹531 heat gun — total ₹649 against a fault that a slower hand fixes for nothing.
Every price on this page is taken from live CSI stock in September 2026 and links to the product page, where current pricing and specification always take precedence. The 80–90°C addition band with two minutes of stirring is the figure stated for Eco Soy CSI 400; Luxury Soy Wax CSI 464 states the same 80–90°C band for melting and pouring. Fragrance load ceilings quoted elsewhere on the site — up to 13% for Luxury Soy Wax Chunks and up to 10% for CSI 464 — are wax manufacturers' formulation limits, not regulatory limits.
The 5 mm and 10 mm thresholds in the backlight test, the 3% fill-weight rule and the 40–50°C vessel target are working bench rules for screening a batch, not laboratory standards; they are set deliberately conservative so that a jar which passes is a jar you can ship. For bulk pricing, GST invoicing, MSDS and IFRA documentation, or help matching a wick to a specific jar diameter, message us on WhatsApp at +91 7397976926.
Frequently asked questions
Customer reviews: The six 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, and only reviews Judge.me records as verified carry the "Verified buyer" label. They are general reviews of the products named — Bamboo Stirring Sticks, Luxury Soy Wax CSI 464, Soy Pillar Wax, Eco Soy CSI 400 and the Hot Air Gun — and not reviews of the technique described on this page. One longer review is shortened with an ellipsis and is otherwise unedited. No location is attached to any review because Judge.me does not record one.
Figures verified September 2026 (CSI live pricing): Bamboo Stirring Sticks for Mixing Wax ₹118 per pack of 5 (₹23.60 each), ₹1,062 per 50 (₹21.24 each), ₹1,888 per 100 (₹18.88 each). CSI Pro Mixer cordless handheld mixer ₹349. Hot Air Gun ₹531. Pen Thermometer for Candle Making ₹354. Candle Making Weighing Scale ₹354. Candle Wax Pouring Jug ₹531. Steel Pouring Pitcher ₹708 for the 600 ml size; the 150 ml size is out of stock on this pull. Mini Electric Wax Melter (1 litre) ₹2,360. Eco Soy CSI 400 ₹399 per 1 kg. Luxury Soy Wax CSI 464 flakes ₹260 per 500 g. Luxury Soy Wax Chunks ₹299 per 500 g and ₹5,999 per 10 kg. Premium Coconut Soy Wax CSI 468 ₹325 per 500 g. White Matte Glass Jar ₹354 per pack of 5 (₹70.80 each). Cheapest in-stock container ₹31.86 each in packs of 5 or 10. Eco Candle Wicks from ₹59 per 10. Candle Accessories collection holds 65 products. Eco Soy CSI 400 specifies fragrance addition at 80–90°C with two minutes of stirring; CSI 464 states melt and pour at 80–90°C. Stated fragrance ceilings of up to 13% (Luxury Soy Wax Chunks) and up to 10% (CSI 464) are wax manufacturers' formulation limits, not regulatory limits. Diwali 2026 falls on 8 November. The 5 mm void threshold, the 10 mm wick clearance, the 3% fill-weight tolerance and the 40–50°C vessel target are CSI bench screening rules, not laboratory standards. Prices and availability change — the linked product pages are always authoritative.