My Candle Looks Perfect for a Week and Then Develops Adhesion Patches — Why Can Wax Pull Away From Glass After the Candle Has Already Set?

Why a set candle is not a finished candle

The intuition that makes this symptom so disorienting is the idea that a candle is done. It looked right, it was lidded, it went on a shelf — surely whatever was going to happen has happened. But a candle is a solid block of fat-derived material pressed against a sheet of glass, and solids expand and contract with temperature. A set soy candle is still doing that, quietly, in both directions, every day. Where it has a firm grip on the glass, nothing shows. Where its grip is marginal, a cycle of expanding and contracting can be enough to break contact, and once contact is broken it does not come back on its own. That is why the mark appears and then stays.

We are going to be strict about what we are not telling you here. No contraction rate. No shrinkage percentage. No coefficient. No temperature above which a gap forms or below which it does not. None of those figures is published anywhere in CSI's data, and a confident-sounding number invented for a blog post would send you buying a fridge or a fan on false precision. What is honest to say is the direction of the thing: contraction is real, it continues in a set candle because temperature continues to change, and the way to find out what your store room is doing to your candles is to measure your store room.

The other thing worth saying plainly, because it changes how much you should spend: this is a cosmetic problem. A void against the glass holds no liquid, does not reach the melt pool — which forms at the top surface and works down — and does not change how the candle burns. What it changes is what a customer concludes when they look through clear glass, which is a real commercial cost and the reason you are reading this. But it is an appearance standard your brand sets, not a safety fault. The exceptions stay exceptions and they are unmistakable: cracked glass, a wick leaning against the wall, a wick that clogs and smokes, or a jar too hot to hold. Those candles do not ship whatever week it is. The symptom itself is explained from scratch in why does my candle have wet spots, and whether gaps affect performance at all is the subject of the cosmetic-or-performance post.

The post-set suspect list

Here is the differential for the post-set window. Every row is a candidate and an observation that separates it — "could" and "check", never "is". The striking thing about this table is that almost nothing on it is a raw material. It is nearly all geography and habit, which is why it is nearly all free to fix.

The 30-day hold log

This is the protocol. It is yours — every duration, temperature and checkpoint below is a choice you are making, not an industry standard, and there is no conversion between a held month and a customer's shelf life. What the method gives you is a dated, grouped, photographed record of your own storage conditions, which is something no product page can supply.

Where in the stack: tagging position

This is the part that surprises people and the reason this post uses position as its unit of analysis. Candles in a stack do not experience the same conditions. A jar on the outer face of a carton follows the room closely. A jar in the middle, insulated by twenty other jars and a layer of cardboard, lags behind it — warming later, cooling later, and never reaching the same extremes. If the room swings and your gaps correlate with position, you have learned something you can act on this afternoon, for nothing.

A practical note on stacking while warm. Candles that are still above room temperature when they go into a carton take far longer to settle, and the centre of that carton is the slowest place in your building. Whatever your hold log tells you about position, a simple rule costs nothing and removes a whole class of confusion: nothing gets boxed until it has been photographed at 24 hours. That single habit also gives you the day-one baseline the hold batch depends on. Where cooling position matters on pour day rather than in storage, the sibling post on cold glass and fast solidification covers the other end of the same argument.

The wax and the jars for a long hold

A 30-day hold wants materials you can reorder for years, because the log you build is only valuable if the next batch is comparable. One wax whose page states nothing — which is honest and matters — and two vessels at opposite ends of the visibility question. Each card shows every listed pack with the per-unit maths.

Moving inspection from pour day to dispatch day

If your gaps are born in week two, then a pour-day inspection is not a quality system — it is a formality. It passes candles that will fail in a customer's hands, and it does so with complete confidence. The commercial fix is structural rather than technical, and it costs shelf space rather than money.

The third row is the one to adopt. Grade at 24 hours to gate the pour, grade again at dispatch to decide what ships, and record both. Over a few months the two columns tell you exactly where your rejects are being created, and that is the number that justifies any spending at all. The length of the settling period is your decision — note that the Luxury Soy Wax Chunks page recommends a cure of a minimum 48 hours and ideally one to two weeks before burn testing, so a settling window that runs alongside cure costs you almost nothing you were not already spending. Where to set the pass line is the subject of the sibling post on how much adhesion variation is normal, and there is no published figure for it anywhere — it is your brand's call.

Frequently asked questions