My Fragrance Oil Is Collecting at the Bottom of the Candle Jar — Does That Mean I Added It at the Wrong Temperature or Failed to Mix It Properly?

Step 0: prove it is oil, not a wet spot

Look through the base of a clear jar and two very different things can look "wet". A wet spot is a patch where the wax has pulled away from the glass as it cooled and shrank, leaving a thin gap. Through the glass it reads as a shiny, silvery or mirror-like patch, usually with crisp, irregular edges, and it often appears on the walls as well as the base. Wet spots are common with soy, commonly linked to cooling rate and the temperature difference between wax and glass, and to how clean the glass was. They are cosmetic, not a defect in how the oil was mixed. Oil at the base is a liquid: darker and glossier, with soft edges, and it gathers at the lowest point.

Four checks, in this order, cost nothing:

If the answers point to wet spots, stop here: the fixes are clean, warmed jars and slower, even cooling, and the clear-glass wet-spot guide covers them. The CSI 464 page claims "excellent glass adhesion" and "Low frosting, minimal blooming"; that is the page's claim, and a wet spot in your jar is still about your jar and your room. If the checks point to oil, or you can't tell, sacrifice one candle for the hub's blot-and-look check and then run the 2×2 below. If the oil sits only in a ring around the wick tab, that is a different story (a tab gap or the wick drawing oil) and the wick-pooling post owns it.

How oil ends up under the wax

When you add oil to melted wax and stir, you are trying to take it fully into the wax before it leaves the pot. Two things work against you. Temperature: wax that has cooled towards its setting point is thicker and on its way to crystallising, and oil added then tends to incorporate less well; separation, mottling and droplets are the common signs (tendencies, not certainties). Time: a few seconds of stirring may leave streaks of oil that have not dispersed. Oil that is not taken in can end up low in the pot or the jar, or come to the surface, depending on the oil; CSI does not publish oil composition or densities, so we can't say which way a given oil goes. Either way, once the wax sets there is no second chance to mix; we know of no evidence that oil "settles" out of a candle that has already set, and we don't claim it.

That is why your question is the right one, and also why it can't be answered by looking. A cool add and a short stir leave the same mark. The 2×2 separates them by pouring both, deliberately, beside a control.

There is a third, less common story: too much oil for this wax. At a load near or past what the wax holds for a particular oil, oil can be pushed out wherever it finds room, including at the glass. That one does not care how hot or how long you stirred, which is exactly how the 2×2 catches it: all four cells, the control included, show oil. Load is stated here on CSI's total-weight basis: 8% of a 150 g candle is 12 g of oil and 138 g of wax (8.70% of the wax weight).

Base pattern → likely cause → what rules it out

Ranked most common first as editorial judgement; CSI publishes no failure statistics. Two orders are in play on this page: this table ranks by likelihood, while the checks run cheapest first (Step 0 is free, the 2×2 costs a few hundred rupees).

Rows 2 and 3 are the two halves of your question. They coexist often, in our judgement, because a maker who adds cool is often also the maker who pours quickly before the wax sets further. That is why the 2×2 crosses them instead of testing each alone. Rows 6 and 7 are other posts' territory; if your pattern matches either, go there instead.

The clear-jar 2×2, step by step

The design crosses two factors at two levels each. When the oil goes in: at the wax page's figure (Eco Soy CSI 400: 85°C) or cooler (a worked-example 65°C, the top of the range one oil FAQ suggests for soy; not a CSI wax figure). How long it is stirred: the page's "stir gently for 2 minutes" or about 20 seconds (a worked-example short stir). Each pot is 150 g at 8% of total weight, 12 g of oil and 138 g of wax, and pours two 70 g votives (5.6 g of oil and 64.4 g of wax in each), leaving about 10 g in the pot. A clean 150 g pot is the smallest we would trust to hold temperature long enough to stir for two minutes (judgement).

Costs are worked examples and exclude labour, freight, packaging and GST. With your own oil, already bought, the materials used come to ₹569.53. If you own nothing, the outlay is ₹1,359.60 (wax, a 50 g oil, ten votives, five sticks and the Pen Thermometer); with your own oil and thermometer, ₹835.60. The 50 g bottle leaves only 2 g after the four pots, so a pooled oil you want to retest at a lower load needs a second bottle. For scale, one unwicked 70 g test votive at these prices is ₹76.60 in materials.

Reading the bases and the decision rule

Call a cell oily if either of its two votives scores 2 or 3 at any of the three readings. Then read the pattern:

Two checks keep the reading honest. The pair must agree: if A1 is clean and A2 oily, the control is unstable and nothing else can be read; look at where A2 cooled. And day 7 outranks 24 hours: soy keeps setting over days (a tendency), and a base that looks clean the next morning can change. If day 7 shows oil where 24 hours didn't, read the day-7 pattern and note the delay; slow-appearing surface oil in storage is the storage post.

When the 2×2 names the cause, change only that in your next production batch, and keep one retained candle from it to compare with a retained candle from the old batch at day 7. That is your proof the fix held at scale. A kilo batch at 8% is 80 g of oil and 920 g of wax (the load-basis hub has the conversions); on a batch that size, a clock on the wall is worth more than a stronger arm.

What not to buy for a pooled base

For the first-level version of this question (what separation looks like and the usual causes) see the separation hub; for "won't mix in" as a temperature question, the oil-or-temperature post. Stearic Acid, sometimes suggested as a hardener, is out of stock on 26 September 2026; nothing in this 2×2 needs it.

The five items the 2×2 uses

In the order you would reach for them: the wax whose page defines cell A, a clear window on the base, a clean stick per pot, the thermometer that answers the temperature half, and a second oil only for the all-four-oily branch. Chosen on page statements, price per unit and stock on 26 September 2026; nothing is ranked on performance, because CSI publishes no measured data. The oil table shows every size, 150 g candles at 12 g and 70 g votives at 5.6 g.

Frequently asked questions