My New Melter Has a Bottom Dispensing Tap and the Last Wax Leaving the Vessel Behaves Differently — Could Material Distribution Inside the Tank Explain It?

A tap reverses the order in which your tank empties

When you pour by ladle or pitcher, you take wax from the top of the tank and work down. The first candles get the surface layer and the last candles get whatever was near the floor. A bottom dispensing tap turns this upside down. The first candles get wax from just above the outlet, the middle of the run drains the bulk, and the last candles receive the layer that sat at the surface from the moment you finished stirring.

If the batch were perfectly uniform in composition and temperature from top to bottom, the order would not matter. In practice a large, still mass is rarely perfectly uniform for long. So a tap does not create the differences; it arranges them into a predictable pour sequence, which is exactly why the last candles can look or smell different.

A note on equipment: CSI's Electric Wax Melter page does not describe a bottom tap, and CSI does not sell heated pouring pots with taps. Everything below applies to bottom-tap melters in general. Check your own unit's manual for how its outlet is built.

Four reasons the tail of the batch can differ

1. The surface layer has had the most air

The top of a warm tank is the only wax in contact with air. Volatile top notes evaporate from that surface throughout the run, and a thin skin can form if the surface cools. Wax in the middle is protected by the wax above it. With a tap, the surface layer is the last wax to reach a jar, so it has had the longest exposure of any wax in the batch. Citrus oils make this more noticeable: the Citrus Lemon page itself notes that citrus top notes are volatile and fade faster than warmer families.

2. The last kilograms heat differently

As the level drops, the heater is warming a smaller and smaller mass. With less wax to absorb the same heat, the remaining layer can climb above the temperature the bulk held earlier, and more of it is in contact with hot walls. The last few candles may therefore be poured hotter than the first, even with the setting unchanged.

3. Anything not fully incorporated leaves at one end

If oil, dye or an additive was not completely mixed in, it does not stay evenly suspended in a still tank. Depending on its density relative to the wax, it can drift towards the surface or towards the floor. Either way, a tap concentrates it into the first or last candles rather than spreading it through the run.

4. The dead leg delivers the first draw

The tap and its short outlet pipe hold wax that is not stirred with the batch. That wax may have been sitting there since before the oil went in, and it is often cooler than the tank. The first draw after mixing can therefore be unscented or under-scented, cool and cloudy. This is the opposite end of the run, but it shows the same principle: the tap's geometry decides which wax goes where.

A falling level also changes the stream

One more effect is mechanical rather than chemical. A full tank pushes wax out of the tap quickly because of the weight of wax above the outlet. As the level falls, that pressure drops and the stream slows. Operators who were filling a jar in a few seconds at the start may take noticeably longer near the end, and a slower stream loses more heat between the tap and the jar. If your team fills to a fixed count or a fixed time rather than on a tared scale, late fills can also drift in weight. Weighing each fill and holding a thermometer in the stream removes both uncertainties.

None of these effects is guaranteed in any particular melter. Tank shape, heater position, insulation, lid design and outlet size all change how strongly they show. That is why the answer to this question comes from sampling your own tank rather than from a general rule.

Worked example: proving where the difference sits in a 20 kg batch

Suppose a maker drains a 20 kg batch of CSI 464 with Citrus Lemon at 8% of total weight (18.4 kg of wax plus 1.6 kg of oil) through a bottom tap into 100 candles of 200 g. Candles 91–100 have noticeably weaker cold throw and slightly rougher tops. Candles 1–3 look hazy. Everything in between seems fine.

Instead of guessing, the maker runs the next batch with a sampling plan. The draw points and sample counts below are suggested starting points to validate, and the temperature readings shown are illustrative.

The maker holds a pen thermometer in the tap stream at each marked draw, labels each candle, and after the page's stated minimum cure of 48–72 hours, and again at later check points they choose, compares the marked candles side by side for cold throw and appearance, then burns the 25, 50 and 100 candles together.

If candles 25 and 50 match and the tail differs, the tank is layering during the run. If the first draw alone differs, the dead leg is the cause. If differences appear throughout, the problem is mixing, not the tap.

What to do with the candles already poured

Treat the batch in three groups based on draw position, because the evidence says they are not the same product.

On the cost side, ten tail candles are 2 kg of scented wax; recovering them into the next Citrus Lemon batch keeps the material, and only the jars and wicks are lost.

Process controls for a tap-drained melter

Where this page fits

CSI already has guides that cover parts of this question. This page covers the failure investigation for the question above; these go deeper on the rest:

Three things to buy for this job

Frequently asked questions