Overpowering scent in a small tin, and how to bring the dose back down
Counting drops overdoses a small batch and a sealed tin concentrates it. Dose arithmetic in grams, a dilution rescue and the ceilings that make this a safety issue.
An overwhelming balm is usually an arithmetic problem wearing a sensory disguise. Almost every case traces back to two things: the dose was counted in drops rather than weighed, and it is being judged from a sealed tin rather than from skin. Both are fixable, and the first step is to find out whether you have a preference problem or a limits problem, because those have different answers.
A drop of essential oil weighs between about 20 and 50 mg depending on the dispenser, so 20 drops in a 25 g batch is anywhere from 1.6 to 4.0%. Reweigh the batch on paper first. If the result is above roughly 1% for lips or 2% for body, dilute with unscented base at one to one or one to two, recombine warm, and repour.
Three different complaints, one phrase
"It smells too strong" covers three situations that need separating before you touch the batch.
The first is a genuine overdose: the finished product carries more fragrance than the arithmetic ever intended, and the smell is proportionate to what is actually in there. The second is a perception artefact: the dose is correct, but a closed tin accumulates volatiles for hours and delivers them in one breath when the lid comes off, so the tin reads far stronger than the film on skin ever will. The third is the serious one, where the smell is a symptom rather than the fault, because the level that smells overpowering is also above a regulatory ceiling for one of the constituents in the oil.
The three are told apart on paper, not by nose. Recalculate the batch by weight, look up the oils you used, and only then decide whether you are diluting a product or correcting a formula. The full picture of what binds at each level is in essential oils in balms.
The drop is not a unit of measurement
The only place a drop has a defined mass is in the pharmacopoeia, and it is defined for water. The European Pharmacopoeia standard dropper, chapter 2.1.1, has a nominal 3 mm outlet and is specified to deliver 20 drops of water weighing 1.000 g, that is 50 mg per drop, within a tolerance of 50 mg on the gram. Nothing about that transfers to an essential oil.
Drop mass is governed by surface tension, not by volume or by density. Tate's law gives the mass of a slowly formed pendant drop as the tip circumference times the surface tension divided by gravity, with a correction factor of roughly 0.7 for the fraction that stays behind on the tip. Water has a surface tension near 72 mN per metre. Most essential oils sit between 26 and 30. From the same orifice, an essential oil drop is therefore around 40% of the mass of a water drop.
| Dispenser and oil | Tip diameter | Drop mass | Drops per gram |
|---|---|---|---|
| Ph. Eur. standard dropper, water (the calibration) | 3 mm | 50 mg | 20 |
| Orifice reducer, thin citrus oil | 3 mm | 18 mg | 55 |
| Orifice reducer, typical essential oil | 3 mm | 21 mg | 48 |
| Wide orifice reducer | 4 mm | 28 mg | 36 |
| Glass pipette or dropper insert | 5 mm | 35 mg | 29 |
| Cold or resinous oil tipped from the bottle lip | variable | up to 50 mg | 20 |
The practical span is a factor of nearly three between the lightest and the heaviest realistic drop. That is why the "20 drops per gram" shortcut that circulates in recipe collections is wrong in both directions: it is right for water from a pharmacopoeial dropper, and it overstates a citrus oil from an orifice reducer by nearly a factor of three. Any recipe written in drops is carrying that error into every batch made from it.
Two bottles from the same supplier can have different reducer bores, and a cold oil forms bigger drops than a warm one. A recipe that worked last winter and reads overpowering this summer may have had nothing changed except the temperature of the bottle. Never treat a drop count as reproducible between bottles, between oils, or between rooms.
Recalculate the batch on paper
Before deciding anything, work out what you actually made. Multiply drops by a plausible drop mass, divide by batch weight, and express it as a percentage of the total.
| Drop mass | 10 drops | 20 drops | 30 drops |
|---|---|---|---|
| 20 mg, thin oil from a fine reducer | 0.8% | 1.6% | 2.4% |
| 30 mg, typical | 1.2% | 2.4% | 3.6% |
| 50 mg, wide bore or viscous oil | 2.0% | 4.0% | 6.0% |
Small batches are where this bites hardest, because the fragrance is a fixed number of drops while the base shrinks. Twenty drops is a sensible addition to a 250 g body balm and an overdose in a 25 g test pot, and the test pot is exactly where most people develop a formula. Scaling in the other direction is just as unforgiving, which is one of the reasons a formula that felt right at bench scale misbehaves in production, covered in batch scaling. If you would rather not do the arithmetic by hand, the essential oil dilution calculator takes a batch weight and a target percentage and returns grams.
Check it against the ceiling, not against your nose
Once you have a percentage, look up what the oils in it are allowed to contribute. This is where a smell complaint can turn into a compliance one, and it happens more often than makers expect because the binding limits are set per constituent rather than per oil.
- Lip products are IFRA Category 1, the strictest category there is.
- Citral: 0.11% in lip products, 0.15% leave-on body. Lemongrass and litsea are 40 to 80% citral.
- Eugenol: 0.45% lip, 0.64% body. Clove bud is around 85% eugenol.
- Isoeugenol: 0.019% lip. That is nineteen thousandths of one percent.
- Methyl salicylate: EU Annex III caps it at 0.06% leave-on and 0.03% on lips. Wintergreen is 85 to 99% methyl salicylate.
- Expressed bergamot: 0.40% for phototoxicity. Expressed lime 0.70%.
- The summation rule: each material as a percentage of its own limit, added up, must not exceed 100.
Work one of those backwards and the scale of the problem is obvious. A lip balm at 4% lemongrass, which is what 20 heavy drops in 25 g gives you, carries about 2.8% citral against a limit of 0.11%. That is twenty-five times over, and it is not a matter of taste. Clove bud at the same level puts eugenol near 3.4% against 0.45%. A muscle balm at 1% wintergreen sits at roughly 0.9% methyl salicylate against an EU leave-on cap of 0.06%. In each case the overpowering smell and the overdose are the same fact seen from two directions.
The reverse is also true and worth saying plainly. Some materials smell enormous at entirely legal levels. Vetiver, patchouli, blue tansy, oakmoss and most absolutes are perceptible at hundredths of a percent, so a balm can be genuinely unpleasant and still be nowhere near a ceiling. A strong smell is a reason to check the numbers, not proof that they are wrong. The allergen declaration thresholds are a separate matter again, starting at 0.001% in a leave-on product, and are handled in fragrance allergens.
Why a sealed tin smells stronger than the balm is
Assume for a moment the arithmetic came out fine. There is still a reliable reason the product seems too strong, and it is the container.
A closed tin is a small sealed headspace above a reservoir of volatiles. Over the hours after filling, the aroma compounds partition between the balm and the air above it until the air is effectively saturated at that temperature. Opening the lid delivers all of that at once, straight up into your face, at a concentration nothing on skin will ever reach. The film you actually wear is a fraction of a gram spread over a large area in open air, losing its top notes within minutes.
Two things make the tin worse than it needs to be. Volatility climbs steeply with temperature, so a tin that has sat in a warm room or a car reads dramatically stronger than the same tin from a cool shelf. And a deep, narrow tin releases its accumulated headspace as a single slug, where a shallow wide one has already been losing it. None of that changes what the balm does in use.
Judge fragrance the way the product is used, not the way it is stored. Take a rice-grain amount on the back of the hand, rub it in, and assess at 30 seconds and again at 10 minutes, in a room where the tin has not just been opened. Then ask someone who has not been in the workshop, because your own nose adapts within about half a minute and a maker is the least reliable judge of their own batch. The protocol belongs in sensory testing.
Dilute it: the arithmetic
Dilution is the whole rescue, and it is one equation. Adding n parts of unscented base to one part of over-fragranced balm divides the fragrance percentage by n plus one. Nothing more sophisticated is needed, because the fragrance is dissolved in the fat phase and goes wherever the fat phase goes.
| Starting level | Plus 0.5 parts | One to one | One to two | One to three |
|---|---|---|---|---|
| 2.0% | 1.33% | 1.00% | 0.67% | 0.50% |
| 3.0% | 2.00% | 1.50% | 1.00% | 0.75% |
| 4.0% | 2.67% | 2.00% | 1.33% | 1.00% |
| 6.0% | 4.00% | 3.00% | 2.00% | 1.50% |
So a 25 g pot at 4% becomes 50 g at 2% if you blend it with 25 g of base, or 75 g at 1.33% if you use 50 g. For a lip product aiming at 0.5 to 1%, one to three is usually the honest ratio, and the batch quadruples. That is the real cost of the fault: not the ingredients, but suddenly owning four times as much of a product you were unsure about.
Make the diluent to the same formula minus the fragrance, including its antioxidant, so that the tocopherol level stays where it was rather than being halved along with the smell. Diluting with plain oil instead will soften the balm and change everything else about it, which then needs correcting in hardness as well.
Recombining without making it worse
- Decide the target first and weigh to it. Work out the diluent mass on paper before anything is warm. Reheating twice while you make up your mind costs more volatiles than the dilution does.
- Melt the unscented base separately and bring both to the same temperature. A butter-based balm with little or no wax combines happily at about 50 C, which is low enough to keep volatile loss modest. A beeswax formula cannot: beeswax melts at 61 to 65 C, so both parts have to reach 65 to 70 C before they will homogenise.
- Combine off the heat and stir for a full minute. A streaky blend gives you tins of two different strengths, which is worse than the original fault because it is invisible.
- Top up any heat-sensitive additions. Tocopherol and flavour go in below 50 to 60 C, as with any batch.
- Cool with stirring and pour into the tin window. Tins at 60 to 68 C, tubes at 65 to 72 C, per pour temperatures, then cool fast and uncovered.
- Reassess at 48 hours, on skin. Freshly poured balm reads sharper than the same balm a fortnight later, because the lightest molecules leave first. Judging a rescue on the day it was made is how a batch gets diluted twice.
Every remelt costs top notes, and the loss is not even across the blend. Citrus and other light materials go first, heavier woods and resins stay, so a twice-remelted balm is not just quieter, it is a different accord. If the original blend was carefully built, expect to adjust it rather than expect it to survive. The opposite fault, where a scent disappears within weeks, has its own causes in scent fading.
When cutting the smell is the wrong move
Some strong-smelling materials are in the formula for what they do, not for how they smell, and their level is set by the function. Menthol and camphor are the clearest case. Menthol at 0.5 to 2% in a foot or body balm is doing measurable work on cold receptors, and a US monograph chest rub sits at roughly 4.7 to 5.3% camphor with 2.6 to 2.8% menthol. Halving those to make the tin pleasant produces something that smells like a chest rub and no longer behaves like one, which is a labelling problem as much as a formulation one. The dose logic is set out in menthol, camphor and actives and in the chest rub formula.
They are also dose limited for reasons that have nothing to do with smell. Exposure to as little as 500 mg of camphor has been cited as a cause of mortality, and 750 to 1000 mg is associated with seizures and death. A 5% camphor balm carries 500 mg in ten grams of product, roughly a blob the size of a large grape, and it is normally sold in a tin that a small child can open. That is not a fragrance decision.
Three more cases where dilution alone is the wrong answer. If the smell has changed rather than merely being loud, going sharp, flat or faintly like turpentine, the oil has oxidised and the hydroperoxides that formed are more sensitising than the parent terpenes, which is a discard rather than a dilution: see when an essential oil smell turns. If the balm also stings, the level is doing something to skin and the triage is in balm stings or burns. And if the product has already been sold, an above-limit constituent is a recall question, not a reformulation one, which is the point at which selling balms in the UK and EU becomes the relevant page.
Prevent it: weigh, and premix
The prevention is dull and completely effective. Buy a scale that reads to 0.01 g, calibrate it, and stop counting anything. Fragrance is the one ingredient in a balm where a tenth of a gram matters, and a scale that reads to 0.1 g cannot resolve the difference between 0.5% and 1% in a 50 g batch.
There is a limit to what any scale can do, though: readability is not accuracy, and a sensible working minimum is about ten times the readability, so 0.1 g on an 0.01 g scale. At 0.5% that means a batch of at least 20 g before the addition is weighable at all, and comfortably more before it is reliable. The fix for smaller batches is a premix. Dissolve one part essential oil in nine parts of a stable carrier such as fractionated coconut or jojoba, label it with the date and the ratio, and weigh the premix instead: your effective precision improves tenfold, and a 0.5% addition to a 20 g batch becomes a 1.0 g weighing rather than a 0.1 g one. Scale calibration and the readability trap are covered in weighing and calibration.
Then set the target rather than discovering it. Aim for 0.5 to 1% total fragrance in a lip product and 1 to 2% in a body balm, treat those as ceilings you work down from, and check each individual oil against its restricted constituents before you commit. Add below 50 to 60 C so you are not evaporating what you just paid for, and write the actual weighed mass into the batch record, not the intended one. If flavour rather than scent is what you were chasing on a lip product, the materials are different and the levels are different too: see lip balm flavour and sweeteners. And a balm with no fragrance at all is not a lesser product, as the fragrance free formula sets out.
The decision rule
Run it in this order. Reweigh the batch on paper. If any constituent is above its limit, dilution is mandatory and the batch cannot be sold as it stands, whatever it smells like. If everything is inside the limits and you simply dislike it, smell it on skin before you touch the pot, wait 48 hours, and ask someone whose nose has not been in the workshop all day. Only if it still reads too strong under those conditions is it worth remelting, and then dilute in one move to a target you have already written down.
The honest limit of this page is that there is no objective threshold at which a balm becomes too strong. Odour perception varies enormously between people, adaptation happens within seconds, and there is no published figure for a preferred fragrance level in an anhydrous product. What is objective is the ceiling: the constituent limits, the phototoxicity limits and the active dose windows are numbers, they are checkable, and they are where the word "too" actually means something.
Frequently asked questions
How many drops of essential oil should go in a 30 g balm?
The question cannot be answered in drops, because a drop weighs anywhere from about 20 to 50 mg depending on the dispenser and the oil. Work in grams: 1% of a 30 g balm is 0.30 g, and 0.5% is 0.15 g. Weigh that on a scale reading to 0.01 g, or make a one in ten premix in carrier oil and weigh 3.0 g of that instead.
Why does my balm smell strong in the tin but faint on my skin?
Because a sealed tin accumulates volatiles in its headspace for hours and releases them all at once when you open it, at a concentration no skin film reaches. On skin the same balm is a fraction of a gram spread over a large area in open air, and the lightest molecules leave within minutes. Judge the product on the back of the hand, not at the lid.
Can I fix an over-fragranced balm without remelting the whole thing?
Not properly. Airing the open tin loses a little top note from the surface and nothing from the body of the balm, so the smell returns as soon as the lid goes back on and the headspace refills. The only real fix is blending with unscented base, and that needs both parts molten and stirred together for a full minute to avoid tins of two different strengths.
Will the smell fade if I just leave it alone?
Partly, and predictably unevenly. Light citrus and herbaceous materials fade over weeks while heavier woods, resins and vanillins stay, so the balm gets quieter and also changes character. That is not a plan for a batch that is above a regulatory limit, because the restricted constituent is still there in the same amount whether you can smell it or not.
Is a strongly scented balm actually dangerous?
Not by smell alone, but the level often is. The limits that bind are per constituent, not per oil: citral is capped at 0.11% in lip products, isoeugenol at 0.019%, and methyl salicylate at 0.06% leave-on under EU Annex III. A lip balm at 4% lemongrass sits about twenty-five times over the citral limit, so the smell and the overdose are the same fact.
What percentage of essential oil is right for a lip balm?
As a working target, 0.5 to 1% total, treated as a ceiling to work down from. That is a formulation window rather than a legal limit. Lip products are IFRA Category 1, the strictest category, and what actually binds is the limit on each restricted constituent in each oil, plus a summation rule if you use more than one.
Why is the "20 drops per gram" rule wrong?
Because it is the pharmacopoeial figure for water from a 3 mm dropper. Drop mass depends on surface tension, and essential oils are around 26 to 30 mN per metre against water's 72, so the same orifice delivers roughly 40% of the mass. A typical essential oil runs nearer 48 drops per gram, and a thin citrus oil nearer 55.
Sources and further reading
- European Directorate for the Quality of Medicines, European Pharmacopoeia general chapter 2.1.1, Droppers, Strasbourg, for the standard dropper and the 20 drops per gram of water specification.
- Harkins, W. D. and Brown, F. E., The determination of surface tension and the weight of falling drops, Journal of the American Chemical Society, 41(4):499-524, 1919, for the drop weight correction factor.
- International Fragrance Association, IFRA Standards, 51st Amendment, product category definitions and restricted constituent limits.
- European Parliament and Council, Regulation (EC) No 1223/2009 on cosmetic products, Annexes II and III, consolidated text, including entry 324 on methyl salicylate.
- European Commission, Regulation (EU) 2023/1545 amending Regulation (EC) No 1223/2009 as regards labelling of fragrance allergens, 2023.
- Love, J. N. and colleagues, Are one or two dangerous? Camphor exposure in toddlers, Journal of Emergency Medicine, 27(1):49-54, 2004, PMID 15219304.
- European Commission Scientific Committee on Consumer Safety, Opinion on fragrance allergens in cosmetic products (SCCS/1459/11).
Reviewed and updated 6 September 2026. Spotted an error? Tell us and we will fix and log it.