How to Burn Test a New Fragrance Oil Before You Sell It
Share
How long before I can judge a new oil? Roughly three weeks from pour to decision. Some CSI oil pages give a cure of 24 to 48 hours, and that is the earliest a candle can be lit at all — but the Luxury Soy Wax Chunks page itself asks for a minimum of 48 hours and ideally one to two weeks before burn testing, and 14 to 21 days is standard practice for soy container candles. Pour on day zero, gate at day fourteen, burn four cycles across four days, decide on day eighteen.
What counts as a pass? A melt pool that reaches the glass by the end of the second or third four-hour burn, a pool that stays shallow rather than deepening past roughly 10 to 13 mm, a flame that stays small and steady, no soot on the glass, a jar you can still hold, and a scent you can smell from the far side of the room you sell into. Any one of those failing is a fail — you change one variable and burn again.
Can I skip this if the oil is already in my range in another wax? No, and this is the most expensive shortcut in the business. The test approves a combination: oil, wax grade, load, jar and wick together. Change the wax from CSI 464 to CSI 468, change a 150 g jar for a 70 g votive, move from Thin (C1) to Thick (C2) at ₹9.44, or move from 8% to 10%, and you are back to a fresh set of three jars.
What a burn test actually tests
A fragrance oil has no burn behaviour of its own. What burns is a system: a wax grade with its own melt behaviour, a load of oil dissolved into it, a wick sized for a diameter, a jar with a wall thickness and a mouth, and a room with a temperature and an air current. Change any part and the result moves. This is why “is this a good oil” is not a question a test can answer, and why “does this oil work in my 150 g jar, in CSI 464, at 8%, on a Thin (C1) wick” is.
It is also why the load has to be written down as a percentage of wax weight, the way the CSI pages state it. 12 g in a 150 g fill is 8%. The same 12 g in a 200 g fill is 6%, and it will not smell the same. Two ceilings apply at once: the wax grade’s stated maximum and the oil page’s stated range, and the lower of the two governs. Luxury Soy CSI 464 states up to 10%; British Rose states up to 10% for soy; so 10% is the ceiling for that pair and 8% is a sensible place to start. Put Japanese Cherry Blossom, whose page states 6–13%, into the same wax and the wax still holds you to 10%.
The last thing a test does, and the one that matters commercially, is give you a record. When a stockist in Bengaluru asks how long your candles burn, or a corporate buyer asks whether the candle is safe on a wooden table, the answer is a sheet of paper with dates, times, pool measurements and a lot number on it. That sheet is worth more than any adjective in your product description, and it is a by-product of testing properly.
The rig: three jars per load, one variable
Two jars is not enough, because the jar you burn is destroyed as evidence and the jar you sniff is opened. Three is the working minimum, and each has a different job.
| Jar | What it does | What it tells you | What it costs at 8% |
|---|---|---|---|
| A — the burn jar | Burns four cycles of four hours across four consecutive days, wick trimmed to about 5 mm before each light | Pool width and depth, flame behaviour, soot, mushrooming, glass heat, hot throw over the life of the candle | ₹213.23 |
| B — the cold jar | Stays lidded and unlit until day fourteen, then is opened once for a cold-throw reading | Whether the candle sells itself on the shelf before it is lit — and, opened once, it is still usable as a second burn jar if A fails early | ₹213.23 |
| C — the retain sample | Sealed, labelled with oil, lot, load, wax, wick, jar and date, and never opened during the test | What you actually approved, months later, when a customer or a new batch raises a question | ₹213.23 |
Pour all three from one pot, in one session, at one temperature. If you melt two batches you have introduced a second variable and you will never be sure which one moved. Weigh the oil on a scale rather than counting drops — any scale reading to a gram will do, and CSI’s own Candle Making Weighing Scale at ₹354.00 is down to its last couple of units — and add at the temperature the wax page asks for. The Luxury Soy Wax Chunks page asks for fragrance at 80–90°C with slow stirring for two full minutes; Eco Soy CSI 400 asks for 85°C and two minutes of gentle stirring. A Pen Thermometer at ₹354.00 is what turns that instruction into something you actually did.
Label the jars before you pour, not after. Masking tape on the base, in pen: oil, load, wax, wick, jar, date, and your initials if more than one person works in the room. Three unlabelled jars on a shelf at day fourteen are three jars you have to throw away.
The schedule, from pour to decision
Eighteen days is the shape of it, and the only part that feels wasteful is the part that matters most: the fortnight of doing nothing. A soy candle read at day two is a candle whose fragrance is still finding its level, and a decision taken there is a decision you will revisit.
You can compress this if a festival forces you to, but compress the burns rather than the cure: three burns instead of four still shows you the second-burn pool, which is where most wick faults appear. Skipping the fortnight does not save you two weeks, it just moves the discovery of the problem into your customer’s living room.
What to record, and what each bad reading means
Nine columns, one row per burn, one sheet per jar. Write it while the candle is burning, not afterwards — at day seventeen you will not remember whether burn two reached the glass or nearly reached it, and “nearly” is the whole question.
| Record | How to take it | What a bad reading usually means |
|---|---|---|
| Burn number and clock time | Light and extinguish times, to the minute | Short or uneven burns make every other column unreadable |
| Melt pool width | How far the liquid has reached across the jar at the end of the burn, in mm | Not reaching the glass by burn two or three: wick too small, or the load is stiffening the wax |
| Melt pool depth | Against a skewer, in mm | Deepening past roughly 10–13 mm: wick too large. Very shallow with a wide pool: usually fine |
| Flame height and steadiness | Eye measure against a ruler behind the jar | Tall, flaring or dancing flame: wick too large, a draught, or a fan overhead |
| Mushrooming on the wick tip | Look before each trim | A carbon ball every burn: wick too large for the load, or the load is at the top of the range |
| Soot on the glass | Look at the upper inside wall after cooling | Any soot at four hours is a fail. Re-wick down one size and retest |
| Jar temperature | Back of the hand on the glass at 3 h 45 min | Too hot to hold: wick too large for this jar. This is a safety reading, not a comfort one |
| Hot throw, scored 0–5 | At 1 m and at 3 m, in a stated room size, at the end of hour two and hour four | A 4 at 1 m and a 1 at 3 m is a jar-candle problem, not an oil problem — check the pool first |
| Room temperature and air movement | Thermometer reading and fan setting | A 34°C room and a fan on 2 will flatter the pool and punish the flame. Record it or the test is not repeatable |
Two readings deserve a comment. The first is the jar temperature: this is the closest thing to a safety measurement a small maker takes, and it is the reason a test is not optional. The second is the throw score. Score it before you discuss it with anyone else in the room, because the second opinion always drags the first towards it, and write the room size next to it — a 3 in a 3 by 4 metre bedroom is not the same product as a 3 in an open-plan living room.
Pass and fail: the criteria, written before you start
Set these before the first light, because criteria invented at day seventeen are criteria bent to fit a result you have already paid for. The figures below are general candle-making practice for soy container candles rather than CSI test data, and they are a starting point you should tighten to your own range.
| Reading | Pass | Fail | What to change first |
|---|---|---|---|
| Pool reach | Reaches the glass by the end of burn two or three | Still tunnelling at burn three | Wick up one size — Thin (C1) to Thick (C2) at ₹9.44 — before you touch the oil |
| Pool depth | Settles around 10–13 mm and stays there | Keeps deepening, or liquid to the base by burn three | Wick down, or reduce the fill in a narrow jar |
| Flame | Small, steady, roughly 25–40 mm, no flare on relight | Tall, flickering, or visibly tilting | Wick down; also check for a draught or a ceiling fan |
| Soot | None on the glass at four hours | Any grey film or black tip on the inside wall | Wick down one size and retest. Do not sell a sooting jar |
| Glass temperature | Warm, holdable at the end of the burn | Too hot to hold comfortably | Wick down, or move to a thicker-walled jar |
| Cold throw at day 14 | Clearly identifiable with the lid off at arm’s length | Faint or ambiguous | Try the higher load within the lower of the two ceilings, then retest |
| Hot throw at hours 2 and 4 | Scores 3 or better at 3 m in your stated room | Under 3 at 3 m with a full, correct pool | This is now an oil or load question, not a wick one |
| Surface after cooling | Acceptable to your own finish standard, consistent across all three jars | Wet spots, sink holes or frosting that differ between jars | A pour or cooling fault, not an oil fault — fix the process and repour |
Notice the order of operations built into that last column. When something is wrong, the wick is almost always the first thing to change and the oil is almost always the last, because the wick is the cheapest variable in the jar at ₹5.90 and the oil is the most expensive. Change one thing, pour three more jars, and keep the old sheet beside the new one.
What the test costs, and what it protects
The reason most makers skip testing is not laziness, it is that six jars feel like six jars you cannot sell. Put the two numbers side by side and the argument ends.
| Item | Unit | Count | Total |
|---|---|---|---|
| Jar — Clear Glass with Golden Lid, 150 g fill | ₹94.40 | 6 | ₹566.40 |
| Wax — Luxury Soy CSI 464, 150 g a jar | ₹76.11 | 6 | ₹456.66 |
| Wick — Eco Candle Wicks Thin (C1) | ₹5.90 | 6 | ₹35.40 |
| Oil at 8% — 12 g of British Rose | ₹36.82 | 3 | ₹110.45 |
| Oil at 10% — 15 g of British Rose | ₹46.02 | 3 | ₹138.06 |
| Six-jar test, two loads | — | 6 jars | ₹1,306.97 |
| The run it protects — fifty candles at 8% | ₹213.23 | 50 | ₹10,661.30 |
The six-jar test is about an eighth of the fifty-candle run. It is also less than the cost of one unhappy stockist, one refunded corporate order, or one Instagram reel in which a customer films a jar that has sooted its own glass black by the third burn. And the six jars are not waste: jars A and B become in-house testers, market-stall demonstration candles or gifts for the people who keep giving you feedback. Only jar C stays untouched, and it is the one with the most value in it.
If cash is genuinely tight, shrink the test rather than dropping it. Three jars at one load is ₹639.68; a 50 gm bottle of oil at ₹212.40 covers the 36 g those three jars need for ₹152.93 with a little left for blotters; and a 15 gm bottle at ₹94.40 is enough for a single 150 g jar at 8% with change. What you must not do is test one jar and call it a protocol.
Testing in Indian conditions, and testing for a festival
Every number in a burn test is a number taken in a room, and Indian rooms vary more than most test protocols admit. A 34°C workshop in Chennai gives a wider, faster pool than a 19°C December morning in Shimla, on the same wick, with the same wax. If you test only in one season you will discover the other one through complaints.
Three habits fix most of it. Record the room temperature with every burn, so a set of readings can be compared with another set. Run at least one burn under the conditions your customers actually use — which, in most of India, means a ceiling fan running somewhere in the room, not a sealed studio. And if your range sells into both an air-conditioned Mumbai flat and an open Jaipur courtyard in December, be honest in your own notes about which one the candle was approved for.
A fan deserves its own line, because it is the single most common cause of a test that disagrees with reality. Moving air pulls the flame sideways, deepens the pool on one side and is a frequent cause of soot on the glass and of a candle burning down one wall. That is not a fault in the oil and no amount of load adjustment will fix it. Test with the fan off to get a clean reading of the wick, then test again with it on low to see what your customer will see.
Now the calendar, because this is September. Diwali falls on 8 November 2026, and with a 14 to 21 day cure plus courier time, festive candles want to be poured by about mid-October. Work backwards through the protocol: eighteen days from pour to decision means a new oil you want in the festive range has to go into test jars in the second half of September. An oil that arrives in your workshop in the second week of October cannot be tested and sold this season — it can be tested for next year, which is a perfectly good use of six jars and a great deal better than launching it untested into your busiest fortnight.
When you have to test the same oil again
An approval is for a combination, not for a bottle. Retest when any of these changes: the wax grade, the jar, the wick size, the fill weight, the load, or the oil itself — including a new lot of the same oil if the last one was bought a long time ago. A new lot does not need the full eighteen days if nothing else has moved; one jar, the same fourteen-day gate and two burns against the old sheet will tell you whether it behaves like the one you approved.
Retest also when the reason for a complaint is unknown. A jar that tunnels for a customer and not for you is usually the room, the wick or a trimming habit, but the way to find out is to reproduce it: pour three, burn them in conditions closer to theirs, and compare with the retain sample you kept. That sealed jar C is why this protocol asks you to make three instead of two — it is the only fixed point you have.
Finally, keep the paperwork with the test. The record sheet, the oil lot, and the IFRA and MSDS documentation CSI sends on request belong in the same folder, named for the product. When a buyer with a procurement checklist asks what you hold, you want to be the maker who attaches a file rather than the one who promises to look for it.
CANDLEMAKINGSUPPLIESINDIA supplies raw materials rather than finished candles, and the honest thing to say on a page like this is that a burn test regularly fails the oil you just bought — and that when it does, the first thing to change is the wick at ₹5.90, not the fragrance at ₹36.82 a jar. A supplier that tells you to solve every throw problem by buying more oil is selling, not advising. Test jars are also not a guarantee: they are evidence, and evidence is only as good as the notes you wrote while it burned.
Every price on this page comes from live CSI stock in September 2026 and the linked product pages are authoritative. The wax working figures quoted are the ones the pages print: Luxury Soy CSI 464 states up to 10% fragrance load with 80–90°C given as the working range for melting and pouring; Luxury Soy Wax Chunks states up to 13% of total wax weight, fragrance at 80–90°C with slow stirring for two full minutes, and a cure of a minimum of 48 hours and ideally one to two weeks before burn testing; Eco Soy CSI 400 asks for fragrance at 85°C. Wick sizing by jar diameter is not published per wick, which is precisely why this page is a test protocol rather than a chart. Burn durations, pool depths, flame heights and scoring bands are general candle-making practice, stated as such, and not CSI test data.
For bulk pricing, GST invoicing, MSDS and IFRA documentation, or help working out which trial sizes a test batch needs, message CSI on WhatsApp at +91 7397976926. Send your jar, your wax grade and your pour date and you will get a parts list rather than a catalogue.
Frequently asked questions
- How to Test Hot Throw Before Launching a Candle Fragrance
- How to Compare Two Fragrance Oils Using a Controlled Burn Test
- Why Your Candle Needs A 14-Day Cure
- Wrong Wick Size Is Killing Your Hot Throw
- Why Does My Candle Tunnel Even With The Correct Wick?
- Why Is My Candle Wick Mushrooming? The Wick Sizing and Fragrance Load Fix
- Best Wick for Candle Making: Complete Guide by Diameter
- Why Your Candle Smells Strong Cold but Weak When Lit (Real Reasons & Fixes)
- Top Fragrance Oils for Candles: Throw, Load, and Cure Science
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. One longer review is shortened with an ellipsis; wording is otherwise unedited. One is below five stars and four are not recorded as a verified purchase — shown as recorded. Not recorded by Judge.me as a verified purchase, so shown without a verified-buyer badge: Ashwini, Lalitha Jagan, LJ, Ashwini. No location is shown against any review because Judge.me does not store one.
Figures verified September 2026 (CSI live pricing). Test unit used throughout: one Clear Glass Jar with Golden Lid at ₹94.40 from the pack of 10 at ₹944.00 (page states 150 g fill capacity, 8 cm height, hot-pour safe to 85°C), 150 g of Luxury Soy Wax CSI 464 at ₹76.11 from the 1 kg pack at ₹507.40, one Eco Candle Wicks Thin (C1) at ₹5.90 from the 100-pack (Thick C2 ₹9.44), and British Rose at ₹3.07 a gram from the 1 kg pack at ₹3,068.00. Loads: 12 g is 8% of 150 g (₹36.82), 15 g is 10% (₹46.02); one jar ₹213.23 at 8% and ₹222.43 at 10%; three jars at 8% ₹639.68; six jars across two loads ₹1,306.97; fifty candles at 8% ₹10,661.30. Trial sizes: British Rose 15 gm ₹94.40, 50 gm ₹212.40 (₹4.25/g), so 36 g for three jars is ₹152.93. Other items priced: Clear Glass Votive Jar 70 gram ₹31.86 each from the pack of 10, Amber Shiny Glass Jar with Black Lid ₹47.20 (page gives 120 gm capacity, 6 cm diameter, 6.5 cm height, so a full pool of about 28.3 cm²), Pen Thermometer ₹354.00, Candle Making Weighing Scale ₹354.00, Wick Centering Device ₹118.00, Candle Wick Stickers ₹2.50 each from the 40-sticker sheet. Page wording quoted: Luxury Soy CSI 464 states up to 10% fragrance load and gives 80–90°C as the working range for melting and pouring; Luxury Soy Wax Chunks states up to 13% of total wax weight, fragrance at 80–90°C with slow stirring for two full minutes and a cure of a minimum of 48 hours, ideally one to two weeks before burn testing; Eco Soy CSI 400 asks for fragrance at 85°C with two minutes of gentle stirring; British Rose states up to 10% for soy and 0% vanillin; Japanese Cherry Blossom states 6–13%. Burn durations, 5 mm trim, pool depth bands of roughly 10–13 mm, flame heights of roughly 25–40 mm and the 0–5 throw scoring are general practice, not CSI measurements. Batch totals and the fifty-candle run are illustrative. Diwali 2026 is 8 November 2026. Prices and availability change — the linked product pages are always authoritative.