My Custom Blend Performs Well at 8% but One Component Discolours the Wax — How Do I Identify Which Fragrance Is Causing It?
Share
Can you tell me which oils discolour? No, and nobody should guess. CSI publishes no composition or discolouration data per oil. The single-component test on this page answers it for your wax, your storage and your light. See why a test beats a list.
What if no single oil discolours but the blend does? Then it is an interaction between components, or the extra load. Test the three pairs. If the unscented control also changes, the cause is the wax or the process, not the fragrance. See reading the results.
How do I fix it without losing the blend's performance? Lower the culprit's share and rebalance, replace it with a similar oil that tests clean, colour the wax on purpose, or protect the candle from light. Re-test performance after any change. See the fixes.
Why a test beats a list
Discolouration in candles is real and common. Wax that was cream at pouring turns yellow, tan, pinkish or brown over days or weeks, or fades unevenly where light reaches it. In general candle-making practice, it is commonly linked to certain aroma materials in fragrance oils, to heat during production, to light and air during storage, and to interactions with dyes. But the same oil can behave differently in different waxes, at different loads and under different storage, and a fragrance house can reformulate an oil without changing its name.
That is why this page does not tell you which oils discolour. CSI publishes no composition, vanillin or discolouration data for any oil, and we will not guess. The example blend below is an example of a structure, not a suspect list: nothing in our data says Mango Coconut, Honeydew Melon or Fresh Strawberries changes colour. What we can give you is a method that answers the question for your blend in about four weeks, for the cost of a few jars and some oil. For documents on any oil, the MSDS and IFRA certificate are available on WhatsApp; they are safety and compliance documents, not colour predictions.
| Possible driver | What it usually looks like | Which control in this protocol isolates it |
|---|---|---|
| A fragrance component | Jars containing one oil shift; more of it, more shift | Singles at share and at 8%, leave-one-out |
| An interaction between components | Only combinations shift, no single oil does | Full blend vs singles; then pairs |
| Light exposure | Faces towards the window change first, or light jars more than dark | Daylight duplicates vs dark set |
| Wax lot or process heat | Even the unscented jar shifts | Unscented control |
| Dye reacting or fading | Only dyed candles change | Add a dyed unscented control if you colour |
The single-component protocol
All jars are 100 g of wax plus oil, poured from one lot of wax, on one day, with one process. The blend is 50% A, 30% B and 20% C at 8% total load, so a 100 g jar holds 8 g of blend: 4.0 g A, 2.4 g B and 1.6 g C (CSI's convention: load is a share of the finished weight). The singles-at-share jars hold exactly those grams of one oil, so they show what that oil does at its real concentration. The singles-at-8% jars exaggerate it, which makes a slow effect visible sooner. The leave-one-out jars keep the other two oils at their blend grams and simply omit one.
| Jar | What it is | A | B | C | Total oil | Wax | Stored |
|---|---|---|---|---|---|---|---|
| 1 | Unscented wax | — | — | — | 0.0 g | 100.0 g | dark |
| 2 | Full blend at 8% | 4.0 g | 2.4 g | 1.6 g | 8.0 g | 92.0 g | dark |
| 3 | A alone at its blend share | 4.0 g | — | — | 4.0 g | 96.0 g | dark |
| 4 | B alone at its blend share | — | 2.4 g | — | 2.4 g | 97.6 g | dark |
| 5 | C alone at its blend share | — | — | 1.6 g | 1.6 g | 98.4 g | dark |
| 6 | A alone at 8% | 8.0 g | — | — | 8.0 g | 92.0 g | dark |
| 7 | B alone at 8% | — | 8.0 g | — | 8.0 g | 92.0 g | dark |
| 8 | C alone at 8% | — | — | 8.0 g | 8.0 g | 92.0 g | dark |
| 9 | Blend without A | — | 2.4 g | 1.6 g | 4.0 g | 96.0 g | dark |
| 10 | Blend without B | 4.0 g | — | 1.6 g | 5.6 g | 94.4 g | dark |
| 11 | Blend without C | 4.0 g | 2.4 g | — | 6.4 g | 93.6 g | dark |
| 12 | Unscented wax | — | — | — | 0.0 g | 100.0 g | light |
| 13 | Full blend at 8% | 4.0 g | 2.4 g | 1.6 g | 8.0 g | 92.0 g | light |
| 14 | A alone at 8% | 8.0 g | — | — | 8.0 g | 92.0 g | light |
| 15 | B alone at 8% | — | 8.0 g | — | 8.0 g | 92.0 g | light |
| 16 | C alone at 8% | — | — | 8.0 g | 8.0 g | 92.0 g | light |
Two design choices matter. First, the leave-one-out jars do not top the other oils back up to 8%. Keeping their grams identical to the full blend means the only difference between jar 2 and jar 9 is the absence of A. Second, the daylight set duplicates only the jars that carry the most information: unscented, full blend and each single at 8%. That separates "this oil changes colour over time" from "this oil changes colour in light", which need different fixes.
Watching and scoring colour
Colour is easy to misjudge by eye, especially when you are looking for it. Always judge against the unscented control from the same storage, on a white background, and score each jar as a difference from that control, not on its own. A phone photograph at every checkpoint, taken in the same place, gives you a record you can compare side by side later.
| Jar | Day 1 | Day 3 | Day 7 | Day 14 | Day 28 | Colour seen (yellow / tan / pink / brown / faded) |
|---|---|---|---|---|---|---|
| 1 · Unscented wax | ||||||
| 2 · Full blend at 8% | ||||||
| 3 · A alone at its blend share | ||||||
| 4 · B alone at its blend share | ||||||
| 5 · C alone at its blend share | ||||||
| 6 · A alone at 8% | ||||||
| 7 · B alone at 8% | ||||||
| 8 · C alone at 8% | ||||||
| 9 · Blend without A | ||||||
| 10 · Blend without B | ||||||
| 11 · Blend without C | ||||||
| 12 · Unscented wax (light) | ||||||
| 13 · Full blend at 8% (light) | ||||||
| 14 · A alone at 8% (light) | ||||||
| 15 · B alone at 8% (light) | ||||||
| 16 · C alone at 8% (light) |
Also note where the colour appears. Change at the top surface first suggests air or light. Change on the side facing the window points to light. Uniform change through the whole jar in the dark set points to time. Change only near the wick hole of a burned candle is a different problem, related to heat, and is not what this protocol tests; if that is your issue, add burned versions of the blend and singles.
Reading the results
Read the pattern across the jars, not any single jar. Most results fall into one of five patterns:
| Pattern | Meaning | Next step |
|---|---|---|
| Every jar with one component changes, most at 8%, and the blend without it matches unscented | That component is the culprit | Go to the fixes |
| Two components each change a little on their own | Two contributors, adding up in the blend | Fix the stronger one first, re-test |
| No single changes; the full blend does; leave-one-out jars are partly changed | Interaction between components | Pour the three pairs (A+B, A+C, B+C) at blend grams |
| Unscented control changes too | Wax lot, storage or process heat | Check wax lot, melting temperature and storage; fragrance may be innocent |
| Only daylight jars change | Light-driven change | Protect from light: storage, jar choice, packaging |
The single-at-8% jars are the most sensitive, so an effect often shows there first. Do not convict a component on its 8% jar alone, though: an oil that yellows slightly at 8% may do nothing visible at the 1.6 g it contributes to the blend. The singles-at-share and leave-one-out jars tell you whether it matters at the concentration you actually use. Checking physical compatibility between components covers separation and haze, which can look like colour change in a clear jar.
Fixing it without losing performance
Your blend performs well at 8%, so every fix should change as little as possible and be re-tested for performance. In order of how little they disturb the blend:
| Fix | What you change | Performance risk | Re-test |
|---|---|---|---|
| Protect from light | Storage, jar colour or packaging | None | Daylight set only |
| Colour the wax deliberately | Add a dye that hides or suits the shift | None to scent; check dye dose on its page | Dyed jars, dark and light |
| Lower the culprit's share | e.g. A from 50% to 40%, raising B and C | Moderate: the scent shifts | Colour + blind scent test |
| Replace the culprit | A similar-direction oil that tests clean | Higher: new balance | Full blend ladder + colour |
| Accept and disclose | Tell customers colour may vary naturally | None | None |
If you lower the culprit's share, keep the total at 8% and rebalance: for a 200 g candle (16 g of oil), moving A from 50% to 40% and letting B and C share the difference in their existing 3:2 proportion gives B 36% and C 24%: 6.40 g A, 5.76 g B and 3.84 g C. Test it with the small-step method in I Want to Make My Existing Candle Fragrance Stronger by Blending in Another Fragrance Oil — How Should I Test the New Combination Without Ruining the Original Profile?, scored for both colour and scent. If you replace the culprit, run the replacement through this protocol before it goes into a blend, then re-ladder the blend; I Have a Fragrance That Is Too Sweet — Can I Blend It with Woody or Fresh Fragrance Oils to Make It More Sophisticated for Candles? and I Have a Floral Fragrance That Smells Flat in Candles — What Type of Fragrance Oil Can I Blend with It to Improve Perceived Freshness and Diffusion? show ladders with real CSI oils.
Colouring the wax on purpose is an honest and common fix when the shift is mild and even. A warm tan or amber tone can make a slight yellowing invisible and suits many gourmand and woody blends. CSI's liquid dyes include Brown at ₹236.00 and Red at ₹236.00 for 15 g; follow the dye's product page for dose, since CSI does not publish dose rates per kilo. Accepting and disclosing is also legitimate: many makers print "natural colour variation may occur" on the label. What matters is that the customer is not surprised.
| Pack | Price | Per gram | Diagnostic sets covered |
|---|---|---|---|
| 15 g | ₹93.22 | ₹6.21 | 0.4 |
| 50 g | ₹250.00 | ₹5.00 | 1.4 |
| 100 g | ₹500.00 | ₹5.00 | 2.8 |
| 500 g | ₹1,982.00 | ₹3.96 | 13.9 |
| 1 kg | ₹3,964.00 | ₹3.96 | 27.8 |
| Pack | Price | Per gram | Diagnostic sets covered |
|---|---|---|---|
| 15 g | ₹105.02 | ₹7.00 | 0.5 |
| 50 g | ₹217.12 | ₹4.34 | 1.8 |
| 100 g | ₹435.42 | ₹4.35 | 3.6 |
| 500 g | ₹1,699.20 | ₹3.40 | 17.9 |
| 1 kg | ₹3,398.40 | ₹3.40 | 35.7 |
| Pack | Price | Per gram | Diagnostic sets covered |
|---|---|---|---|
| 15 g | ₹113.28 | ₹7.55 | 0.6 |
| 50 g | ₹212.40 | ₹4.25 | 2.1 |
| 100 g | ₹377.60 | ₹3.78 | 4.2 |
| 500 g | ₹1,770.00 | ₹3.54 | 20.8 |
| 1 kg | ₹3,540.00 | ₹3.54 | 41.7 |
Shopping list: a 16-jar diagnostic set
Oil use across all 16 jars is 36.0 g of A, 28.0 g of B and 24.0 g of C. A 50 g of each covers that. For B and C, a 50 g is also cheaper than enough 15 g testers: Honeydew Melon 50 g is ₹217.12 against two 15 g at ₹210.04 (only 30 g), and Fresh Strawberries 50 g is ₹212.40 against two 15 g at ₹226.56. Wax is 1,512.0 g from a single lot; the 1 kg and 500 g bags cover it with a margin. No wicks are needed.
| Item | Quantity | Line total |
|---|---|---|
| Mango Coconut 50 g | 36.0 g needed (component A) | ₹250.00 |
| Honeydew Melon 50 g | 28.0 g needed (component B) | ₹217.12 |
| Fresh Strawberries 50 g | 24.0 g needed (component C) | ₹212.40 |
| Luxury Soy Wax Chunks 1 kg + 500 g | 1,512.0 g needed (same lot) | ₹898.00 |
| Yankee Glass Jar 130 ml with clear cap × 16 | 16 clear test jars with caps, 100 g each | ₹755.20 |
| Candle Making Weighing Scale | 1 | ₹354.00 |
| Candle Wax Pouring Jug | 1 | ₹531.00 |
| Bamboo Stirring Sticks pack of 5 | 1 | ₹118.00 |
| Total | ₹3,335.72 |
Without the scale, jug and sticks, the consumables come to ₹2,332.72. If you already hold the three oils, the test costs only the wax and jars. For a two-oil blend, drop the C jars (jars 5, 8, 11 and 16): 12 jars in all. The Yankee jars come with clear caps, which keeps air exposure the same for every jar in the dark set.
CANDLEMAKINGSUPPLIESINDIA supplies raw materials, not finished candles, and it does not sell a blending service. CSI sells single fragrance oils; any blend you build from them is your formula. This page deliberately names no oil as one that discolours: CSI has no data that would support such a claim, and a list would send you to replace the wrong component. The method finds it in your wax instead.
Every oil and wax price on this page is CSI live pricing, pulled on 24 September 2026. Wick, jar, scale and thermometer prices are those verified for Blog 760 on 10 September 2026; the jug, stirring sticks and other equipment were pulled on 24–25 September 2026. Out-of-stock sizes are flagged, never offered as buyable. Grams, ratios, per-gram and per-candle figures are plain arithmetic on those prices, shown to two decimals so you can check them.
CSI publishes no throw, cure-time, composition or discolouration data for any individual oil, so nothing here ranks oils by performance. Test ratios, cure checkpoints and tolerances are labelled working assumptions for your own trials, not specifications. For MSDS and IFRA documentation, a restock date, or a quote on a production quantity of any component, message us on WhatsApp at +91 7397976926.
Frequently asked questions
- I'm Combining Three Candle Fragrance Oils from Different Suppliers — How Do I Know Whether They Are Physically Compatible in the Same Wax?
- My Custom Fragrance Blend Causes Sweating Even Though Each Fragrance Works Individually at the Same Load — Why?
- I Want to Make My Existing Candle Fragrance Stronger by Blending in Another Fragrance Oil — How Should I Test the New Combination Without Ruining the Original Profile?
- I Want to Create My Own Signature Candle Fragrance Using Ready-Made Fragrance Oils — How Many Components Should I Start with Before Making the Blend Too Complicated?
- I'm Scaling a Custom Fragrance Blend from 100 g to 5 Kg — How Do I Preserve the Exact Ratios and Minimise Weighing Errors?
- How Do I Document a Custom Candle Fragrance Formula so My Production Team Can Reproduce the Same Scent and Strength Every Time?
- I'm Switching Fragrance Oil Suppliers to Reduce Cost — Which Cold Throw, Hot Throw, Colour, Wick and Stability Tests Should I Repeat Before Production?
- My Fragrance Supplier Sent a New Batch of My Bestselling Scent — What Should I Compare Against the Previous Batch Before Making Hundreds of Candles?
Customer reviews: The six reviews at the top of this page are genuine, published Judge.me reviews left by CSI customers on the product pages named on each card — Eco Candle Wicks, Premium Candle Wicking Needle and Bamboo Wick Holder. They are general product reviews, not assessments of the guidance set out here. Names, star ratings, dates and products are as recorded by Judge.me and wording is unedited. "Verified buyer" appears only on the three reviews Judge.me recorded as verified — Monika Deep, Subhankar Roy and Sneha Tamang; the reviews from Ashwini, LJ and Lalitha Jagan are published but not recorded as verified, and carry no badge.
Figures pulled 24 September 2026, CSI live pricing: Fragrance oils (size by size, as quoted on this page): Mango Coconut ₹93.22 / 15 g, ₹250.00 / 50 g, ₹500.00 / 100 g, ₹1,982.00 / 500 g, ₹3,964.00 / 1 kg; Honeydew Melon ₹105.02 / 15 g, ₹217.12 / 50 g, ₹435.42 / 100 g, ₹1,699.20 / 500 g, ₹3,398.40 / 1 kg; Fresh Strawberries ₹113.28 / 15 g, ₹212.40 / 50 g, ₹377.60 / 100 g, ₹1,770.00 / 500 g, ₹3,540.00 / 1 kg; Eucalyptus Spearmint ₹68.44 / 15 g. Wax, equipment and other catalogue items pulled 24–25 September 2026: Luxury Soy Wax Chunks ₹599.00 / 1 kg; Luxury Soy Wax Chunks ₹299.00 / 500 g; Yankee Glass Jar 130 ml with clear cap ₹47.20 / each; Candle Wax Pouring Jug ₹531.00 / each; Bamboo Stirring Sticks ₹118.00 / pack of 5; Candle Dye Liquid Brown ₹236.00 / 15 g; Candle Dye Liquid Red ₹236.00 / 15 g. Figures verified 10 September 2026 (Blog 760): Candle Making Weighing Scale ₹354.00 / each.
Non-price figures and assumptions: Load is a share of the finished weight (oil = W × L): a 100 g jar at 8% holds 8 g of blend (A 4.0 g, B 2.4 g, C 1.6 g); leave-one-out jars keep the other components at their blend grams. The 16-jar design, the day 1/3/7/14/28 checkpoints, the 0–3 scoring scale and the two-minute stir are working practices, not specifications. Causes of discolouration listed are general candle-making practice, not CSI measurements. Yankee 130 ml fill weight is not published; 100 g per jar is a test pour. CSI's CSI 464 guidance is 80–90°C; confirm on the product page, and follow the wax maker's instructions for other waxes. Dye dose rates are not published by CSI; follow the dye's product page. Per-candle oil cost is at 100 g rates and excludes wax, wick, jar, labour, packaging, freight and GST. Prices and availability change — the linked product pages are always authoritative.