My Candle Formula Works Perfectly at Small Scale — How Do I Build a Commercial Scale-up Protocol That Controls Raw-Material Lots, Weighing Accuracy, Fragrance Distribution, Mixing, Temperature, Holding Time, Filling, Retained Samples and Batch Release?
Aktie
How is it different from an SOP? The SOP tells the operator how to make the batch. The protocol defines what must be proven — at validation and on every batch — and what happens when a check fails.
Do I need to validate every batch? You validate the process once at production size. After that, every batch is verified: records complete, in-process checks within limits, beginning-middle-end release tests passed.
When must I revalidate? When a material lot, supplier, equipment, batch size, fragrance or key parameter changes, or after an investigation shows the process was not under control.
What a scale-up protocol is for
Your formula works at small scale because, at small scale, the process looks after itself. A kilogram of wax heats and cools quickly, a few stirs distribute the oil, the pour is over in minutes and you are watching every candle. At 25 kg none of that is automatic. The formula has not changed; the conditions around it have.
A scale-up protocol names those conditions and puts each one under control. It has two phases. Validation happens once, when you first make the product at production size: you prove that the process, run strictly as written, makes candles that match the approved reference. Verification happens on every batch afterwards: you check that the batch was made within the validated limits and that its output still matches.
The protocol sits above the SOP. The SOP tells the operator what to do. The protocol says what must be proven, by whom, with what evidence, and what happens when the evidence is missing or wrong.
The nine control points
| Control point | Risk at scale | Control | Evidence | Suggested starting point |
|---|---|---|---|---|
| Raw-material lots | A new pack or bottle behaves differently | Approve each new lot with a 500 g control beside a retained reference | Receiving code, retained sample, control result | One control per new lot before production use |
| Weighing accuracy | Tare errors, wrong percentage basis, drifting scale | Scale checked on a known object; oil weighed in and out; basis written in words | Daily check entry; actual weights on the record | Check before every production day |
| Fragrance distribution | Oil uneven through a large mass | Validated mixing method proven by retains | B-M-E throw ranks at validation | Repeat proof when method or batch size changes |
| Mixing | Effort that suffices at 1 kg falls short at 25 kg | Written tool, pattern and duration | Operator initials; routine followed | Duration set by validation, not copied from 1 kg |
| Temperature | Large mass holds heat; pitcher loads cool | Separate thermometer at melt, oil addition, pour start and end | Temperature log | Readings every few pitcher loads until drift is understood |
| Holding time | Oil held at heat too long before pouring | Maximum time from oil addition to last pour | Times on the record | Set from validation; hold the batch for QC if exceeded |
| Filling | Fill weight drifts as wax thickens | Weigh a candle at set intervals | Fill-weight log | An interval that catches drift within one pitcher load |
| Retained samples | No evidence when questions arise | Retains from beginning, middle and end | Retained-sample register | Three per batch, three per melt for multi-melt runs |
| Batch release | Shipping before proof | Written release rule | Signed release record | No release without all evidence present |
Every figure in the right-hand column is a suggested starting point. Your validation batch replaces each one with a number you have measured.
Two controls makers underestimate
Weighing accuracy
Weighing error works the opposite way to intuition. A 1 g reading error on the 40 g of oil in a 500 g control is 2.5% of the dose; on the 2 kg in a 25 kg batch it is 0.05%. Small controls therefore need the most careful weighing, because they are the reference everything else is judged against. At scale, the danger is not resolution but process: forgetting to tare a bucket, reading the wrong unit, or calculating 8% of the wax instead of 8% of the total.
CSI's scale page does not publish capacity or resolution, so check yours: weigh the same known object several times and note whether the readings agree. If a load is beyond what the scale handles comfortably, weigh in portions and record each portion on the sheet.
Raw-material lots
CSI's product pages do not publish lot numbers or lot-change notices, so lot control is yours. Give every pack and bottle a receiving code on arrival, keep a sealed retained sample, and approve each new one with a 500 g control beside a control made from the retained reference before it enters a production batch.
Worked example: validating 25 kg of Dark Vanilla in CSI 464
The product is a 200 g jar at 8% of total weight. At 25 kg that is 23 kg of CSI 464 and 2 kg of Dark Vanilla, 125 candles before losses. At ₹507.40 per kg of wax and ₹7,646.00 per kg of oil: 23 × ₹507.40 = ₹11,670.20 and 2 × ₹7,646.00 = ₹15,292.00, a total of ₹26,962.20 in wax and oil. Per candle: ₹26,962.20 ÷ 125 = ₹215.70 before wick, jar and labour.
Dark Vanilla's page publishes a flashpoint of approximately 191–193°C, well above melter range, so the addition limit is set by your validated window, not by flashpoint. The page also says the oil contains vanillin and will yellow white or light-coloured wax without affecting performance. The protocol therefore needs a colour criterion as well as a throw criterion: beginning, middle and end candles should match each other and the reference in tone at the same age.
The validation batch is run strictly to the SOP, with temperatures logged every few pitcher loads, fill weighed at intervals, and retains taken at candles 3, 63 and 122. Those retains, cured for at least the page's stated 48–72 hours in soy and compared blind with a fresh 1 kg reference, decide whether the process is validated. If the end candle is lighter, rougher or weaker, the relevant control is tightened and the validation batch is repeated.
Running the protocol on every batch
When a check fails, the protocol already says what happens, so nobody has to decide under pressure. A temperature outside the window means the pour stops until the wax is back in range. A fill-weight drift means the pour technique is corrected and the out-of-tolerance candles are set aside for review. A missing record entry, a release candle that does not match, or a hold time exceeded means the batch is labelled HOLD and investigated, starting with a fresh 1 kg control made from the same materials. The batch is released, reworked or written off only on evidence.
Over months, the verification records become the protocol's own quality check. If one control keeps failing, it is either set too tightly or the process genuinely cannot hold it; either way, the records show where to look.
When to revalidate, and what not to buy yet
Revalidate whenever an input the protocol depends on changes: a new wax, a new fragrance, a new supplier, a different jar, a larger batch, a new melter, a changed mixing method, or an investigation that shows a control was not working. Revalidation is the same exercise as the first time, usually faster because the records show where to look.
Be sceptical of spending as a fix. A mechanical mixer, a larger melter or a filling machine may eventually be right, but only once the protocol's evidence shows that the process itself, properly controlled, is the limit. Until then, a thermometer, a scale, a pitcher, numbered trays and a written release rule are the equipment that matters. When the protocol does point to a material question, bring the evidence to CSI on WhatsApp on +91 7397976926, along with a request for the IFRA and MSDS documents for the oils you use.
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
| Option | Price | Status on the September 2026 pull |
|---|---|---|
| 500 g | ₹260.00 | In stock |
| 1 kg | ₹507.40 | In stock |
| 5 kg | ₹2,537.00 | In stock |
| 10 kg | ₹5,074.00 | In stock |
| Option | Price | Status on the September 2026 pull |
|---|---|---|
| 15 g | ₹130.00 | In stock |
| 50 g | ₹410.00 | In stock |
| 100 g | ₹790.00 | In stock |
| 500 g | ₹3,840.00 | In stock |
| 1 kg | ₹7,646.00 | In stock |
| Option | Price | Status on the September 2026 pull |
|---|---|---|
| Pen thermometer | ₹354.00 | In stock |
CANDLEMAKINGSUPPLIESINDIA supplies raw materials and production equipment, not finished candles. This guide separates general process physics from what CSI's own product pages state, and labels every mixing time, temperature window, hold limit, loss allowance or sample count as a starting point for your own validation.
Prices are live CSI prices from September 2026; product pages are authoritative. Where a spec is not published — such as the scale's capacity or the thermometer's range — this guide says so rather than filling the gap.
For bulk quotes, GST invoicing, IFRA and MSDS documents, message us on WhatsApp at +91 7397976926.
Frequently asked questions
- My 1 Kg Test Formula Is Approved — What Intermediate Batch Size Should I Make Before Jumping Directly to 25 Kg Commercial Production?
- Should I Scale From 1 Kg to 5 Kg to 10 Kg to 25 Kg or Can I Safely Jump Directly to a Much Larger Batch Once Percentages Are Established?
- I Need to Produce 500 Candles From a New Formula — How Many Pilot Candles Should I Make at Commercial Batch Size Before Releasing the Full Run?
- I Have a 1,000-Candle Corporate Order and the Formula Has Only Been Tested at 1 Kg — Should I Validate It at Production Scale Before Buying All Materials?
- My Formula Passed 1 Kg and 5 Kg Tests — What Should I Specifically Evaluate When Moving to a 25 Kg Production Batch?
- I’m Scaling a Bestseller From 100 to 1,000 Candles a Month — Should I Lock a Maximum Batch Size to Preserve Production Consistency?
- Is One 50 Kg Batch More Efficient Than Five 10 Kg Batches If Smaller Batches Give More Consistent Candle Quality?
- How Do I Calculate the Optimal Candle Batch Size by Balancing Labour Efficiency, Equipment Capacity, Mixing Uniformity and Risk of Losing an Entire Batch?
- How to Maintain Fragrance Consistency Across Commercial Candle Batches
- I'm Scaling from 1 Kg Test Batches to 25 Kg Commercial Batches — What Fragrance Formulation Variables Can Change Even When I Scale the Percentages Correctly?
Customer reviews: The reviews at the top are genuine, published Judge.me reviews — Raajull Bhatt (5★, verified); Arminder Kaur (4★, verified); Tina Raha (4★, verified); Lalitha Jagan (5★, not recorded as verified); Judy Roy (5★, verified); Akanksha Sharma (5★, verified). Names, ratings, dates and products as recorded by Judge.me.
Product facts (September 2026, CSI live store): Luxury Soy Wax CSI 464 (flakes): 500 g ₹260.00, 1 kg ₹507.40, 5 kg ₹2,537.00, 10 kg ₹5,074.00 — 100% soy container wax in flakes · single pour · stated fragrance load up to 10% · stated shelf life 60 months (cool, dry storage). Dark Vanilla Fragrance Oil: 15 g ₹130.00, 50 g ₹410.00, 100 g ₹790.00, 500 g ₹3,840.00, 1 kg ₹7,646.00 — top rich vanilla and spice, middle caramel and cream, base dark amber and musk · flashpoint approximately 191–193°C · vanillin PRESENT (percentage not published) — product page says it will yellow white or light-coloured wax and that this does not affect performance · soy 6–10% (cure 48–72 h), paraffin 6–8% (cure 24–48 h). Pen Thermometer: Pen thermometer ₹354.00 — range and accuracy not published.
Assumptions: Fragrance load is a percentage of total candle weight (200 g = 184 g wax + 16 g oil at 8%; a 25 kg batch at 8% = 23 kg wax + 2 kg oil). Melted soy wax is taken as roughly 0.9 g per ml for volume conversions, so the melter's stated 9.5 litres is about 8.5 kg — weigh one real fill. Mixing times, temperature windows, hold limits, loss allowances and sample counts are suggested starting points to validate, not measured CSI data. Example logs and readings are illustrations. Prices and availability change — product pages are authoritative.