How Do I Scale a Tested Reed Diffuser Formula from a 100 g Laboratory Trial to 1 kg, 5 kg, 10 kg and 25 kg Commercial Batches While Maintaining Accuracy and Quality?
शेयर करना
Do I need 1, 5 and 10 kg steps? Only where the vessel, mixer, scale or lots change. Often a 1 kg pilot with production lots is enough.
What keeps quality? A gate at every step: clarity repeated on each batch, wicking and throw confirmed against a remade control, uniformity and fill checks at scale.
What does it cost? A 1 kg pilot is ₹4,740.00; the 25 kg run ₹1,11,912.00 in CSI materials. See the cost ladder.
One formula at five sizes, solved to two decimals
Fragrance load in this series is the share of the total finished blend by weight. A tested 100 g trial at 20% — 20 g of oil and 80 g of base — is not a recipe of 20 g and 80 g; it is the ratio 20:80, and every larger batch is that ratio applied to a new total. Formula 1 (forward) at each size gives the table below. Formula 4 (actual load) gives exactly 20.00% for every row. If your load is not 20%, replace 0.20; if your trial was not 100 g, it does not matter — the ratio is what scales.
This page is the capstone of the section on scaling, so it does three things the individual posts do not: it carries one formula through every size, it shows how each common error grows with the batch, and it sets out the gate that should stand between each step and the next. Each step links the post that owns its detail. The pillar for the load itself is how much fragrance oil to add; the business pillar is starting a profitable reed diffuser business; the general method from the earlier series is converting a 100 g trial into a 10 kg batch and developing a trial and scaling it to 25 kg.
| Size | Oil | Base | Scale step → load band | Volume | 500 ml base at 0.95 / 1.00 / 1.05 | Oil packs |
|---|---|---|---|---|---|---|
| 100 g trial | 20.00 g | 80.00 g | 1 g → 19.500%–20.500% | about 0.10 L | 1 / 1 / 1 | 2 × 15 g |
| 1 kg pilot | 200.00 g | 800.00 g | 0.1 g → 19.995%–20.005% | about 1.01 L | 2 / 2 / 2 | 2 × 100 g |
| 5 kg | 1,000.00 g | 4,000.00 g | 1 g → 19.990%–20.010% | about 5.05 L | 9 / 8 / 8 | 1 × 1 kg |
| 10 kg | 2,000.00 g | 8,000.00 g | 1 g → 19.995%–20.005% | about 10.10 L | 17 / 16 / 16 | 2 × 1 kg |
| 25 kg | 5,000.00 g | 20,000.00 g | 5 g → 19.990%–20.010% | about 25.25 L | 43 / 40 / 39 | 5 × 1 kg |
Two columns deserve a second look. The load band shrinks as the batch grows: even a 5 g platform scale holds 25 kg within about ±0.010 of a point, while the original 100 g trial on a 1 g kitchen scale could have been anywhere from 19.50% to 20.50%. Precision at scale is easy; the trial is the least precise batch you will ever make. And the base column is in bottles: CSI lists its bases in 500 ml and publishes no density, so 20,000 g of base is 40 bottles at an example 1.00 g/ml and 43 if your base is 5% lighter. The section hub starts the ladder at 1 kg; a 250 g formula at 10 kg starts it from a different trial size.
How the classic errors grow with the batch
Three errors cause most scale-up failures, and each one gets more expensive as the batch grows. The table puts them in grams of oil at every size, using the same formula, so you can see what the error would cost you at the size you are about to make.
| Size | Base taken as total: load | Base taken as total: oil missing | Mixed by volume: oil missing | × (1 + loss) instead of ÷: bottles short |
|---|---|---|---|---|
| 100 g trial | 16.67% | 5.00 g | 1.58 g | — |
| 1 kg pilot | 16.67% | 50.00 g | 15.84 g | 0.00 |
| 5 kg | 16.67% | 250.00 g | 79.21 g | 0.02 |
| 10 kg | 16.67% | 500.00 g | 158.42 g | 0.04 |
| 25 kg | 16.67% | 1,250.00 g | 396.04 g | 0.10 |
Base taken as total — weighing the batch size as base and adding 20% of it as oil — gives 16.67% at every size, and the oil missing grows from 5.00 g at 100 g to 1,250 g at 25 kg; the 15 kg weighing sheet is built to stop it and 8 kg of base at 20% shows the fixed-base formula. Mixing by volume — 80 ml + 20 ml — is 20% by weight only if the base and oil weigh the same per ml; with the example densities it is 18.42%, and at 25 kg the difference is 396 g of oil. Bottle volume into batch weight with measured density covers the conversion. Multiplying by (1 + loss) instead of dividing by (1 − loss) leaves you a fraction of a bottle short at 2%, more at higher allowances; the 3% filling-loss post shows why.
What actually changes at each step
Each step up brings a new vessel, a different way of stirring, usually a different scale, often new lots of base and oil, and longer gaps between weighing, resting and filling. Those are the things a gate has to test. The table uses labelled example equipment; yours may differ, and the rule is to add a step to the ladder whenever one of these changes — not because the next round number is bigger.
| Size | Scale step | Vessel (example) | Stirring (example) | Base and oil lots | Retained sample |
|---|---|---|---|---|---|
| 100 g trial | 1 g | 250 ml beaker | glass rod | same as trial | — |
| 1 kg pilot | 0.1 g | 2-litre jug | handheld mixer or rod | same as trial | 100 g per batch |
| 5 kg | 1 g | 10-litre bucket | long paddle | new lots likely | 100 g per batch |
| 10 kg | 1 g | 15-litre drum | paddle, timed | new lots likely | 100 g per batch |
| 25 kg | 5 g | 40-litre drum | paddle or drum mixer, timed | new lots likely | 100 g per batch |
So you may not need all five sizes. If your 1 kg pilot uses the production lots and a vessel and paddle like the 25 kg ones, the pilot may be the only step between trial and production. If the 25 kg drum is new, a 5 kg batch in it is cheaper than discovering at 25 kg that the paddle does not reach the bottom. Mixing time and equipment from 500 g to 20 kg shows how to prove a stirring time; 1 kg good, 10 kg different shows how to tell a vessel problem from a lot problem; a clear 100 g trial turning cloudy at 5 kg is the page for when clarity fails on the way up.
The gate between each step
A gate is the test a batch must pass before the next, larger one is weighed or bought for. It uses CSI's change-control vocabulary: KEEP a result that cannot change, CONFIRM with a short side-by-side against a control, REPEAT the full test, or add a NEW check that only exists at scale. What to retest after the pilot explains each status in detail; the table below is the whole ladder at a glance.
| Batch | Status | Test | What passing proves |
|---|---|---|---|
| 100 g trial | APPROVE | 7-day clarity hold + cool night; blind side-by-side against your earlier best; IFRA 10A limit read from the certificate | the formula exists |
| 1 kg pilot | REPEAT clarity · CONFIRM wicking, throw · NEW uniformity | Production lots; remade 100 g control on a 0.1 g scale; P1–P3 from first, middle, last fill; days 0, 1, 3, 7 | the formula scales |
| 5 kg | REPEAT clarity, uniformity · CONFIRM short | Level samples T/M/B before final stir; two bottles vs control for 7 days | the vessel and stirring work |
| 10 kg | REPEAT clarity, uniformity · CONFIRM short · record lots | As 5 kg, plus base and oil order dates on the sheet | the batch is repeatable |
| 25 kg | REPEAT clarity, uniformity, filling · NEW retained sample | T/M/B, F1/F5/F9 fill weights, sealed retained sample at 7 days | the batch can ship |
One weighing sheet for every size
The same six lines work from 1 kg to 25 kg; only the numbers change. Print it once, fill in T (the total) and L (the load), and compute line 1 at a desk before the oil is opened. The "Scale should read" column is the cumulative figure on the display with the vessel tared, which lets anyone walking past check it. The page cannot run a calculator, so line 2 is a blank for a phone calculator, with the formula printed beside it.
| Line | Ingredient | Formula | Target g | Scale should read | Actual g | Tare / check | Initials |
|---|---|---|---|---|---|---|---|
| 0 | Header | product · load (exact) · total · base lot · oil lot · vessel · scale step | — | — | ______ | Batch code | ____ |
| 1 | Base | Formula 1 (forward): total × (1 − L) | T × 0.80 | T × 0.80 | ______ | Write actual before line 2 | ____ |
| 2 | Recompute oil | Formula 2 (fixed base): actual base × L ÷ (1 − L) | = ______ g | — | ______ | Phone calculator, full precision | ____ |
| 3 | Oil | line 2, rounded to the scale step | line 2 | line 1 actual + line 3 | ______ | Pour straight into the vessel | ____ |
| 4 | Check | Formula 4 (actual load): line 3 ÷ (line 1 + line 3) | L | — | ______ | Second person | ____ |
| 5 | Samples | T, M, B before final stir; retained sample after | — | — | ______ | 0 h · 48 h · 7 days | ____ |
If a batch is split across vessels, each vessel gets its own sheet and its own total — 50 kg across five fragrances shows how to divide a large target into vessel-sized batches. If the formula uses two or three oils, the total load applies to the combined oils and each oil's grams are total oil × its share (Formula 8 (blends of oils)); scaling a two-oil trial to 5 kg and 10 kg with three oils at 20% cover blends. If oil or base is the limiting stock, 3 kg of oil at 25% and 5 kg of jasmine with only 750 g of oil show what the stock allows.
Losses and retained samples at each size
Every batch above the trial loses some liquid to the vessel, the funnel and the filling, and every batch should keep a retained sample. Formula 10 (losses) applies them once, in order: retained sample first, then the loss allowance, then round up to real packs. This page uses an example 2% transfer and filling loss and one 100 g retained sample per batch, with 120 ml Gold bottles filled to an example 100 ml at the measured 0.99 g/ml — 99.00 g per bottle.
| Batch | Made | Retained sample | Loss | Into bottles | Bottles |
|---|---|---|---|---|---|
| 1 kg pilot | 1,000 g | 100 g | 20.00 g | 880.00 g | 8 |
| 5 kg | 5,000 g | 100 g | 100.00 g | 4,800.00 g | 48 |
| 10 kg | 10,000 g | 100 g | 200.00 g | 9,700.00 g | 97 |
| 25 kg | 25,000 g | 100 g | 500.00 g | 24,400.00 g | 246 |
If you start from a bottle count instead of a batch size, run the chain the other way: bottles × fill weight = net; (net + retained samples) ÷ (1 − loss) = gross; gross × load = oil. The bottle-count section owns that direction — the purchasing hub starts there. Measure your real loss on the first production batch (made minus bottled, divided by made) and use the measured figure next time; there is no standard one.
| If a bottle loses | Days to empty | Weeks |
|---|---|---|
| 1.5 g per day | 66.0 | 9.4 |
| 2.0 g per day | 49.5 | 7.1 |
| 2.5 g per day | 39.6 | 5.7 |
| 3.0 g per day | 33.0 | 4.7 |
The cost of each step, and the priced 25 kg run
Priced from listed, in-stock CSI sizes: Lavender's packs rounded to the fewest that cover each need, and Softness Base sized for a base 5% lighter than an example 1.00 g/ml so a weighing never runs dry. The last column is the liquid's value at the 1 kg oil rate and the example density — what a failed batch at that size would throw away.
| Size | Oil | Oil packs | Oil cost | Base | Liquid to buy | Liquid value at risk |
|---|---|---|---|---|---|---|
| 100 g trial | 20.00 g | 2 × 15 g | ₹210.04 | 1 × 500 ml = ₹1,099.00 | ₹1,309.04 | ₹280.26 |
| 1 kg pilot | 200.00 g | 2 × 100 g | ₹1,080.00 | 2 × 500 ml = ₹2,198.00 | ₹3,278.00 | ₹2,802.60 |
| 5 kg | 1,000.00 g | 1 × 1 kg | ₹5,221.00 | 9 × 500 ml = ₹9,891.00 | ₹15,112.00 | ₹14,013.00 |
| 10 kg | 2,000.00 g | 2 × 1 kg | ₹10,442.00 | 17 × 500 ml = ₹18,683.00 | ₹29,125.00 | ₹28,026.00 |
| 25 kg | 5,000.00 g | 5 × 1 kg | ₹26,105.00 | 43 × 500 ml = ₹47,257.00 | ₹73,362.00 | ₹70,065.00 |
| Item | Need | Buy | Price |
|---|---|---|---|
| Lavender | 5,000.00 g | 5 × 1 kg | ₹26,105.00 |
| Softness Base | 20,000.00 g | 43 × 500 ml | ₹47,257.00 |
| Gold Reed Diffuser Bottle 120 ml | 246 bottles | 25 × pack of 10 | ₹26,550.00 |
| Fiber Diffuser Reeds (Black) | 1,476 reeds (6 per bottle) | 3 × pack of 500 | ₹12,000.00 |
| CSI materials | ₹1,11,912.00 |
| Line | Quantity | Cost |
|---|---|---|
| Fragrance oil | 19.80 g | ₹103.38 |
| Base | 79.20 g | ₹175.84 |
| Gold bottle 120 ml | 1 | ₹106.20 |
| Black fibre reeds | 6 | ₹48.00 |
| CSI materials per diffuser | ₹433.42 |
Lavender is ₹5.40 per gram at 100 g and ₹5.22 at 1 kg — a 3.3% step, typical of most CSI oils, which are priced almost linearly above 100 g. Buying bigger mostly buys fewer packs to open and stock security, not a lower rate; a few oils have real breaks and a few are cheaper at 500 g than at 1 kg, so compute yours. CSI lists no 5 kg or 25 kg oil or base pack: quantities above the listed packs are quoted on WhatsApp, and no volume discount is published. The complete purchase list for 1,000 diffusers and ordering wholesale without overbuying take the order further.
The decision rule
| If this is true | It means | Do this |
|---|---|---|
| The 100 g has not passed a 7-day clarity hold and its IFRA 10A check | There is no formula yet | Buy nothing for scale. 8724 · 8729 |
| The formula is written as grams, not a ratio | Chain-scaling risk | Rewrite it as L by weight with lots and density |
| No pilot with production lots and equipment | Scale-up untested | Buy the 1 kg pilot only (₹4,740.00) |
| Pilot fails clarity or differs from the remade control | Gate failed | Fix at 1 kg — 8748 · 8755 |
| The production vessel, mixer or scale is new | One more step needed | Make one batch at the next size in the new equipment |
| Every gate passed | Ready for production | Order 5 × 1 kg oil, 43 × 500 ml base, bottles and reeds; weigh on the sheet |
The rule never adds a step for its own sake and never skips the clarity repeat. If a batch goes wrong on the way up, the correction maths is in its own section: the correcting-mistakes hub starts with an over-oiled 100 g, and reconstructing a mis-weighed 10 kg batch works from the records. For choosing the load before you scale, the fragrance-load hub; for the single-bottle arithmetic, the exact-calculations hub.
The CSI products in this ladder
| Size | Price | Per unit | Stock (15 Oct 2026) |
|---|---|---|---|
| 500 ml | ₹1,099.00 | ₹2.20 per ml · ₹2.22 per g at an example 0.99 g/ml | In stock |
| CSI HyperDiffusion Base™ (different product) | ₹1,499.00 | — | Out of stock — not offered |
| Pack | Price | Per gram | Bottles at 19.8 g oil |
|---|---|---|---|
| 15 g | ₹105.02 | ₹7.00 | 0.8 |
| 50 g | ₹270.00 | ₹5.40 | 2.5 |
| 100 g | ₹540.00 | ₹5.40 | 5.1 |
| 500 g | ₹2,620.00 | ₹5.24 | 25.3 |
| 1 kg | ₹5,221.00 | ₹5.22 | 50.5 |
| Variant | Price | Per bottle | Stock (15 Oct 2026) |
|---|---|---|---|
| 50 ml · pack of 10 | ₹826.00 | ₹82.60 | In stock |
| 120 ml · pack of 10 | ₹1,062.00 | ₹106.20 | In stock |
| Pack | Price | Per reed | Bottles covered |
|---|---|---|---|
| Pack of 25 | ₹236.00 | ₹9.44 | 4 bottles at 6 reeds |
| Pack of 50 | ₹400.00 | ₹8.00 | 8 bottles at 6 reeds |
| Pack of 100 | ₹800.00 | ₹8.00 | 16 bottles at 6 reeds |
| Pack of 500 | ₹4,000.00 | ₹8.00 | 83 bottles at 6 reeds |
| Item | Price | Note |
|---|---|---|
| Weighing Scale | ₹354.00 | Resolution not published — check the display steps |
| Silicon Measuring Beaker 250 ml | ₹600.00 | For batches above about 100 g |
| CSI Pro Mixer (cordless handheld) | ₹349.00 | Optional; a clean glass rod or spoon also works |
Every post in this section, and the other hubs
This capstone carries one formula through every size. Each post below owns one question from the ladder, starting from a reader's own numbers.
| Your numbers | What the post settles | Post |
|---|---|---|
| 100 g at 20 g + 80 g → 1 kg | The ×10 card | 8741 |
| 5 kg at 22% | Oil and base for one target weight | 8742 |
| 250 g formula → 10 kg | Keeping strength at ×40 | 8743 |
| 25 kg of jasmine at 20% with filling losses | Quantities plus loss | 8744 |
| Two-oil lavender trial → 5 kg | Scaling each component | 8745 |
| 1 kg at 18% — allowance before or after? | Order of operations | 8746 |
| 500 g → 20 kg: mixing time and equipment | Uniformity checks | 8747 |
| 100 g clear, 5 kg cloudy | What to investigate before 25 kg | 8748 |
| 10 kg with three oils at 20% | Each ingredient of a three-oil blend | 8749 |
| 3 kg of oil at 25% | Maximum finished liquid | 8750 |
| 15 kg with a ready-made base | A weighing sheet that blocks % errors | 8751 |
| 120 ml formula → 1,000 bottles | Bottle volume to batch weight by measured density | 8752 |
| 8 kg of base in stock | Oil to order for 20% | 8753 |
| 50 kg across five fragrances | Dividing into production batches | 8754 |
| 1 kg good, 10 kg different | Mixing uniformity and lot consistency | 8755 |
| 5 kg of jasmine with only 750 g oil | The maximum load you can reach | 8756 |
| 100 g → 25 kg | What to retest after the pilot | 8757 |
| 20 kg at 21.5% | Precise weights without rounding errors | 8758 |
| 500 diffusers, 3% filling loss | Extra liquid without changing the formula | 8759 |
| Hub | Owns | Post |
|---|---|---|
| Exact batch calculations (hub) | Single-bottle arithmetic by weight and volume | 8701 |
| Fragrance load and performance (hub) | Whether 15%, 20% or 25% is right | 8721 |
| Scaling 100 g to 25 kg (hub) | This section's starting point | 8741 |
| Bottle-count purchasing (hub) | Turning bottle orders into kilograms | 8761 |
| Correcting formula mistakes (hub) | Rescue maths and costing | 8781 |
CANDLEMAKINGSUPPLIESINDIA supplies raw materials, not finished diffusers. The easy answer is "multiply by ten, then five, then two and a half". This page shows how that compounds, recommends making only the steps your equipment needs instead of selling you about ₹1,20,877.00 of ladder, and says plainly that we publish no density, clarity, throw, evaporation or lot data for any base or oil.
Diffuser bases, fibre reeds, Gold bottles, beakers, scale and mixer are CSI live pricing, re-checked 15 October 2026. Fragrance oils are CSI live pricing, pulled 24 September 2026, stock re-checked 14 October 2026. Every gram split, ml-to-g conversion, per-bottle cost and kit total on this page is computed and shown to two decimals so you can check it; densities are labelled examples, never CSI specifications.
Bottle counts per arm, clarity checkpoints, weighing schedules, room-scoring scales and loss allowances are labelled working practices, not industry standards — none exists for reed diffuser performance. The scale steps, vessels and stirrers, the 0.99 g/ml blend density, the 1.00 ± 5% g/ml base density, the 0.93/1.03 g/ml volume-error densities, the 0.4% pour overshoot, the 100 ml fill, the 2% loss, the 100 g retained samples and the 6-reed count are labelled examples you replace with your own. For help building a test kit, bulk quotes, GST invoicing, MSDS or IFRA documents, message us on WhatsApp at +91 7397976926.
Frequently asked questions
- I Successfully Made a 100 g Reed Diffuser Formula with 20 g Fragrance Oil and 80 g Base — How Do I Scale It to 1 kg Without Changing the Ratio?
- I Want to Make 5 kg of Reed Diffuser Liquid at 22% Fragrance Load — How Much Fragrance Oil and Base Should I Weigh?
- I Want to Make 25 kg of Reed Diffuser Liquid with 20% Jasmine Fragrance Oil — How Do I Calculate the Raw-Material Quantities and Allow for Filling Losses?
- I Am Scaling a 500 g Reed Diffuser Formula to 20 kg — How Do I Check Whether Mixing Time and Equipment Affect Uniformity?
- I Am Making a 15 kg Reed Diffuser Batch Using a Ready-Made Base — How Do I Create a Weighing Sheet That Prevents Fragrance Percentage Errors?
- I Want to Scale My 120 ml Reed Diffuser Formula to 1,000 Bottles — How Do I Convert Bottle Volume into Total Batch Weight Using Measured Density?
- My 1 kg Reed Diffuser Batch Has Good Throw, but the 10 kg Version Performs Differently — How Do I Check Mixing Uniformity and Raw-Material Consistency?
- I Am Scaling a Reed Diffuser Formula from 100 g to 25 kg — Should I Retest Stability, Wicking and Fragrance Throw After the Pilot Batch?
- I Want to Make 20 kg of Reed Diffuser Liquid with 21.5% Fragrance Oil — How Do I Calculate Precise Ingredient Weights Without Rounding Errors?
- I Am Making 500 Reed Diffusers and Expect 3% Filling Loss — How Do I Calculate the Extra Liquid to Manufacture Without Changing the Formula?
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 re-checked 15 October 2026 and pulled 24 September 2026: Diffuser bases, reeds, bottles and tools — CSI live pricing, re-checked 15 October 2026: Pure Performance Base ₹325.00 / 100 g, ₹1,099.00 / 500 ml; Longevity Base ₹1,296.00 / 500 ml; Lift Base ₹1,099.00 / 500 ml; Depth Base ₹1,099.00 / 500 ml; Softness Base ₹1,099.00 / 500 ml; CSI HyperDiffusion Base™ ₹1,499.00 (out of stock); Fiber Diffuser Reeds (Black) ₹236.00 / 25, ₹400.00 / 50, ₹800.00 / 100, ₹4,000.00 / 500; Fiber Diffuser Reeds (White) ₹265.00 / 25, ₹450.00 / 50, ₹900.00 / 100, ₹4,500.00 / 500 (both 20.3 cm long); Gold Reed Diffuser Bottle 50 ml ₹826.00 / pack of 10, 120 ml ₹1,062.00 / pack of 10; Weighing Scale ₹354.00; Silicon Measuring Beaker 100 ml ₹300.00, 250 ml ₹600.00; CSI Pro Mixer ₹349.00. Fragrance oils — CSI live pricing, pulled 24 September 2026, stock re-checked 14 October 2026: Lavender Fragrance Oil ₹105.02 / 15 g, ₹270.00 / 50 g, ₹540.00 / 100 g, ₹2,620.00 / 500 g, ₹5,221.00 / 1 kg.
Non-price figures and assumptions: Fragrance load is a share of the total blend by weight (oil g = blend g × L; base g = blend g × (1 − L)). No density is published for any CSI base or oil: every ml-to-g conversion uses a labelled example density and shows ±5%; weigh 100 ml of your own blend to get yours. Batch sizes 100 g, 1 kg, 5 kg, 10 kg and 25 kg are the reader's; 20% load is a labelled example. Example scale steps 1 g (original trial), 0.1 g (remade control and pilot), 1 g (5 and 10 kg), 5 g (25 kg). Example measured blend density 0.99 g/ml; base density 1.00 g/ml ± 5% for bottle counts; example densities 0.93 (oil) and 1.03 g/ml (base) for the volume-mixing error; example 0.4% pour overshoot for chain scaling; example vessels and stirrers. Labelled working practices: 100 ml fill per 120 ml bottle, 2% transfer and filling loss, 100 g retained sample per batch, 6 black reeds per bottle, clarity checks at 0 h, 48 h and 7 days plus one cool night, weighing on days 0, 1, 3 and 7. KEEP / CONFIRM / REPEAT / NEW is CSI's change-control vocabulary. Per-bottle cost uses the 1 kg oil rate and Softness Base at the example measured density. What CSI's base pages state (quoted as claims to test, not CSI test results): every base is a "Professional … Reed & Car Diffuser Base", crystal clear per its page, used at 10–30% fragrance of the total blend with a 24–48 hour rest before bottling. Pure Performance — neutral, "no added fixatives, no complex enhancements", for all families, "start at 10% … up to a maximum of 30%", "20% recommended", a "typical working blend 75% base + 25% fragrance", the example "for a 100 ml finished diffuser at 25%: 75 ml base + 25 ml fragrance oil", and "flip reeds after the first hour". Longevity — oriental/amber, woody, resinous/balsamic, spicy and gourmand; "avoid light florals, fresh citrus, aquatics — use Lift Base"; slower "cold diffusion"; "if clouding occurs, reduce fragrance percentage and retest"; "allow 2–3 hours before the first flip". Lift — floral, citrus, green and fruity. Depth — chypre, leather, smoky and rich gourmand; "stabilises tricky sweet and spicy fragrance oils". Softness — soft florals, powdery/musky, light gourmand. Reed page — 20.3 cm long; "for a standard 50–100 ml diffuser bottle, 6–8 reeds gives a good balance". Bottle page — variants 50 ml and 120 ml (its text says "50 and 100ml"); neck diameter, fill line and any reducer are not stated. CSI publishes no composition, density, viscosity, evaporation rate, flash point or shelf life for any diffuser base; no throw, evaporation or longevity result for any base, oil or reed; no reed diameter; no bottle neck size or fill line; no IFRA percentage for any oil; and no bulk base price. The five bases are told apart only by what their pages say they are for — never by a measured result. IFRA: reed diffusers are listed under Category 10A in current IFRA amendments as commonly summarised in suppliers' guides; read the Category 10A limit on the certificate for your oil, which CSI provides on WhatsApp request. Material costs exclude labour, packaging, freight and GST. Diwali 2026 falls on Sunday 8 November, 24 days after this page was written on 15 October 2026.