I Am Making Rose Reed Diffusers, but the Floral Character Changes After Several Weeks in Summer — Should I Test a Different Rose Accord?
शेयर करना
I Am Making Rose Reed Diffusers, but the Floral Character Changes After Several Weeks in Summer — Should I Test a Different Rose Accord?
Then what do I do? Split your next batch. Bottle the product as usual, and fill one extra bottle from the same mixing vessel, cap it, put it in the coolest dark cupboard you have and do not open it. At fixed weeks, pour a little of each into two identical unmarked cups and have someone else hand them to you in random order. You are not grading the smell. You are answering one question: can you tell which cup is which?
Why not just trust my nose on the working bottle? Because your memory of week 0 is not evidence. Six weeks of living with a scent, a different room temperature, a different time of day and the knowledge that you are checking for a problem all push the answer in the same direction. A same-batch control removes every one of those at once.
What does the change cost to fix if it is real? Nothing to ₹118 a bottle, depending on which diagnosis you land on. Fewer reeds is −₹16.00 of hardware per bottle and reversible the same day. A base change to Depth is ₹0 per bottle, because Depth and Pure Performance are both ₹2.198 a millilitre. Rebuilding the floral around a second oil is where the real money is.
Why the control bottle answers this and a new oil does not
Think about what buying a different rose accord would prove. You would mix it, bottle it, live with it for six weeks and arrive at a second impression — formed in the same room, at the same temperature, by the same nose, with the same six-week gap in it. If that one also seemed to drift you would have learned nothing except that you have two oils and one unreliable measuring device. The measuring device is the problem, and it costs ₹207.80 to replace.
A same-batch control works because it collapses the time axis. The control and the working bottle came out of the same mixing vessel on the same afternoon, which means they started chemically identical — not similar, identical. Whatever differs between them at week six was produced by the only thing that differed: one sat capped in a cool cupboard, one sat open on a shelf in an Indian summer with reeds in it. You are no longer comparing today's smell to a memory. You are comparing two liquids in front of you, at the same moment, at the same temperature, in identical cups. That is the one comparison in this whole series that your nose is actually good at.
It also lets you eliminate the uncomfortable possibility nobody likes to test. Four separate things change across six weeks, only one of them is the formula, and the other three are free to rule out. Work down the table in order of cost to you, and notice that the formula — the expensive suspect — is last.
| What changed | Why it reads as a character change | The test | What it costs you |
|---|---|---|---|
| Your reference point | You are comparing a cup in your hand to a memory six weeks old, formed in a cooler room, and you already suspect a fault | Blind paired judgement against the cool-stored control, somebody else handing over the cups | ₹207.80 once — the control bottle and its liquid |
| The room | Warmer air and moving air both carry more away from an exposed reed surface, so the same bottle presents differently at 2 pm in May than at 9 pm in December | Read a thermometer at every checkpoint and note whether a fan or AC was running; repeat one checkpoint at night with everything off | ₹0 — a note in the log |
| The reeds | Reeds have been wicking for weeks and carry whatever the blend deposits; CSI's fibre reed pages state "consistent diffusion from day 1 to day 30" and say nothing about day 45 | Fit fresh reeds to the working bottle and judge again the next day, same blind pairing | ₹64.00 for eight black reeds at the ₹8.00 pack rate |
| The liquid | The genuine case: the blend in the bottle is no longer the blend you mixed | Only provable once the three rows above are controlled, and only by the panel separating control from working bottle | A reformulation and a full retest — the most expensive answer on this page |
One more reason to run the control before buying anything: it is the only version of this investigation whose result is useful whichever way it falls. If the panel cannot separate the bottles you have saved the cost of a reformulation and learned that your product is steadier than you feared — and you can say so to a customer without overclaiming, because you tested it rather than hoped it. If the panel separates them cleanly, you now have a real finding, a date on which it appeared, and a weight log sitting beside it. That is a brief you can hand to a supplier. "My rose smells funny in summer" is not.
What the rose shelf actually is
This is the part of the question with a hard, checkable answer, and it is worth getting it out of the way before you plan a comparison you cannot buy. The CSI range runs to 405 purchasable product pages. Search them for rose and you get two fragrance oils, a handful of silicone moulds and a clay powder. Of the two oils, one has stock.
| Listing | What it is | Buyable today | What that means for your comparison |
|---|---|---|---|
| British Rose Fragrance Oil | Fragrance oil, five pack rungs from 15 gm to 1 kg | Yes — all five variants available | Your current oil, and the only rose you can run a trial on |
| Smoked Rose Royale | Fragrance oil, three rungs (100 g, 500 g, 1 kg) | No — every variant at zero units | Named here for completeness; it cannot be ordered, so it is not linked and not costed anywhere on this page |
| Rose moulds, rose bud mould, rose cube mould, rose fondant moulds, rose jar mould | Silicone moulds for candle making | Yes | Not fragrance. They appear in a rose search and are irrelevant to a diffuser |
| Rose clay powder | A soap and cosmetic ingredient | Yes | Not fragrance either |
| Nearest listed florals | Gardenia, Ylang Ylang Flower, Jasmine | Yes | A different floral direction, not a different rose. Treat any of them as a new product with its own trial, not a substitution |
So the shopping half of your question has a one-line answer: there is nothing to buy. That is not a sales problem, it is a relief — it means the money you were about to spend on a second accord is available for the thing that will actually settle the matter. And the British Rose ladder is unusually kind if you do end up committing: the per-gram price falls at every single rung, which is less common across this catalogue than makers expect.
Running the blind same-or-different judgement
The method below is deliberately narrow. It does not grade intensity, it does not score adjectives, and it never asks anyone to say whether a smell is nicer. All of those are opinions that drift with mood, room and time of day. It asks one question with a right answer — which cup is the cool one — and because there is a right answer, the result can be counted. If you are also screening several candidate oils at once, use the full rig in the screening guide and treat the steps here as the judgement that sits inside it.
Reading the four outcomes
A score on its own is half the answer. Cross it with the weight log and the diagnosis sharpens considerably, because the two readings fail in different ways. The panel tells you whether the liquid is separable by nose. The scale tells you how much of it has left the bottle. Read together, they distinguish a formula that has genuinely moved from a bottle that is simply three-quarters gone and presenting what remains.
| Panel result | Weight log | Most likely reading | Next action, cheapest first |
|---|---|---|---|
| Near 6 of 12 — not separable | Consumption steady across checkpoints | Nothing has changed. The impression was yours, not the bottle's | Stop. Write the result down, publish nothing new, and keep the control for the next batch comparison. ₹0 |
| Near 6 of 12 — not separable | Consumption rising, bottle well down | The liquid is the same; there is simply much less of it, with loaded reeds and a lower fill | Fresh reeds and a placement change first — ₹64.00 of reeds, or −₹16.00 a bottle if the answer is two reeds fewer. Formula untouched |
| 11 or 12 of 12 — clearly separable | Consumption steady | Something in the warm, open bottle has changed that is not explained by mass loss | Repeat at the next checkpoint to confirm, then vary one thing: reeds, cap discipline, placement. Only then the formula |
| 11 or 12 of 12 — clearly separable | Consumption high and rising | Heavy loss plus a separable difference: the strongest case for a formula or base decision | Rule 1 applies — judge any base swap on both consumption and throw. Depth is ₹0 a bottle against Pure Performance; Longevity is ₹35.46 more at 25% |
| 7 to 10 of 12 | Either | Inconclusive, and the most common honest result | Add a checkpoint and a judge rather than a purchase. Record it as inconclusive in your own log — do not round it up into a finding |
The bottom row is the one makers skip, and it is the one that keeps a product honest. An inconclusive result is a result. Writing "9 of 12 at week six, inconclusive, re-tested at week eight" into a batch file costs nothing and gives you a record you can stand behind when a customer or a retailer asks what testing you did. Rounding it up to "we found the fragrance drifts in heat" is how a business ends up reformulating a product that was fine and advertising a change nobody can detect.
If the change is real: what the floral rebuild costs
Suppose the panel lands on twelve of twelve twice in a row and the reeds and the placement have already been ruled out. Now the purchase question is live — and it is not "which rose instead", because there is no other rose in stock. It is "what do I build around the rose I have". The cards below are the listed florals that can carry part of a rose blend, priced at every rung with the per-bottle arithmetic worked out at the published 25% ratio, where a 120 ml bottle is 90 ml of base and 30 g of oil.
| Pack | Price | Per gram | Oil cost per 120 ml bottle at 25% |
|---|---|---|---|
| 15 gm | ₹94.40 | ₹6.29333 | ₹188.80 (half a bottle per pack) |
| 50 gm | ₹212.40 | ₹4.248 | ₹127.44 |
| 100 gm | ₹342.20 | ₹3.422 | ₹102.66 |
| 500 gm | ₹1,593.00 | ₹3.186 | ₹95.58 |
| 1 kg | ₹3,068.00 | ₹3.068 | ₹92.04 |
| Pack | Price | Per gram | Cost of 3 g in a 90:10 build |
|---|---|---|---|
| 15 gm | ₹106.20 | ₹7.08 | ₹21.24 |
| 50 gm | ₹330.40 | ₹6.608 | ₹19.82 |
| 100 gm | ₹590.00 | ₹5.90 | ₹17.70 |
| 500 gm | ₹2,714.00 | ₹5.428 | ₹16.28 |
| 1 kg | ₹5,310.00 | ₹5.31 | ₹15.93 |
| Pack | Price | Per gram | Oil cost per 120 ml bottle at 25% |
|---|---|---|---|
| 15 gm | ₹105.02 | ₹7.00133 | ₹210.04 |
| 50 gm | ₹223.02 | ₹4.4604 | ₹133.81 |
| 100 gm | ₹433.00 | ₹4.33 | ₹129.90 |
| 500 gm | ₹2,166.00 | ₹4.332 | ₹129.96 |
| 1,000 gm | ₹4,332.00 | ₹4.332 | ₹129.96 |
| Variant | Price | Per bottle | Packs in stock |
|---|---|---|---|
| 50 ML Pack of 10 | ₹826.00 | ₹82.60 | 16 (160 bottles) |
| 120 ML Pack of 10 | ₹1,062.00 | ₹106.20 | 24 (240 bottles) |
Buying for the test, not for the fix
The cheapest version of this investigation buys one pack of oil, one pack of bottles and one scale, and most makers already own the scale. The table below is the shopping list at three sizes of ambition, from a single control alongside your current production run to a four-arm floral comparison. Note that the first row changes nothing about your product: you are buying one bottle's worth of extra liquid out of a batch you were mixing anyway.
| Version | What you buy | Cost | What it answers |
|---|---|---|---|
| One control, this batch | British Rose 50 gm ₹212.40 + 50 ML Pack of 10 ₹826 | ₹1,038.40, of which ₹207.80 is the control itself and nine spare bottles remain | Whether your rose is separable from its own cool control. Nothing about any other oil |
| Control plus reed and placement arms | Add black fibre reeds Pack of 50 ₹400 | ₹1,438.40 | Adds the two free-to-test suspects: loaded reeds and bottle position, each as its own one-variable run |
| Three-floral coded comparison | Add Ylang Ylang 15 gm ₹106.20 + Gardenia 15 gm ₹105.02 + 120 ML Pack of 10 ₹1,062 | ₹2,711.62 | Whether a rebuilt floral is separable from rose after the same exposure. Still says nothing about product life |
| Scale, if you do not own one | Candle Making Weighing Scale ₹354 | +₹354 once | Turns every checkpoint into grams per day instead of an impression |
| Thermometer and a notebook | Whatever you already have | ₹0 | The room temperature and the fan state, which is half the explanation in an Indian summer |
On the base question: if you do reach for a different base, the rule is the same one that governs this whole series. A slower base is not automatically a better base. If it restricts wicking or fragrance release, your diffuser lasts longer and smells of less, and you have traded one complaint for another. So read every base swap on two axes at once — grams lost per day and what the panel says about throw at the same checkpoint. The money involved is smaller than makers assume: Depth, the one base whose page claims to slow evaporation, costs exactly the same ₹2.198 a millilitre as Pure Performance, so the swap is ₹0 a bottle. Only Longevity carries a premium at ₹2.592 a millilitre — ₹35.46 more on a 90 ml fill. One honest complication worth knowing: Depth's own page lists light florals under what to avoid and routes them to Lift, so for a floral the slow-evaporation base is the one CSI's copy steers away from. And Softness, at the same ₹1,099, is the page written for gentle and comforting scents. Those are page positions, not measurements, and they cannot rank two bases numerically.
| Base | ₹ per ml | 90 ml of base | + 30 g British Rose at ₹3.422 | + hardware ₹170.20 |
|---|---|---|---|---|
| Pure Performance 500 ML | ₹2.198 | ₹197.82 | ₹300.48 | ₹470.68 |
| Depth 500 ml | ₹2.198 | ₹197.82 | ₹300.48 | ₹470.68 |
| Lift 500 ml | ₹2.198 | ₹197.82 | ₹300.48 | ₹470.68 |
| Softness 500 ml | ₹2.198 | ₹197.82 | ₹300.48 | ₹470.68 |
| Longevity 500 ml | ₹2.592 | ₹233.28 | ₹335.94 | ₹506.14 |
Four of the five bases are the same price, which means the base decision in this range is almost never a cost decision — it is a performance decision you have to test, on both axes, with the control bottle you have now built. Hardware is the ₹170.20 floor: a 120 ml bottle at ₹106.20 plus eight black reeds at ₹8.00 each, before a drop of liquid. And the published load window is where the real money sits: at 20% the oil in a bottle drops to 24 g and ₹82.13 at the 100 gm rung, at 30% it rises to 36 g and ₹123.19. Both are inside what the base pages publish; neither is a claim about how the product will smell in September.
CANDLEMAKINGSUPPLIESINDIA supplies raw materials, not finished diffusers. That is why this page contains no statement about what British Rose, Ylang Ylang or Gardenia does in a warm room over six weeks. Nobody here has smelled your batch, no CSI product page publishes a temperature, a flash point used as a working temperature, a viscosity, an evaporation rate, a shelf life or an odour-stability figure, and a page that invented one would be selling you a number instead of a method. The honest commercial position is the awkward one: the oil you might have wanted to buy — a second rose — is out of stock, and the thing that will actually answer your question costs ₹207.80.
Every price, pack size and stock figure here comes from live CSI data in October 2026 and links to the product page, where current pricing always takes precedence. Per-variant availability was checked before each link, and the one rose listing that is live but at zero units across all three variants is named and deliberately not linked. Stated specifications are quoted only where a page publishes them: the 10–30% blend window, the 25–30% working ratio and the 24–48 hour rest on the base pages; 20.3 cm reeds, 6–8 reeds for a 50–100 ml bottle, "uniform pore structure", "consistent diffusion from day 1 to day 30" and the instruction to avoid direct sunlight and heat sources on the fibre reed pages; "slows evaporation" and the light-floral avoid list on Depth; "extended cold diffusion" on Longevity. Those are CSI's words about its own products, not measurements, and they cannot rank two products numerically. Blend percentages are formulation windows, not safety limits — nothing here is certified or approved by IFRA, and category limits, labelling and compliance are matters for your supplier's documentation and a qualified adviser.
For help designing a control trial for your own batch, for bulk pricing, GST invoicing, MSDS or IFRA documentation, message us on WhatsApp at +91 7397976926.
Frequently asked questions
- The full sample-screening rig: coded bottles, two controls, four checkpoints
- Lavender turning sharp in summer — one arm unbuyable, the other a build
- A jasmine diffuser that becomes overpowering — reed count before formula
- Fifty days in AC, twenty in a hot flat — the fixed-checkpoint fix
- Customers report shorter life in summer — complaint to single-variable test
- Reeds dark and sticky after a month — what residue does to the comparison
- The 30-day weighed test: weight loss, intensity, reed condition, temperature
Customer reviews: The ten 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; wording is unedited. They are candle and hardware observations and none of them concerns odour stability in a warm room.
Figures verified October 2026. Control bottle ₹207.80 = 50 ML bottle ₹82.60 (₹826 ÷ 10) + 37.5 ml Pure Performance at ₹2.198/ml (₹1,099 ÷ 500) = ₹82.425 + 12.5 g British Rose at ₹3.422/g = ₹42.775. British Rose live: 15 gm ₹94.40, 50 gm ₹212.40, 100 gm ₹342.20, 500 gm ₹1,593, 1 kg ₹3,068 — per gram ₹6.29333, ₹4.248, ₹3.422, ₹3.186, ₹3.068, the steps between rungs being ₹2.045, ₹0.826, ₹0.236 and ₹0.118, and the kilo saving ₹118 against two 500 gm packs (₹3,186). Its 500 gm stock figure reads as a placeholder and no depth claim is built on it. Smoked Rose Royale is ACTIVE with 100 g, 500 g and 1 kg all at zero units and availableForSale false — named, never linked. Ylang Ylang Flower ₹106.20 / ₹330.40 / ₹590 / ₹2,714 / ₹5,310, per gram ₹7.08, ₹6.608, ₹5.90, ₹5.428, ₹5.31, kilo saving ₹118. Gardenia ₹105.02 / ₹223.02 / ₹433 / ₹2,166 / ₹4,332, per gram ₹7.00133, ₹4.4604, ₹4.33, ₹4.332, ₹4.332, kilo saving ₹0, both top rungs at three packs. Jasmine's cheapest rung is its 500 gm at ₹3.922/g. Bottles: 50 ML Pack of 10 ₹826 (₹82.60 each, 16 packs = 160 bottles), 120 ML Pack of 10 ₹1,062 (₹106.20 each, 24 packs = 240 bottles); CSI lists no 100 ml and no 200 ml reed bottle. Black fibre reeds ₹236 / 25 (₹9.44 each), ₹400 / 50, ₹800 / 100, ₹4,000 / 500 (₹8.00 each from the 50-pack up); white ₹265 / ₹450 / ₹900 / ₹4,500. Hardware floor ₹170.20 = ₹106.20 bottle + 8 × ₹8.00. Bases: Pure Performance 100 g ₹325 and 500 ML ₹1,099, Depth ₹1,099, Lift ₹1,099, Softness ₹1,099, Longevity ₹1,296 — ₹2.198/ml for the first four and ₹2.592/ml for Longevity, so 90 ml is ₹197.82 or ₹233.28 and the Longevity premium on a 120 ml bottle at 25% is ₹35.46. Oil per bottle = 30 g × per-gram price; at 20% it is 24 g and at the 30% published maximum 36 g. One kilogram of oil fills 1,000 ÷ 30 = 33.3 bottles at 25%. Weighing scale ₹354. Published specifications quoted as published: "Blend with 10-30% fragrance oil", a 25–30% working ratio and a 24–48 hour rest before bottling on the base pages; 20.3 cm length, "uniform pore structure — no swelling, no clogging, no splitting", "consistent diffusion from day 1 to day 30", 6–8 reeds for a 50–100 ml bottle, and "avoid placing in direct sunlight or near heat sources which accelerate evaporation" on the fibre reed pages; "slows evaporation" and a light-floral avoid list on Depth; "extended cold diffusion" on Longevity; "performs well in Indian climates by balancing scent throw with longevity" on Pure Performance. CSI publishes no temperature, flash-point working range, viscosity, volatility, evaporation rate, humidity effect, shelf life or odour-stability figure for any oil, base, reed or bottle in this range, and none is stated, estimated or implied here. Every week count, temperature and panel score on this page is the reader's own observation or an illustration of the arithmetic; the twelve-judgement guessing average of six is the arithmetic of a two-way forced choice, not a statistical threshold, and the 11-of-12 and 12-of-12 cut-offs are this page's recommendation for your own log rather than a published standard. Prices and availability change — the linked product pages are always authoritative.