I Am Making Hanging Car Perfumes With Expensive Fragrance Oils, but Their Throw Is Weak — How Do I Evaluate Diffusion Separately From Fragrance Quality?
Aktie
I Am Making Hanging Car Perfumes With Expensive Fragrance Oils, but Their Throw Is Weak — How Do I Evaluate Diffusion Separately From Fragrance Quality?
Weighing scale ₹354 · the whole diffusion instrumentBottles ₹600 for 12 · ₹50.00 eachStocked in India · ships nationwide
CSI diagnostic guides · car perfume diffusion
"Weak throw" is two complaints wearing one word. One of them a ₹354 scale can settle in thirty days. The other no instrument can settle at all.
★★★★★
"Good quality wood fragrance oil with a warm, long-lasting scent and strong performance in candles."
Payal PatelVerified buyer · Mar 2026
Woody Oud Fragrance Oil
★★★★★
"Good product👍"
Sooraj RVerified buyer · Jul 2026
Reed & Car Diffuser Base
★★★★★
"Thankyou so much candle supplier M happy with the smell of dark vanilla M obsessed with the vanilla fragrance💕 I ll definitely shop again☺️"
SUMAN BISHTJul 2025
Dark Vanilla Oud
★★★★★
"Candles dyes are very good quality and I loved the way they work with all kinds of waxes. Very good"
Antara Srinivas IyerVerified buyer · May 2025
Candle Dye Liquid
★★★★★
"Absolutely loved it!! Smells so perfect!! Go for itt! Packing is done very carefully and happy for that!❤️"
Kavitha murthyOct 2025
Fresh Strawberries Fragrance Oil
★★★★★
"Fragrance is awesome. Hot throw is also very good after burn candle."
AkanshaVerified buyer · Dec 2025
Lavender Fragrance Oil
★★★★★
"Very good"
Reena BagalVerified buyer · Oct 2025
Sandalwood Fragrance Oil
★★★★★
"Quality is good with good fragrance"
Avanthi CheeliVerified buyer · Nov 2025
Citrus Lemon Fragrance Oil
★★★★★
"I like it."
AshwiniSep 2025
British Rose Fragrance Oil
★★★★★
"Will buy more.. wonderful"
Punam SarmaVerified buyer · Dec 2024
Cinnamon Vanilla Fragrance Oil
★★★★★
"Good quality wood fragrance oil with a warm, long-lasting scent and strong performance in candles."
Payal PatelVerified buyer · Mar 2026
Woody Oud Fragrance Oil
★★★★★
"Good product👍"
Sooraj RVerified buyer · Jul 2026
Reed & Car Diffuser Base
★★★★★
"Thankyou so much candle supplier M happy with the smell of dark vanilla M obsessed with the vanilla fragrance💕 I ll definitely shop again☺️"
SUMAN BISHTJul 2025
Dark Vanilla Oud
★★★★★
"Candles dyes are very good quality and I loved the way they work with all kinds of waxes. Very good"
Antara Srinivas IyerVerified buyer · May 2025
Candle Dye Liquid
★★★★★
"Absolutely loved it!! Smells so perfect!! Go for itt! Packing is done very carefully and happy for that!❤️"
Kavitha murthyOct 2025
Fresh Strawberries Fragrance Oil
★★★★★
"Fragrance is awesome. Hot throw is also very good after burn candle."
AkanshaVerified buyer · Dec 2025
Lavender Fragrance Oil
★★★★★
"Very good"
Reena BagalVerified buyer · Oct 2025
Sandalwood Fragrance Oil
★★★★★
"Quality is good with good fragrance"
Avanthi CheeliVerified buyer · Nov 2025
Citrus Lemon Fragrance Oil
★★★★★
"I like it."
AshwiniSep 2025
British Rose Fragrance Oil
★★★★★
"Will buy more.. wonderful"
Punam SarmaVerified buyer · Dec 2024
Cinnamon Vanilla Fragrance Oil
Reviews collected and published through Judge.me on candlemakingsuppliesindia.store. Wording is unedited.
✓ No throw or duration figure quoted for any CSI product✓ Page contradictions named, not resolved✓ Bulk pricing & GST invoicing on WhatsApp
Car Perfume Guides · Diffusion Measurement
I Am Making Hanging Car Perfumes With Expensive Fragrance Oils, but Their Throw Is Weak — How Do I Evaluate Diffusion Separately From Fragrance Quality?
Diffusion is how much material leaves the bottle per day, and it is weighable to the gram. Odour character is whether the material reads as expensive, and no instrument on earth measures it. Mixing the two is why an expensive oil gets blamed for a hardware problem — so run them as two separate studies on the same seven bottles, one on a ₹354 scale and one on a blind panel.
30 weighings
One weighing a day for thirty days, per bottle — the only diffusion data that will ever exist for your product, because no CSI oil, base or bottle page publishes a throw, evaporation or longevity figure · CSI live pricing, 26 October 2026
Quick answers — read this first
How do I measure diffusion? By mass. A hanging bottle loses weight as material leaves it, so grams lost per day is diffusion expressed in the one unit you can actually read off an instrument. Weigh the capped bottle on day 0, then at the same time each day, and log the running total.
How do I measure whether it smells expensive? You cannot measure it. You can only poll it — coded bottles, three or more judges, written scores before anyone speaks. That is a different study with a different output, and it runs alongside the weighing rather than inside it.
Is there a published throw figure I can compare mine against? No. No CSI oil, base, bottle or kit page publishes a throw, evaporation, substantivity or longevity figure. The car freshener kit page does say "long-lasting (weeks, not days)" in its marketing copy; treat that as marketing copy, because nothing in the range supports it with a number.
How long does the log need to run? That depends on your scale, not your patience — and here is the catch: the ₹354 Candle Making Weighing Scale page publishes neither a resolution nor a capacity. So you measure your own noise floor on a sealed control bottle first, then set the window from it. At ±0.2 g of noise and a 0.20 g/day rate, that is a 20-day minimum.
The short answer
Two complaints, two instruments: diffusion is grams per day on a scale; character is a rank from a blind panel. A bottle can be high on one and low on the other, and the fix is different in each case.
Run a sealed control bottle: an identical, capped, un-vented bottle weighed on the same days. Its readings are your instrument noise, and nothing smaller than that noise is a finding.
Change one thing per bottle: seven bottles from a ₹600 pack of twelve gives you a control plus six single-variable tests — fill mass, load, base, cap seating, hang position, oil.
The two tracks share the same seven bottles and the same thirty days, and they answer different questions with different kinds of evidence. Track A produces numbers you can put in a spreadsheet and defend. Track B produces ranks and hit rates. The dashed box is what the whole exercise cannot give you — which is also everything a product page would have to publish for you to skip the exercise, and no page in this range publishes any of it.
My hanging car perfumes use expensive oils but the throw is weak. How do I evaluate diffusion separately from fragrance quality?
Run two studies on the same seven bottles, and never let one conclude for the other. Study A is diffusion and it is a weighing. A hanging bottle loses mass as material leaves it, so grams per day is the honest expression of throw — the only one you can read off an instrument rather than off an opinion. Weigh the capped bottle on day 0, then at the same hour every day for thirty days, log the running total, and divide. Study B is odour character and it is a poll. Code the bottles, use three or more judges, have everyone write before anyone speaks, and run rounds at day 0, 7, 14 and 30. The output is a rank and a hit rate, not a number — and it is the only evidence that bears on "does it read as expensive". Keeping them separate matters because the two faults have opposite fixes. A bottle losing almost nothing per day has a hardware or formulation problem: cap seating, fill level, hang position, base choice, load. A bottle losing plenty per day that still nobody notices has an odour character problem, and buying a dearer oil is a reasonable response to the second and a waste of money against the first. Two live facts shape the method. There is no benchmark to compare against: no CSI oil, base, bottle or kit page publishes a throw, evaporation, substantivity or longevity figure, so the only numbers that will ever exist for your product are the ones you log. And the instrument itself is under-specified: the ₹354 Candle Making Weighing Scale page publishes neither resolution nor capacity, which is why the first thing you weigh is not a test bottle at all but a sealed control, repeatedly, to find out how much your readings wander. That noise floor sets how long the log has to run — at ±0.2 g and a 0.20 g/day rate, twenty days is the minimum window that produces a trustworthy rate. Everything below is that method, costed: the rig comes to ₹1,708.73 including the oil and the base.
One line: weigh the bottle for diffusion, poll the room for character, run a sealed control for noise — and never fix one of those two faults with a purchase aimed at the other.
The whole instrument set: Candle Making Weighing Scale ₹354, wooden-cap car bottles ₹600 for 12 which is seven test bottles and five spares at ₹50.00 each, and a Silicon Measuring Beaker 100 ML ₹300 so every fill is weighed rather than poured by eye. Reused for every batch after this one.
A bottle that releases nothing and a bottle that releases plenty of something forgettable both get described as weak.
The word hides the diagnosis. When a customer says the throw is weak they are reporting one perception, but that perception has two independent causes, and you cannot tell which from the complaint. The first is a delivery problem: not enough material is leaving the bottle into the cabin. That is physical, it is about the cap, the opening, the fill, the carrier and the position, and it shows up on a scale as a small number of grams per day. The second is a perception problem: material is leaving perfectly well, and either nobody registers it or it does not read as the thing you were selling. That one is invisible to a scale and shows up only in a panel.
Separating them is not a methodological nicety, it is where the money is. An expensive oil costs real money per bottle — Khamrah Qahwa at ₹15.06 a gram is ₹45.18 of oil in a 3.0 g charge, three times what a ₹4.94-a-gram oil costs for the same charge — and upgrading the oil is the most common response to a weak-throw complaint. If the real fault is a cap that seats too tightly or a fill level that leaves no headspace, you will have tripled your oil cost and changed nothing, and you will conclude the dearer oil "did not work" when you never tested its delivery at all.
The two axes · instrument, unit, output, limit
What each study settles, and what it has no standing to say
Axis
Instrument and unit
What it settles
What it cannot settle
Diffusion
A scale · grams lost per day, per bottle
Whether material is leaving, how fast, whether the rate is steady or falling, and which single variable changed it
Whether anyone will like the result, or notice it in a moving car
Detectability
A triangle test · hit rate out of five or more rounds
Whether two bottles are distinguishable at all by someone who does not know the codes
Which of the two is better. Discrimination is not preference
Character
A blind forced choice · a rank
Which bottle a panel prefers, and whether the preference holds across rounds
Anything about quantity. A preferred bottle can be the one losing least mass
Throw distance
No instrument you have
Nothing. There is no field method for this in a workshop, and no CSI page publishes a throw figure
Everything. Do not put a distance on a label
Duration
Arithmetic on your own logged rate
Days to empty = fill mass ÷ measured grams per day — your figure, your conditions
A duration for anyone else's conditions, or a claim the range supports. No page publishes one
The mistake: upgrading the oil before weighing the bottle. It is the most expensive wrong move available here, and it is also the most natural one, because the complaint names the fragrance. Spend ₹354 on a scale before you spend ₹1,506 on a 100 g pack of a premium oil. If the log shows your bottles losing very little mass per day, the oil was never the variable — and the same log, run on one bottle with a looser cap and one with a lower fill, will tell you which hardware variable to change. Note also what the hanging bottle is: its own page publishes the mechanism, a wooden lid that absorbs oil and gently releases scent over time, with a braided cord. No reed is supplied and none is needed — the fibre diffuser reeds in the range are for a different product entirely.
The 30-day log, column by column
The log is the whole of Study A, and it is deliberately boring. Ten columns, one row per bottle per day, filled in at the same hour. Everything in it exists because leaving it out has caused somebody to misread their own data.
The log sheet · one row per bottle per day
Every column, and the error it prevents
Column
What goes in it
Why it is there
Date and hour
26-10-2026, 08:30
A reading taken at 08:30 one day and 22:00 the next is a 37.5-hour interval being recorded as one day
Bottle code
A single letter, meaningless
So the sheet can be read without knowing which bottle is the expensive one
Gross mass
The whole capped bottle, in grams
Tare masses drift between bottles; gross mass plus a day-0 reference is self-consistent
Delta since last reading
Yesterday's gross minus today's
The raw daily signal — and the number you must compare against your noise floor, not against zero
Cumulative loss
Day 0 gross minus today's gross
The figure that beats noise. Over 20 days a 0.20 g/day rate is a 4.0 g cumulative loss
The one variable
"cap finger-tight", "fill 10.0 g", "load 15%"
Each bottle differs from the control in exactly one way, and the sheet has to say which
Cap state
Seated / loose / removed and replaced today
Opening a bottle to smell it changes its loss for that day. Record it or lose the day
Position
Hanging in car / hanging indoors / standing on bench
A sealed sunlit cabin is not a room. Two positions are two experiments
Weather note
"hot, parked in sun 6 h" — words, not numbers
Context for the reading. No temperature figure is published for any CSI product, so do not pretend to one
Intensity 0–5, by whom
A number and a name
Keeps the perception data beside the mass data without letting either decide the other
Two rules make the log usable. Weigh capped, every time, with the cap seated exactly as it ships — an uncapped weighing measures a different object. And weigh cold and still: a bottle just brought in from a hot car is warm, the liquid is moving, and a light bottle on a light pan gives a different reading when it is warm from when it is settled. Bring it in, give it an hour, weigh it at the same hour every day.
Depletion arithmetic on a 12.0 g fill · rates you supply, not rates we publish
What each logged rate implies for days to empty, and for a 30-day log
Logged rate
Loss over 30 days
Days to empty a 12.0 g fill
% of fill gone at day 30
What it means for your log
0.05 g/day
1.5 g
240
12.5%
Barely above most noise floors — needs a long window or a finer scale
0.10 g/day
3.0 g
120
25.0%
Readable at ±0.1 g noise over 20 days; marginal at ±0.2 g
0.15 g/day
4.5 g
80
37.5%
Comfortable over a 30-day log at ±0.2 g
0.20 g/day
6.0 g
60
50.0%
Half the fill gone at day 30 — the reference case used throughout this page
0.30 g/day
9.0 g
40
75.0%
Clearly readable; also a bottle your customer will see emptying
0.40 g/day
12.0 g
30
100%
Empty at day 30 — the log and the product end together
Worked out above: days to empty = 12.0 g ÷ rate — 12 ÷ 0.05 = 240, 12 ÷ 0.10 = 120, 12 ÷ 0.20 = 60, 12 ÷ 0.40 = 30. Loss over thirty days is rate × 30, and the percentage is that divided by 12.0 g. Every rate in this table is an illustration of the arithmetic, not a measurement of any CSI product and not a figure any CSI page publishes. The fill is 12.0 g worked by mass at the published 25% working ratio: 3.0 g of oil and 9.0 g of base. The one figure here that will be real is the one your own log produces, and it will be real only for your bottle, your cap, your base, your load and your car.
Your scale's noise floor sets your log length
This is the part almost every maker skips, and it is the part that decides whether the other twenty-nine days mean anything. The Candle Making Weighing Scale page publishes neither a resolution nor a capacity. That is not a complaint about the product — at ₹354 it is the cheapest line in this entire investigation and nothing else here works without it — but it does mean you cannot look up how small a difference it can see. You have to find out.
The way to find out is a sealed control bottle. Fill an identical bottle to the identical mass, cap it, and then seal the cap — tape over the join, or keep the shipping seal on. Weigh it on exactly the same days as the test bottles. In principle it should lose nothing; in practice its readings will wander by some amount, and that wander is the sum of everything your instrument and your procedure contribute. That number is your noise floor, and it is the threshold below which no reading is a finding.
Noise floor to minimum log window · at a 0.20 g/day rate
How long the log must run for the cumulative loss to be ten times the two-reading uncertainty
Noise on your control, ±
Two-reading uncertainty
Minimum days at 0.20 g/day
Minimum days at 0.10 g/day
Verdict
±0.05 g
0.10 g
5
10
A weekly reading is enough; a 30-day log is generous
±0.1 g
0.20 g
10
20
A 30-day log is comfortable at either rate
±0.2 g
0.40 g
20
40
30 days works at 0.20 g/day and is too short at 0.10
±0.5 g
1.00 g
50
100
A 30-day log cannot support a conclusion. Longer log, or a finer scale
±1.0 g
2.00 g
100
200
Unusable for this product. Weigh five bottles together and divide instead
Worked out above: a cumulative loss is the difference between two readings, so its uncertainty is about twice the single-reading noise. Requiring the signal to be ten times that uncertainty gives days ≥ 20 × noise ÷ rate. At ±0.2 g and 0.20 g/day: 20 × 0.2 ÷ 0.2 = 20 days. At ±0.5 g and 0.10 g/day: 20 × 0.5 ÷ 0.1 = 100 days. The factor of ten is this page's choice, stated so you can pick a different one; what is not optional is comparing your daily delta against a measured noise figure rather than against zero. The bottom row's workaround is real and useful: weighing five identical bottles as one group divides the noise across five units, at the cost of losing the per-bottle detail.
Two cheap upgrades to the rig
Keep your sealed control in a 3 ml attar bottle at ₹18.00 rather than a full car bottle if you want a second, smaller control — small sealed glass costs ₹180 for a pack of ten. And weigh every fill in a Silicon Measuring Beaker 100 ML at ₹300 rather than decanting by eye: if your seven bottles do not start at the same mass, the one variable you are testing is no longer the only variable.
What has to be identical across the seven bottles
Seven bottles out of a ₹600 pack of twelve gives you a control, five single-variable tests and one sealed noise bottle — and leaves five spares for the retest you will want. The discipline is the same as any experiment: everything is held constant except the one thing named on the sheet. The list below is what "everything" means for a hanging car bottle, and each row has caught somebody out.
Hold constant, vary one · the seven-bottle layout
What to fix, what each test bottle changes, and what happens if you do not
Bottle
The one change
What it tests
What it exposes if you skip it
A · control
Nothing. Your current formula, current cap, current fill, hanging as sold
The baseline rate every other bottle is read against
Without it, five rates with nothing to compare them to
B · sealed
Cap taped, nothing can leave
Your instrument and procedure noise
Daily deltas judged against zero instead of against noise — the single commonest error
C · cap
Cap seated more loosely, recorded as "finger-tight, quarter turn back"
Whether the cap is the throttle. Its page states the lid itself releases the scent
An oil blamed for a hardware restriction
D · fill
10.0 g of fill instead of 12.0 g, same 25% load
Whether headspace changes the rate
A rate that looks like an oil difference and is a fill difference
E · load
15% oil by mass instead of 25%, inside the published 10–30% window
Whether more oil means more mass leaving, or just more cost
The assumption that load and throw move together, untested
F · base
A different reed and car base, same load, same fill
Whether the carrier is the variable. All five bases publish the same 10–30% window
A base switch made on a product name rather than a measurement
G · position
Same formula as A, hanging indoors instead of in the car
The cabin itself — heat, sun, a sealed volume — as a variable
Thirty days of data that only describes one parked car
Do not open a test bottle to smell it and then say nothing on the sheet. Every uncapping is a loss event that does not belong to the day's diffusion rate, and a log with three unrecorded sniffs in it will show three spikes you will later interpret as weather. If a bottle must be opened for the panel rounds at day 0, 7, 14 and 30, open all of them on those days, for the same length of time, and write it in the cap-state column. Better still, blend a separate small set for the panel and keep the weighed bottles sealed between readings — at ₹50.00 a bottle, duplicating the set costs ₹350 and protects a month of data.
What the rig costs, item by item
Four purchases and one month. The arithmetic is at the end of the cards, and the total is smaller than a single 100 g pack of most premium oils — which is the argument for buying the instrument before buying the upgrade.
1
The instrument, and the one specification it does not publish
Candle Making Weighing Scale
Everything on this page rests on one number you cannot estimate: the mass of a capped bottle today against its mass yesterday. At ₹354 with 98 units live, this is the cheapest line in the investigation and the only one without which none of the rest means anything. Buy it before you buy a dearer oil. One thing to know before you rely on it: its page publishes neither a resolution nor a capacity, so you cannot look up the smallest difference it can read — which is exactly why the method starts with a sealed control bottle weighed repeatedly to establish your own noise floor. Ask for the figures on WhatsApp if you need them before ordering; in the meantime, measure them. A scale whose noise is ±0.2 g still supports a 30-day log at a 0.20 g/day rate, because the cumulative loss by day 20 is 4.0 g.
Seven test bottles, five spares, one published mechanism
Glass Car Hanging Freshener with Wooden Cap (12 ml) Pack of 12
The vessel the whole study is about, at ₹600 per pack of 12 = ₹50.00 each, with 4,960 units live. Its page is unusually informative for this range and worth reading before you blame an oil: it publishes the mechanism — a wooden lid that absorbs oil and gently releases scent over time — plus a braided hanging cord, and it states a minimum order of 12 pieces, which is the only published MOQ anywhere in the range. No reed is supplied and none is wanted; the fibre diffuser reeds CSI sells belong to a different product. One page conflict to name and not resolve: the title says 12 ml while both the handle and the body copy say 20 ml. For a weight-based study it does not matter — you weigh the fill rather than trusting a printed volume — but ask on WhatsApp which the carton is before you print anything.
Size
Price
Per bottle
Live units
Pack of 12
₹600.00
₹50.00
4,960
₹350 of bottles for the seven-bottle layout, from a ₹600 pack of twelveBuy the car bottle pack
3
An expensive oil, costed per bottle
Khamrah Qahwa
If the study is about whether a dear oil earns its place, use one and put a number on it. At ₹15.06 a gram — identical at its 50 g and 100 g rungs — a 3.0 g charge is ₹45.18 of oil per bottle, against ₹14.82 for a ₹4.94-a-gram oil at the same charge. That ₹30.36 difference per bottle is ₹3,036 across a hundred, which is precisely the sum the log exists to justify or refute. 1,199 units live across three rungs; like every brand-referencing title on this shelf it has no rung below 50 g and no 500 g rung, so ₹753 is its cheapest sample, and its kilo at ₹15,056 is ₹4 cheaper than ten 100 g packs. The title is quoted exactly as listed and names an olfactory direction only — no association with or equivalence to any perfume house is stated or implied, and nobody here has compared it with anything.
Size
Price
Per gram
Live units
50g
₹753.00
₹15.06
99
100g
₹1,506.00
₹15.06
1,000
1kg
₹15,056.00
₹15.056
100
₹45.18 of oil in one 12 ml bottle at a 3.0 g chargeBuy Khamrah Qahwa
4
So that seven bottles actually start equal
Silicon Measuring Beaker 100 ML
A seven-bottle study where the bottles do not start at the same mass is a seven-bottle study with two variables in it. At ₹300 with 86 units live, this is how you blend and weigh each fill before it goes into a bottle rather than decanting by eye and hoping. Blend in the beaker, weigh to your target — 3.00 g of oil plus 9.00 g of base for a 12.0 g fill at the published 25% working ratio — stir, then decant, then weigh the capped bottle as your day-0 reading. Note that this listing is titled "Silicon Measuring Beaker 100 ML" and the 250 ML is a separate handle at ₹600; the 100 ml is ample for a 12 g fill and awkward for a 500-bottle batch, so buy the size that matches the job.
Size
Price
Per unit
Live units
Default Title
₹300.00
—
86
₹300 · the difference between seven comparable bottles and seven approximate onesBuy the 100 ML beaker
The rig, totalled: Candle Making Weighing Scale ₹354 + one pack of twelve bottles ₹600 (you cannot buy seven — the page states a 12-piece minimum) + Silicon Measuring Beaker 100 ML ₹300 + oil for seven bottles at 3.0 g each, which is 21 g of Khamrah Qahwa at ₹15.06 = ₹316.26 + base for seven bottles at 9 ml each, which is 63 ml of a ₹1,099 / 500 ml base at ₹2.198 per ml = ₹138.47. Total ₹1,708.73, of which ₹1,254 is instrument and vessel you keep. Against the ₹3,036 that upgrading a hundred bottles from a ₹4.94 oil to a ₹15.06 oil costs, the study is a little over half the price of the decision it informs.
1
Find your noise floor before you fill a single test bottleFill one bottle to 12.0 g, cap it, tape the join, and weigh it five times in a row — lifting it off and replacing it between readings. Write all five numbers down. Then weigh it again at the same hour on each of the next five days. The spread across those readings is your noise floor, and from it the minimum log window follows: days ≥ 20 × noise ÷ expected rate. At ±0.2 g and 0.20 g/day that is twenty days, so a 30-day log clears it.
2
Blend and weigh all seven fills in one sessionSame base pack, same oil pack, same beaker, same afternoon. For the control and five of the test bottles that is 3.00 g of oil and 9.00 g of base to a 12.0 g fill at 25% by mass; bottle D takes 10.0 g of fill and bottle E takes 15% oil. Weigh every fill — do not pour by eye — and record the day-0 gross mass of each capped bottle to the gram. Rest all seven the same measured time inside the published 24–48 hour window before day 1.
3
Code the bottles and hang themLetters that carry no information, written on tape, with the key on a separate sheet. Six go to their positions — five in the car as sold, one indoors — and the sealed control sits with the scale. Note where in the car each one hangs and do not move them: a bottle on a mirror and a bottle in a door pocket are two different experiments, and swapping positions mid-log quietly destroys the data.
4
Weigh daily, at the same hour, capped and coldThirty weighings per bottle. Bring each bottle in, give it an hour to settle if it has been in a hot car, weigh it capped, and fill in all ten columns including the weather note in words. If a bottle was opened for a panel round, say so in the cap-state column — that day's delta is not a diffusion reading and you must be able to see which days to exclude.
5
Read cumulative loss, not daily deltasDaily deltas are mostly noise at this scale; the signal lives in the running total. At day 20, a bottle at 0.20 g/day has lost 4.0 g against a two-reading uncertainty of 0.40 g, which is a ten-to-one margin. Divide cumulative loss by elapsed days for each bottle's rate, subtract whatever the sealed control drifted, and rank the six. Days to empty is then 12.0 g ÷ that rate — 60 days at 0.20 g/day.
6
Write the two conclusions separately, in the same fileOne line for diffusion: "bottle C, cap a quarter-turn loose, 0.26 g/day against the control's 0.14 g/day over 28 days, noise ±0.2 g." One line for character: "panel of three, four rounds, triangle hit rate 4 of 5 between C and A, no consistent preference." Then the decision, and what it cost. Two conclusions in one sentence is how a hardware fix gets spent on an oil.
The character axis, running alongside
Track B is the half of the question a scale has nothing to say about, and it needs its own discipline because it is so easy to contaminate. Four rounds — day 0, day 7, day 14, day 30 — on a duplicate set of bottles blended in the same session as the weighed ones, so the weighed set stays sealed between readings.
Each round answers two questions in order. First, can anyone tell these apart? That is a triangle test: three coded bottles, two identical and one different, and the judge names the odd one out. With three bottles a judge who is guessing is right one time in three, so a single correct answer proves nothing — run at least five rounds with fresh codes and read the hit rate. If your ₹45.18-a-bottle oil cannot be distinguished from your ₹14.82-a-bottle oil in your own hardware, that is the most valuable finding the whole month will produce, and it is worth ₹3,036 per hundred bottles.
Second, and only if the first passes, which do people prefer? Forced choice, judges who are not you, written before anyone speaks, with intensity recorded in a separate column from preference. Keep those two columns apart: a panel will routinely mark the louder bottle as more intense and the quieter one as better, and if you have collapsed the two into one score you will never see it.
Three rules, all of which cost nothing. One bottle at a time with a minute of ordinary air between them, because a strong sample leaves a nose unreliable for the next. Everyone writes before anyone talks, because the first opinion voiced becomes the group's. And the person who blended the bottles pours, times and records but does not score — they are the only one in the room who cannot be blind.
Do not turn a 30-day log into a printed duration or a throw claim. Your log gives you a rate and a days-to-empty figure for your bottle, your cap, your base, your load, your car and that month's weather — which is genuinely useful internally and is not a specification. No CSI oil, base, bottle or accessory page publishes a throw, evaporation, substantivity, longevity or temperature figure, and the one duration-like phrase in the range — "long-lasting (weeks, not days)" on the car freshener kit page — is marketing copy that nothing else supports with a number. Describe your own test conditions if you describe anything, and send labelling and consumer-information questions to a qualified adviser in your market.
The CSI principle
Diffusion is grams per day and you can weigh it. Whether it smells expensive is a vote. Fixing one with a purchase aimed at the other is the whole mistake.
No page in this range publishes a throw figure, which is the honest position: throw is a property of your whole product — your oil, your load, your base, your cap, your fill, your cabin — and no supplier can publish one on your behalf. What anyone can give you is a ₹354 instrument, a method and the arithmetic. The thirty numbers have to be yours.
Running the full layout? ₹1,708.73 all in — scale ₹354, one pack of twelve bottles ₹600, beaker ₹300, 21 g of oil ₹316.26 and 63 ml of base ₹138.47 — against the ₹3,036 that upgrading a hundred bottles from a ₹4.94 oil to a ₹15.06 oil would cost. Bulk pricing and GST invoicing on WhatsApp.
A bottle that loses nothing has nothing to do with the oil in it. Weigh first, then argue about fragrance.
— CandleMakingSuppliesIndia
Why trust this guide
CANDLEMAKINGSUPPLIESINDIA supplies raw materials, not finished car perfumes. That is exactly why there is no throw figure anywhere on this page attached to a CSI product. Diffusion is a property of a complete product — your oil, your load, your base, your cap, your fill mass, your hanging position, your cabin — and no supplier can honestly publish one on your behalf. We would sell more premium oil by telling you which one throws hardest. We are telling you to buy a ₹354 scale and weigh seven bottles for a month, because a maker who measures stops replacing oils by trial and error.
Every price, pack rung and stock figure here was read from live CSI product data on 26 October 2026, variant by variant. Stated specifications are quoted only where a page publishes them: the 10–30% blend window, 25–30% working ratio and 24–48 hour rest that every reed and car diffuser base page carries; the car bottle's published mechanism, braided cord and 12-piece minimum order. Where a page states nothing, we say so — and two silences matter here. The Candle Making Weighing Scale page publishes neither resolution nor capacity, which is why this method begins by measuring its own noise floor. And no CSI oil, base, bottle or accessory page publishes a throw, evaporation, substantivity, longevity, heat-stability or temperature figure; the car freshener kit page's "long-lasting (weeks, not days)" is marketing copy that nothing else in the range supports, and it is not quoted here as a duration. Every rate in the depletion table is an illustration of the arithmetic, not a measurement of any CSI product.
All load arithmetic is by mass on a 12.0 g fill, because no density is published for any CSI oil or base; base costs quoted per millilitre are flagged as such. For bulk pricing, GST invoicing, documentation or the scale's specifications, message us on WhatsApp at +91 7397976926.
Frequently asked questions
How do I measure the throw of a hanging car perfume?
You measure the part of it that is physical: mass lost per day. Weigh the capped bottle on day 0 and at the same hour every day after, log the running total, and divide cumulative loss by elapsed days. A 12.0 g fill losing 0.20 g a day empties in 60 days and will have lost half its mass by day 30. What you cannot measure in a workshop is a throw distance — there is no field method for it and no CSI page publishes a throw figure for any oil, base or bottle — so keep the claim out of your marketing and the rate in your file.
What is the resolution of the CSI weighing scale?
Its page publishes neither a resolution nor a capacity, so the honest answer is that we will not quote one — ask on WhatsApp before ordering if you need the figures confirmed. For this method you can work without them: weigh a sealed, taped control bottle five times in a row and then on five consecutive days, and the spread of those readings is your effective noise floor. The log window follows from it: days ≥ 20 × noise ÷ expected rate, so ±0.2 g of noise at a 0.20 g/day rate needs twenty days, while ±0.5 g needs fifty.
Will an expensive fragrance oil throw better than a cheap one?
Nothing published anywhere in the range supports that, and this page does not claim it. Price per gram runs from about ₹3.74 to ₹24.06 across the fragrance shelf and no page attaches a diffusion figure to any of them. The test is the one above: two bottles identical in base, load, fill mass, cap and position, differing only in oil, weighed on the same days for the same window. Put the money in context first — Khamrah Qahwa at ₹15.06 a gram is ₹45.18 of oil per 3.0 g charge against ₹14.82 for a ₹4.94 oil, which is ₹3,036 across a hundred bottles. That is what the log is deciding.
Should I increase the fragrance load to fix weak throw?
You can test it, inside the published window, and you should test it as one variable rather than assuming it. Every reed and car diffuser base page publishes "blend with 10–30% fragrance oil" with a 25–30% working ratio, so on a 12.0 g fill by mass the range is 1.20 g to 3.60 g of oil. What no page states is that mass loss rises with load — that is a hypothesis, and bottle E in the seven-bottle layout exists to test it at 15% against the control's 25%. Raising the load also raises cost per bottle directly, so a result either way is worth having.
How long will a 12 ml car perfume last?
However long your own logged rate says, and no page in the range will tell you otherwise: no CSI oil, base or bottle page publishes a duration. The car freshener kit page carries "long-lasting (weeks, not days)" and "typically lasts for several weeks" in its copy; treat those as marketing copy rather than specifications, because nothing else in the range supports them with a figure. Your arithmetic is fill mass ÷ grams per day — 12.0 g at 0.20 g/day is 60 days, at 0.40 g/day is 30 — and the honest version of any claim you print names the conditions you measured it under.
Can I compare two bottles that were filled on different days?
Not usefully. Two bottles blended in different sessions differ in base pack, rest time, ambient conditions at filling and your own weighing technique, and the range publishes no batch number or lot code on any oil or base, so you cannot even confirm the two fills came from the same material. Blend the whole set in one session from one pack of each, weigh every fill, record a day-0 gross mass for each capped bottle, and rest them all for the same measured time inside the published 24–48 hour window. If you must add a bottle later, treat it as a separate study with its own control rather than a late entrant to this one.
Weigh it for a month, then decide
₹1,708.73 of rig and thirty weighings — against ₹3,036 of oil upgrade you might not need.
Candle Making Weighing Scale ₹354 (98 units, neither resolution nor capacity published); Glass Car Hanging Freshener with Wooden Cap (12 ml) Pack of 12 ₹600 = ₹50.00 each, 12-piece minimum order, 4,960 units; Silicon Measuring Beaker 100 ML ₹300 with the 250 ML a separate handle at ₹600; Khamrah Qahwa ₹753 / 50 g, ₹1,506 / 100 g (₹15.06 a gram), ₹15,056 / 1 kg; Attar Bottles 3 ml ₹180 for ten = ₹18.00 each for sealed controls; reed and car diffuser bases ₹1,099 / 500 ml = ₹2.198 per ml, Longevity ₹1,296. Message us for bulk pricing, GST invoicing and documentation.
About this guide: Written by the CANDLEMAKINGSUPPLIESINDIA team. Product recommendations reflect the CSI range; the two-track method, the noise-floor rule, the 30-day log sheet and the triangle-then-preference panel apply to any supplier's materials.
Customer reviews: The 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.
Figures verified 26 October 2026, read live per variant: Candle Making Weighing Scale ₹354 (98 units) — its page publishes neither resolution nor capacity. Glass Car Hanging Freshener with Wooden Cap (12 ml) Pack of 12 ₹600 = ₹50.00 each (4,960 units), publishing a wooden lid that absorbs oil and gently releases scent over time, a braided hanging cord and a 12-piece minimum order — the only published MOQ in the range — while its title says 12 ml against a handle and body copy that both say 20 ml. Silicon Measuring Beaker 100 ML ₹300 (86 units), with the 250 ML a separate handle at ₹600. Khamrah Qahwa 1,199 units — 50 g ₹753 (99), 100 g ₹1,506 (1,000), 1 kg ₹15,056 (100), so ₹15.06 / ₹15.06 / ₹15.056 a gram, no rung below 50 g and no 500 g rung, the kilo ₹4 cheaper than ten 100 g packs. Attar Bottles 3 ml ₹180 per pack of 10 = ₹18.00 each (99 packs). Reed and car diffuser bases ₹1,099 / 500 ml = ₹2.198 per ml, Longevity Base ₹1,296 / 500 ml = ₹2.592 per ml. Blue Ocean's ₹4.94 a gram at its 50 g rung is used as the low-cost oil comparison. Arithmetic on this page: a 12.0 g fill at the published 25% working ratio is 3.00 g of oil and 9.00 g of base, worked by mass because no density is published for any CSI oil or base; 3.0 g × ₹15.06 = ₹45.18 and 3.0 g × ₹4.94 = ₹14.82, a ₹30.36 difference per bottle and ₹3,036 across a hundred; days to empty = 12.0 g ÷ rate, giving 240 / 120 / 80 / 60 / 40 / 30 days at 0.05 / 0.10 / 0.15 / 0.20 / 0.30 / 0.40 g per day; minimum log window = 20 × noise ÷ rate, giving 5 / 10 / 20 / 50 / 100 days at ±0.05 / ±0.1 / ±0.2 / ±0.5 / ±1.0 g of noise against a 0.20 g/day rate, where the factor of ten over the two-reading uncertainty is this page's stated choice; the rig totals ₹354 + ₹600 + ₹300 + ₹316.26 of oil (21 g) + ₹138.47 of base (63 ml) = ₹1,708.73. Every depletion rate quoted is an illustration of the arithmetic, not a measurement of any CSI product. Stated specifications as published: every reed and car diffuser base page carries "blend with 10–30% fragrance oil", a 25–30% working ratio and a 24–48 hour rest before bottling. No CSI oil, base, bottle or accessory page publishes a throw, evaporation, substantivity, longevity, heat-stability or temperature figure, no oil page publishes a batch number or lot code, and none of those is stated, estimated or implied here. The car freshener kit page's "long-lasting (weeks, not days)" and "typically lasts for several weeks" are identified as marketing copy and are not quoted as durations. Listed titles referencing perfume names are quoted exactly and name an olfactory direction only; no association with, endorsement by or equivalence to any perfume house is stated or implied. Prices and availability change — the linked product pages are always authoritative.