How Many Fibre Reeds Should I Test in a 100 ml Reed Diffuser Bottle Without Causing Excessively Fast Liquid Consumption?

A reed is a dose, not an accessory

Most people buy reeds the way they buy ribbon. They are the last line on the order, they cost less than the glass, and the number that goes in the bottle is whatever looks right in a photograph. That habit is why reed count is the most common uncontrolled variable in a small diffuser range, and why two bottles of the same blend from the same batch can behave completely differently on two different shelves.

Think about the physical system for a moment. There is liquid in a bottle. There is a neck. There are reeds standing in the liquid with most of their length in the air. Liquid climbs the reed by capillary action and leaves it by evaporating off the exposed length. That exposed length is the entire interface between your product and the room. The bottle does not evaporate. The cap does not evaporate. The liquid surface in the neck contributes very little, because the neck is narrow by design. Essentially all of the fragrance that reaches a nose has gone up a reed and off a reed.

Once that is clear, the arithmetic follows. Add a reed and you have added a fixed increment of evaporating surface. The reeds are the same product — CSI lists one fibre reed in two colours, both at a stated length of 20.3 cm — so eight of them expose a third more length than six of them, and twelve expose twice as much as six. You have not changed the liquid at all. You have changed how fast it is allowed to leave.

Read that table from the right-hand column backwards and the testing order chooses itself. Reed count is the only lever on the list that costs single-digit rupees, acts immediately, and can be undone in ten seconds. Everything else means a new blend, a new rest, a new purchase or new glass. If your bottle is consuming too fast — or not fast enough — this is where to start, not where to finish.

Why your 100 ml is the one fill the band actually covers

There is a quiet gap in this category that is worth knowing about, because it shapes how much weight you should give the published number. The reed pages state their count guidance for "a standard 50-100ml diffuser bottle". The bottle CSI lists alongside those reeds comes in 50 ML and 120 ML. So for the larger listed bottle, the stated reed band does not apply — it stops 20 ml short — and anyone selling the 120 ML SKU is bracketing reed count from scratch with no published starting point at all.

You are in the better position. A 100 ml fill is the upper limit of the range the sentence speaks to, which means the band is guidance for your bottle rather than guidance you are extrapolating into. Use it as what it is: a starting bracket from the supplier, not a specification and not a limit. It tells you where to begin and what the two ends of the first test should be. It does not tell you what your particular blend, in your particular room, will do — nothing published anywhere does that.

One more thing the band does not do: it does not change with the fragrance oil. The sentence is about bottle size, not about what is in the bottle. A 100 ml of a light citrus blend and a 100 ml of a heavy gourmand blend both start at six to eight reeds, and then diverge entirely on the evidence of your own weighed log. If your gourmand SKU is darkening and thickening its reeds over a month, that is a separate question with its own answer — see dark and sticky reeds after one month and a vanilla formula clogging faster than a lemon one.

The millilitres-per-reed ladder

Here is the full ladder for a 100 ml fill, from four reeds to twelve. The second column is the unit this post works in: how many millilitres of your liquid each reed is responsible for moving. The third is the exposed length that count puts into the air, taken from the one reed dimension CSI publishes. The rest is money.

Two readings of this table matter. The first is that the step from six to eight is not large. It is a 33% increase in reed length and a 25% cut in millilitres per reed — meaningful, but not the difference between a quiet bottle and a loud one. If six reeds give you almost nothing in the room, eight will give you almost nothing plus a third, and the problem is elsewhere. The place to look next is the formula and the room, not the reed jar; strong in the bottle and weak in the room works through that differential properly.

The second reading is the one this post is named for. Consumption and throw are the same dial. There is no setting that gives you more fragrance in the room and a longer-lasting bottle at the same time, because the fragrance in the room is the bottle leaving. CSI's own sentence does not pretend otherwise — more reeds for stronger throw and faster diffusion, fewer for a subtler scent and a longer-lasting base. If a 100 ml bottle at eight reeds is emptying faster than your product promise allows, the honest options are fewer reeds, a different promise, or a bigger bottle. They are not "eight reeds that somehow last longer".

What eight reeds cost against what they consume

Now put a rupee figure on both sides of the dose. The reed is a one-off purchase. The liquid it moves is a repeating one — and the whole point of getting the count right is that the count decides how often the repeating purchase happens. Here is the liquid, worked at CSI's published working ratio.

A 100 ml fill at 25% fragrance oil, which is the bottom of the "recommended working ratio is 70-75% base with 25-30% fragrance oil" the base pages publish, is 75 ml of base and 25 g of oil. Keep in mind what the series has had to say all the way through: CSI publishes no density for any base and none for any oil, so the millilitre figure and the gram figure cannot be converted into one another. Weigh the base, then add oil equal to the base weight divided by three, and you have hit 25% by weight without needing a density at all. The millilitre figures below are for costing the base, because that is how the base is sold.

The last column is the number to carry away. At the top of the published band, the reeds are roughly a quarter of what is in the bottle with them — a real line on the bill of materials, not a rounding error, but still the cheapest thing in the system. Below the band they are an eighth. Above it they climb past a third, and at that point you are paying a visible premium in glass fibre for a bottle you will also be refilling more often. That is the double cost of over-dosing reeds, and it is why "just add a few more and see" is not free even though it feels like it.

There is a second cost layer that only shows up when you buy the reeds. The ladder on both reed pages steps exactly once — at the pack of 50 — and is then completely flat to 500. Black is ₹236 for 25 (₹9.44 each), then ₹400 for 50, ₹800 for 100 and ₹4,000 for 500, all at ₹8.00. White runs ₹265, ₹450, ₹900 and ₹4,500, so it is ₹1.16 dearer per reed at the pack of 25 and exactly ₹1.00 dearer at every pack above it. There is no volume reward for going past 50. What there is, is a remainder problem: reeds come in packs, bottles take whole reeds, and the two numbers rarely divide.

Bracketing six against eight: the weighed test

This is the part that costs ₹16.00 and replaces the argument with a number. You are going to fill two identical bottles from one blend, put six reeds in one and eight in the other, and weigh both at fixed checkpoints. What you are measuring is grams lost per day — the only quantity in this whole system that you actually own, because no CSI page publishes an evaporation rate, a throw figure or a life expectancy for any base, reed or blend, and none is implied here.

Thirty days is the right horizon for a first bracket because it matches the only duration claim the reed pages make — "consistent diffusion from day 1 to day 30" — and because it is roughly the service interval a customer will experience. If you want the full protocol, including reed condition and room notes alongside the weight column, the three-days-then-weak post works through an uneven checkpoint schedule designed to catch the front of the curve, which is where most surprises live.

What CSI actually lists, and what a six-bottle trial costs

Four lines go into this test and one of them you may have to source yourself. The cards below are the listed products with every pack rung, so you can see exactly what a six-bottle trial at 100 ml and 25% comes to before you commit.

Frequently asked questions