My reed diffuser is clear in a glass beaker but looks cloudy after filling the bottle — could container residue or lighting be misleading me?
Share
My reed diffuser is clear in a glass beaker but looks cloudy after filling the bottle — could container residue or lighting be misleading me?
- The free test that decides whether you have a formula problem at all
- Nine things changed between the vessel and the bottle, and the liquid is one of them
- The bottle as first suspect: residue, rinse water and what “dry” means
- The rinse-and-dry protocol, built on what CSI actually sells
- Why the vessel itself changes what you see
- The reference-vessel rule: read clarity in one vessel, every time
- If the pour-back says the blend really did change
- What the investigation costs and what it protects
- Buying for the audit, then for the batch
- Frequently asked questions
The free test that decides whether you have a formula problem at all
Before you buy a different base, a different oil or a hygrometer, spend ten minutes moving liquid backwards.
You have made two observations and you are treating them as one. Observation one: the blend in the measuring vessel was clear. Observation two: the blend in the bottle is not. Between them, at least nine things changed, and the composition of the liquid is only one of them. The pour-back test holds eight of those nine still and asks the only question worth asking first — is it the liquid, or is it everything else?
Do it like this. Wash and dry the measuring vessel you originally used, or take a clean one. Pour the bottle's contents straight back into it. Stand it on the same surface, in front of the same background, under the same light, within a few minutes of the bottle reading. Look at it the same way you looked the first time — through the side, not down into it. If the liquid reads clear again, you have your answer, and it is the cheap one: nothing in your blend changed, and what you are fighting is the container or the way you are viewing it.
Two warnings about the test, because it is easy to run badly. First, do it within minutes rather than the next morning: a blend that genuinely changes with temperature or time will change again while you are not looking, and then the pour-back tells you nothing. Second, do not strain, filter, warm or stir it on the way back. The point is to move the same liquid into a different vessel and change nothing else. If you filter it and it comes back clear, you have proved that filtering works and learnt nothing about the cause.
If the pour-back reads clear, keep going down the vessel route in this page's order: the bottle's cleanliness first, because it is the cheapest to fix and the most common; then the bottle as an optical object, because that one is permanent and worth knowing about; and then the viewing conditions, which are where a sibling page in this series takes over — the scenario where a formula looks clear under warm light and cloudy in daylight is a question about standardising how you look, and it has its own page. This one stays on the vessel.
If the pour-back is still not clear, the useful consequence is that you have ruled out the entire vessel branch in one step, and you can stop buying glassware. You are now in the published part of the process. Every CSI base page carries the same instruction when a blend does not hold its clarity: reduce the fragrance percentage and retest. Lift publishes it as “If clouding occurs, the fragrance oil may not be fully compatible at that load - reduce fragrance percentage and retest.”, and Softness almost word for word: “If clouding occurs, the fragrance oil may not be fully compatible at that load - reduce the fragrance percentage and retest.” Read on to the last two sections of this page for what that means in weights and rupees, and read CSI's live page on how to determine whether a cloudy formula is physically compatible for the full diagnosis.
Nine things changed between the vessel and the bottle, and the liquid is one of them
Write the list out once and the diagnosis stops feeling mysterious. Most of it has nothing to do with your formula.
A measuring vessel and a sales bottle are two different optical instruments, and you have swapped one for the other mid-observation. The depth of liquid you are looking through changed. The curvature of the wall changed. The thickness of the wall changed. The colour of the glass may have changed. What is behind the liquid changed, because a beaker on a bench has a room behind it and a bottle has its own far wall. The surface you are standing it on changed. The light reaching it changed, because you moved. The temperature of the glass changed, if the bottle came out of a carton in a warm room. And, yes, the liquid may have changed, if time or temperature did something in the interval.
Rank them by cost to fix rather than by likelihood, because that is how you should spend your afternoon. Residue and rinse water are free to fix and the most common single cause in a small production run, so they go first. Viewing conditions are free to fix and need a rule rather than a purchase. The bottle's own optics cannot be fixed at all — a thick gold-finished bottle shows you less than a measuring beaker does — so they need to be understood rather than solved. The blend itself is the expensive one, and it is the only branch where remaking fifty bottles at ₹244.28 of raw material each becomes the question.
It is worth being specific about depth, because it is the factor makers underestimate most. Reading a 20 mm layer in a wide measuring vessel and reading a 100 ml column in a narrow bottle are not the same measurement at all: a very slight loss of clarity that is invisible across 20 mm can be plainly visible along a path five times longer. Nothing about the liquid needs to have changed for that to happen. This is also why the pour-back is so informative — it puts the path length back to where your clear reading came from.
The published process does not tell you which vessel to read clarity in, and that is not an omission so much as the reason this page exists: the published step is “check for clarity — the blend should remain crystal clear”, and the vessel is left to you. Longevity's page states the outcome you are aiming at plainly — “No clouding, no separation when used correctly.” — and its clarity step gives the action if you do not get it. So the vessel is a decision you have to make and then hold, which is exactly what the reference-vessel rule further down this page does.
| What changed | How you rule it out | Cost | If this is the cause |
|---|---|---|---|
| Residue inside the bottle | Fill a second bottle, straight from the carton, with the same liquid | Free | Rinse-and-dry protocol. ₹1.04 of spray and 20 g of base per bottle |
| Rinse water that did not dry | Dry a washed bottle fully, then refill from the same liquid | Free, plus patience | A longer drying step, and a sacrificial base rinse before filling |
| Path length — 100 ml deep against 20 mm | The pour-back test | Free | Nothing is wrong. Fix the reading, not the blend |
| Wall thickness and curvature | Pour some into a plain thin-walled vessel | Free | Read clarity in the reference vessel and judge the bottle on appearance alone |
| Tinted, frosted or gold-finished glass | Compare against a plain clear vessel | Free | A design decision, not a fault. Gold bottles are ₹106.20 each in a pack of ten |
| Background behind the liquid | White card behind, then black card behind | Free | A fixed background in your clarity check, written into the sheet |
| The light you looked under | Look under both your lights, one minute apart | Free | Standardise the light — a sibling page in this series owns that question in full |
| Glass temperature | Let the bottle sit an hour at room temperature and look again | Free | A rest before reading, and a logged room temperature |
| The blend itself | The pour-back, then a fresh arm at a lower load | ₹325.00 for a trial pack of base | The published step: reduce the fragrance percentage and retest |
The bottle as first suspect: residue, rinse water and what “dry” means
A bottle that looks clean and a bottle that is dry are two different states, and only one of them is safe to fill.
Three kinds of thing can be waiting inside a bottle you are about to fill. There is manufacturing and packing residue — dust, glass fines, release agents, the smear from a gloved hand, whatever travelled with it. There is anything you introduced yourself, which in a small workshop is usually water from a rinse, and sometimes washing-up liquid, which is designed to hold oil and water together and is the single worst thing to leave in a diffuser bottle. And there is residue from a previous fill, if this is a returned, reused or sample bottle.
Water is worth its own paragraph because it is the one most makers get caught by. A reed diffuser blend is an oil-based liquid. Water left in the bottle is a second liquid that does not belong in the formula, and a few millilitres in the bottom of a 120 ml bottle is enough to be visible once the bottle is full. CSI publishes no water-tolerance figure, no maximum moisture content and no drying instruction for any base, so there is no number for you to look up here: the honest version of this advice is a protocol. Dry the bottle until there is nothing to see and nothing to smell, and if you cannot be sure, use a bottle you have not washed.
The straight-from-the-carton control is the experiment that settles it, and it is free. Take one bottle you have washed and dried, and one bottle out of the carton untouched. Fill both from the same vessel of the same liquid, at the same minute. Stand them side by side in the same light. If only the washed bottle is cloudy, your washing process is the cause and you have found it in twenty minutes. If both are cloudy, the bottle is not the cause — which is itself worth ₹325.00 of base you now do not need to spend.
What CSI does publish for this job is the Isopropyl Alcohol Spray. Its page states the purpose plainly: it is for cleaning candle jars, and it names what it does — removing residue and lingering scents from jars. That is a published claim about residue in a glass vessel, which is exactly the problem you have, but note what the page does not say: it does not give a concentration, a method, a contact time, a drying time, or any instruction for a diffuser bottle in particular. So the use below is a protocol built on a published purpose rather than a published method, and it is written to be conservative accordingly.
The pack arithmetic is the one place the choice is clear-cut. The 150 ml pack is ₹383.00, which is ₹2.55 per millilitre. The 1 litre pack is ₹520.00, which is ₹0.52 per millilitre — ₹2.03 per millilitre less, 79.63% lower, for ₹137.00 more on the shelf. At roughly 2 ml a bottle, the 150 ml pack covers about 75 bottles at ₹5.11 each and the litre covers about 500 at ₹1.04. Buy the 150 ml pack to test the protocol; buy the litre the moment the protocol is part of your production routine.
The rinse-and-dry protocol, built on what CSI actually sells
Six steps, two products, and one trick that costs five grams of your own base and removes the last doubt.
The protocol below is assembled from two CSI products and one thing you already own. The Isopropyl Alcohol Spray at ₹383.00 for 150 ml is published for removing residue and lingering scents from jars. A Silicon Measuring Beaker at ₹300.00 for 100 ml or ₹600.00 for 250 ml is published as a measuring and mixing vessel, with a transparent design and clear scales so you can see the liquid inside, and the page names reed diffuser making among its uses. The third component is your own base, used as a final rinse and then discarded — which is the step that removes the last doubt, because it introduces nothing that was not already going into the bottle.
That sacrificial rinse is the part worth paying attention to. After an alcohol wipe and a full dry, pour 20 g of plain base into the bottle, swirl it to wet the whole inside wall, and pour it away. Anything the alcohol loosened and left behind leaves with it, and what remains on the glass is base — the same liquid that is about to fill the bottle. At ₹2.20 a gram from the 500 ml pack that is ₹10.99 a bottle, ₹109.90 across ten bottles and ₹549.50 across fifty. For a product you are selling on appearance, that is a rounding error against ₹395.48 a finished unit.
Drying is the step people rush, so give it a rule you can keep. Invert the bottle on a clean rack, in moving air, out of direct sun, and leave it longer than you think necessary. Then check it two ways: look through the neck at a light for any film or droplet, and smell it. Alcohol has a smell; if you can smell it, it has not finished evaporating. CSI publishes no drying time for the spray, so the test is the sense check rather than a clock, and the sacrificial base rinse is your insurance against getting it wrong.
Do all of this on a batch basis rather than bottle by bottle. Clean and dry ten bottles together, rinse all ten sacrificially, fill all ten from one vessel of one blend, and read all ten side by side against a white card. A row of ten identical bottles is a far better instrument than one bottle: if nine are clear and one is not, you have a bottle problem and you can see it in a second. If all ten are equally not clear, the pour-back has already told you to look at the blend.
| Step | Quantity | Per bottle | Per ten bottles | Per fifty bottles |
|---|---|---|---|---|
| Alcohol spray, 150 ml pack | about 2 ml | ₹5.11 | ₹51.07 | ₹255.33 |
| Alcohol spray, 1 litre pack | about 2 ml | ₹1.04 | ₹10.40 | ₹52.00 |
| Sacrificial base rinse | 20 g | ₹10.99 | ₹109.90 | ₹549.50 |
| Both, at the litre rate | 2 ml plus 20 g | ₹12.03 | ₹120.30 | ₹601.50 |
Why the vessel itself changes what you see
Some of what you are looking at is permanent, and the right response is to stop reading clarity in that vessel.
A measuring vessel is built to be looked into. CSI's Silicon Measuring Beaker page states it directly: with clear scales and a transparent design, you can see the liquid inside, so you can measure the volume accurately. That is a vessel designed around reading a level and a liquid. A finished diffuser bottle is designed around a shelf: a heavier wall, a shaped shoulder, a narrow neck, often a gold or tinted finish, a label across a third of it. Those are the right decisions for a product and the wrong ones for an instrument.
So three of the nine differences are not problems to be solved. Wall thickness scatters and dims. Curvature bends what is behind the liquid into it, which is why a bottle can look softer or milkier than the same liquid on a bench even when nothing is wrong. A gold or tinted glass removes some of the light before it reaches your eye and changes the colour of the rest. And the depth of liquid, as the previous section said, multiplies any very slight loss of transparency along a much longer path.
That has one clean consequence for how you work. Judge clarity in the reference vessel, every time, as a formulation decision. Judge appearance in the production bottle, as a product decision. They are two different judgements with two different standards, and conflating them is what put you in this position. A blend that is crystal clear in the beaker and slightly less brilliant through 4 mm of gold-finished glass has passed the formulation gate and may still fail your own appearance standard — in which case the honest fix is the bottle, the label or the fill level, not the formula.
If appearance in the bottle is the real problem, CSI's base pages give you the published lever. Pure Performance is published as crystal clear, with the note that its clear appearance works with any label design and shows off coloured fragrance blends in glass bottles. Lift's page says the same of coloured fragrance oils. So the published route to a bottle that looks right is a clear base plus whatever colour your fragrance oil brings, inside a bottle you have chosen deliberately. CSI's Gold Reed Diffuser Bottles come in two sizes — ₹826.00 for ten at 50 ml, which is ₹82.60 each, and ₹1,062.00 for ten at 120 ml, which is ₹106.20 each.
| Vessel | Price | What the page publishes | Judge this in it |
|---|---|---|---|
| Silicon Measuring Beaker 100 ml | ₹300.00 | A transparent design with clear scales so you can see the liquid inside; named for candle making and reed diffuser making; non-stick, easy to wash and dry | Clarity, during formulation. This is the reference vessel |
| Silicon Measuring Beaker 250 ml | ₹600.00 | The same published text at a larger capacity — ₹240.00 per 100 ml of capacity against ₹300.00 | Clarity on a bigger trial blend, and mixing a 100 g to 200 g batch |
| Gold Reed Diffuser Bottle, 50 ml | ₹826.00 for ten, ₹82.60 each | A finished reed diffuser bottle in a pack of ten | Appearance of the product you are selling, not clarity of the formula |
| Gold Reed Diffuser Bottle, 120 ml | ₹1,062.00 for ten, ₹106.20 each | A finished reed diffuser bottle in a pack of ten | Appearance, fill level and how the label sits |
| A plain thin-walled glass you already own | Nothing | Nothing published — it is not a CSI product | A free second opinion. If the liquid is clear here too, the bottle is doing the work |
The reference-vessel rule: read clarity in one vessel, every time
The published step says check for clarity. It does not say what to check it in. That part is yours to fix, once.
Every CSI base page runs the same sequence: measure the base, add the fragrance oil, stir until fully combined, check for clarity, rest 24 to 48 hours, bottle. The clarity check sits before the rest and before bottling, which tells you what it is for — a gate on the blend, taken at the bench. It does not specify a vessel, a light, a background or a distance, because those are workshop decisions. Make them once, write them down, and your clarity check becomes a measurement instead of an impression.
Here is a rule you can adopt today, and the cost is one beaker. One vessel, named in your batch sheet, used for every clarity reading in your business. One background, a white card, with a black card as the second look — some losses of clarity show against white and some against black, and two cards cost nothing. One fill depth, so the path length is the same every time: mark a line on the vessel and fill to it. One eye level, through the side. One distance. And the light logged at every reading, which is where the sibling page on warm light against daylight takes the argument much further than this page needs to.
Then the production bottle gets its own, separate standard, and it is a commercial one rather than a formulation one. Fill one bottle from the batch, cap it, stand it where a customer would see it, and decide whether it looks like the product you are selling. That is a judgement about your brand, and it is allowed to be stricter than the formulation gate. What it must not do is overwrite the formulation gate: a blend that is clear in the reference vessel and slightly dulled through gold glass has not failed the published check, and remaking it will not change the glass.
Two readings, two records, one sheet. Date, batch number, base and pack, oil and pack and lot, load as a percentage of the total finished blend, both weights, mixing method, rest hours, and then the two readings: clarity in the reference vessel, and appearance in the bottle. Add the room temperature and the light. When the same question comes back in June with a different oil lot, that sheet is the difference between a diagnosis and a guess — and a single sheet from October is worth more than any number this page could invent.
If the pour-back says the blend really did change
Then the vessel branch is closed, you have spent nothing on it, and the published route is short.
A blend that is clear in the vessel, not clear in the bottle, and not clear again when you pour it back has changed in the interval. That is a formula result and it moves you onto published ground. The clarity step on every CSI base page says the same thing in nearly the same words. Lift: “If clouding occurs, the fragrance oil may not be fully compatible at that load - reduce fragrance percentage and retest.” Softness: “If clouding occurs, the fragrance oil may not be fully compatible at that load - reduce the fragrance percentage and retest.” Depth: “If any clouding occurs, reduce fragrance percentage and retest before proceeding to batch production.” Longevity: “The blend should remain crystal clear. If clouding occurs, reduce fragrance percentage and retest. Heavy oils like oud can cloud at maximum load in some bases.” The instruction is to reduce the fragrance percentage and retest, and the reason the Lift and Softness wording is the most useful is the phrase at that load — compatibility is published as a property of an oil and a base together at a percentage.
Before you change anything, check the load you are actually at, because the convention catches people. Load is published as a percentage of the total finished blend. Twenty grams of fragrance oil with 80 g of base is a 100 g blend at 20% — the published recommended working load. Against the base alone the same mix reads 25%, which is the candle convention and the number a spreadsheet built for wax will give you. Both are arithmetically true; only the first matches the pages. Write both on the sheet so the next batch is unambiguous.
Then run the step-down as arms rather than as a rescue. Mix three fresh 20 g arms from the base you already own at 20%, 15% and 10%, read each in the reference vessel at mixing and again after the published 24 to 48 hour rest, and log the room temperature at every reading. A 100 g trial pack at ₹325.00 covers that, with oil from a 50 g pack. If the lower arms hold and the higher one does not, you have found a working load and a published reason for it. If every arm behaves the same way, the load is not your variable and the pair is — which is where the base question starts.
On the base question, CSI publishes a family map, and it is the only published guidance anywhere in the range about which base suits which oil. The pages name families, and placing your own oil in a family is CSI's editorial judgement rather than something a base page prints. British Rose, for instance, is a floral under that editorial classification, and the Lift page names floral as a family it is built for. The step from “my oil is a rose” to “roses are florals” is the step you make; the step from “floral” to “Lift” is the one the page publishes.
| Family, as the pages name it | Base published for it | 500 ml price | A stocked oil of that family |
|---|---|---|---|
| floral, citrus, green, fruity, aquatic | Lift | ₹1,099.00 | British Rose — a floral. 50 g ₹212.40 |
| oriental, amber, woody, resinous, balsamic, spicy, gourmand | Longevity | ₹1,296.00 | Woody Oud — a resinous profile. 50 g ₹271.40 |
| complex, layered, sweet, spicy, gourmand, chypre | Depth | ₹1,099.00 | Roasted Coffee — a gourmand. 50 g ₹230.00 |
| gentle, comforting | Softness | ₹1,099.00 | Jasmine — a floral, if you want it softer rather than brighter. 50 g ₹230.00 |
| neutral — any family under test | Pure Performance | ₹1,099.00 | Whatever is under test. 100 g trial pack ₹325.00 |
What the investigation costs and what it protects
The whole audit is cheaper than one remade batch, and most of it is free.
Start with what is at stake, because it sets the scale. A 100 g blend of British Rose at the recommended 20% is ₹244.28 of raw material — ₹175.84 of base from the 500 ml pack and ₹68.44 of oil from the 100 g pack, so the oil is 28.02% of the cost in a fifth of the weight. Fifty of those bottles is ₹12,214.00 of raw material. Add a ₹106.20 bottle and ₹45.00 of reeds and a finished 120 ml unit is ₹395.48, so fifty finished units is ₹19,774.00. That is the number a wrong diagnosis puts at risk.
Against it, the audit. The pour-back is free. The straight-from-the-carton control is free. The white-card and black-card readings are free. The hour you let a bottle come to room temperature is free. What costs money is the equipment that makes the result repeatable: a 100 g trial pack of base at ₹325.00 for control arms, a beaker at ₹300.00 or ₹600.00, and the alcohol spray at ₹383.00 or ₹520.00. The lean kit is ₹1,008.00 and the fuller one ₹1,308.00, or ₹1,445.00 with the litre of spray.
The 100 g trial pack is the right size for this and it is worth understanding why it is also the dearest base per gram. It is ₹325.00 per 100 g against ₹219.80 per 100 ml at the 500 ml bottle — 47.86% more per gram. That premium is the honest price of a small quantity, and it is the reason to use the trial pack for arms and controls and the 500 ml pack for production. The audit blend here is small: a 60 g blend at 20% is 12 g of oil and 48 g of base, poured 20 g into each of three vessels, plus 10 g of base for two sacrificial rinses. That is 58 g of base in total, so one 100 g trial pack covers the whole audit with 42 g to spare — and ₹41.06 of British Rose, ₹146.57 for the blend.
One parcel note, and it applies only to the bases, because they are the items whose packed weight is published reliably. A 500 ml base bottle is packed at 1 kg, and CSI's Standard rate is ₹110.00 to 1 kg and ₹180.00 to 2 kg, so two 500 ml bottles in one parcel save ₹40.00 against two separate orders. If the audit sends you on to a base trial, order both bottles together: you also get both arms mixed on the same day, which is what makes them comparable.
| Stage | What you buy | Cost | What it rules out |
|---|---|---|---|
| 1 · the pour-back | Nothing | Free | The entire vessel branch, or the entire formula branch |
| 2 · the carton control | Nothing — one unwashed bottle from stock | Free | Your own washing process as the cause |
| 3 · white card, black card | Two cards | Free | A background artefact, and a reading you cannot repeat |
| 4 · the reference vessel | 100 ml beaker at ₹300.00 or 250 ml at ₹600.00 | ₹300.00 to ₹600.00 | A different vessel every time, which makes every reading incomparable |
| 5 · the cleaning protocol | Alcohol spray, 150 ml or 1 litre | ₹383.00 or ₹520.00 | Residue and your own rinse water, repeatably, on every bottle |
| 6 · control arms | 100 g trial pack of base | ₹325.00 | The load, if the pour-back pointed at the blend |
| The lean kit | Trial pack, 100 ml beaker, 150 ml spray | ₹1,008.00 | 8.25% of ₹12,214.00 — the raw material in fifty bottles |
Buying for the audit, then for the batch
Two purchases in two stages. The first is small and settles the question; the second waits for the answer.
The order is the whole discipline. Pour-back, carton control, cards, then equipment, then — only if the blend turned out to be the issue — base. The temptation after a bad-looking batch is to order a different base and a different oil the same evening, which converts a free afternoon into a several-thousand-rupee week and still does not tell you whether your bottles were clean. If the pour-back comes back clear, you have just saved the whole base order, and the ₹1,008.00 kit pays for itself on the first batch of ten bottles that all look the same.
Nothing in the kit is wasted if the answer goes the other way, either. The beaker is permanent equipment and the vessel every future clarity reading happens in. The spray is production stock you will use on every bottle from now on. The trial pack becomes the base for a load ladder, and the published step-down — reduce the fragrance percentage and retest — is the first thing you would run anyway. Two outcomes, one kit, and the only thing you spent before knowing the answer was ten minutes.
When the formula is settled and the bottles are clean, the purchasing arithmetic belongs to other pages and they are already live. For quantities, CSI's page on how much base a hundred 100 ml bottles at 20% needs does the base maths, and developing a 100 g trial formula and scaling it covers the step from an arm to a run. For the load economics, the most economical fragrance percentage for a commercial reed diffuser owns that comparison. For the rest of the unit, fibre reeds against natural rattan and the luxury hotel-style reed diffuser are the reads, and one base in bulk or different bases for different families is the range decision.
One last habit, and it is the one that makes this page unnecessary next time. Photograph every batch in the reference vessel and in one bottle, against the same white card, on the day you mix it and at 7 and 30 days. Keep the images with the batch sheet. Comparing this October's photographs against next June's is the only evidence you will ever have about your own products in your own climate, and CSI publishes no ageing data, no shelf life and no cloud point for any base — so your photographs are not a poor substitute for published data, they are the data.
Frequently asked questions
- My reed diffuser is cloudy after mixing fragrance oil with the base — how do I determine whether the formula is physically compatible?
- How to test fragrance oil compatibility before buying 1 kg or 5 kg
- How do I develop a 100 g reed diffuser trial formula and scale it to 25 kg for wholesale production?
- How do I test reed diffuser performance over 30 days?
- My fragrance oil has become cloudy during cooler weather — temperature-related or a quality problem?
- My fragrance oil appears cloudy after a cool room — reversible physical change or deterioration?
- What is the most economical fragrance percentage for a commercial reed diffuser?
- How much reed diffuser base should I buy for 100 bottles of 100 ml at 20% load?
- Should I use fibre reeds or natural rattan reeds with a ready-made diffuser base?
- How do I make a luxury hotel-style reed diffuser for commercial sale?
- How to make long-lasting reed diffusers in Indian weather
- A complete guide to reed diffuser bases
- Pure Performance Base — the complete buyer's guide
- Reed diffusers 101 — everything a new maker needs to know