In an ice water wash, quality is decided by three things: how cold the water stays, how gently it is stirred, and how fast the wet hash dries.
Ice water hash, also called bubble hash, is resin knocked off cold cannabis material in ice water. It is filtered through a stack of mesh bags, then dried and graded. This part covers water and ice, stirring, the wash cycles, collection and drying, the melt grading scale, yield, common faults, storage and pressing.
In Thailand, a cannabis extract above 0.2 percent THC by weight is a Category 5 narcotic, and producing one requires a licence under the Ministerial Regulation on Cannabis or Hemp Extracts B.E. 2569 (2026), in force since 26 April 2026. Ice water hash is far above that line. The method below is written for places where processing is lawful; in Thailand, it is for licensed processors and for readers who want to understand quality.
This is Part 2. Part 1: Ice Water Extraction and Bubble Hash: What Separates a Good Wash From a Bad One
6. Water, Ice and Stirring
6.1 Temperature and water quality
The common target is water as close to freezing as the ice can hold it. Washers usually describe this as below about 4 °C, with ice present the whole time so the temperature never creeps up. This figure is craft practice, not a published parameter.
Washers prefer water low in minerals, such as reverse osmosis water (water pushed through a very fine membrane that removes most dissolved salts) or distilled water. Their reasoning is that minerals and chlorine by-products show up in the finished hash as haze or an off taste. This is also common practice, not measurement.
Water also carries microbes, so clean water matters in the tropics. Ice is used generously: at least as much ice by weight as plant material.
Translator’s note: At 30 to 35 °C outside, ice melts much faster than the temperate guides assume. Work in the coolest room you have, keep extra ice in reserve, and check the water with a thermometer during every cycle, so you know the water really stayed cold.
6.2 Stirring strength is the main quality control
How hard you stir (agitation) is the biggest single factor in quality. Most operators make the same mistake: they stir too hard.
Gentle stirring removes only the heads that are ready to come off. Hard stirring also tears off stalk pieces, leaf tissue and chlorophyll (green pigment). Yield rises, but quality drops sharply.
The same material, the same bags and the same water can give full melt hash or a green paste. The only difference is how hard the slurry was moved.
What to do: Start with the gentlest stirring that still moves the material, and increase it only in later cycles.
7. The Wash
7.1 Method, duration and successive cycles
You can stir by hand with a paddle, with a gentle washing machine, or by stirring in a lined container. The aim is a slow, even circulation that lifts and turns the material, not one that beats it.
The material first soaks in the cold water. The usual pattern is short first cycles of a few minutes, then longer, firmer cycles later.
Each cycle gives a different product:
- First wash: pulls the ripest, most exposed heads. It gives the cleanest, palest result.
- Each later wash: reaches deeper into the material. Yield falls, colour darkens and melt quality declines.
Most washers run three to five cycles and keep the fractions from each cycle separate.
7.2 When to stop
Stop when any of these happens:
- the water coming off turns green instead of amber;
- the extra yield is no longer worth the contamination it brings;
- the material breaks down into pulp.
Running one more cycle to get more yield is the most common way to spoil a good batch. Low-grade hash contaminates every fraction it is mixed with.
What to do: Decide your stopping signs before you start, and keep late-cycle hash in its own container.
8. Collection and Drying
8.1 Breaking up the wet hash
The wet hash collected in each bag forms a flat cake, often called a patty. Rinse each patty with cold water, press it gently to squeeze out free water, then break it down.
Washers usually push the wet hash through a fine sieve or a micro-grater onto a cold surface, to make small, even particles. This increases the surface area, so water leaves fast and evenly.
A thick patty dries from the outside in. The wet centre is where problems start.
8.2 Freeze drying compared with cold air drying
Drying is where most quality is lost. The resin came through the wash intact; drying is the step that costs aroma and gives microbes a chance to grow.
Freeze drying means freezing the hash and then removing the ice under vacuum, so it turns straight to vapour without melting (sublimation).
| Factor | Freeze drying | Cold air drying |
|---|---|---|
| How it works | Frozen water turns to vapour under vacuum | Water evaporates into cool, moving air |
| Typical time | Roughly one to two days | Several days to two weeks |
| Aroma kept | Highest | Lower; long exposure costs aroma |
| Colour | Usually lightest | Often darkens |
| Risk of microbes | Low; frozen, dries fast | Higher; long time at a water activity microbes can use |
| Main obstacle | Cost of the equipment | Needs a cold, dry, clean space |
| Best for | Full melt fractions | Hash meant for pressing |
The times and results in this table are common practice. The original article found no accessible published study of drying parameters.
Air drying needs a cold, dark, low-humidity space with light air movement and no direct heat. Spread the hash thinly and turn it. In the tropics, that means a refrigerated space with dehumidification, not an ordinary room.
8.3 Microbial risk when wet hash dries badly
Wet hash sits at the top of the water activity range where, according to the 2023 review by Gwinn and colleagues described in Part 1, most fungi grow best. A dense patty left at room temperature in humid air spends hours or days in conditions where those fungi thrive.
Such a batch cannot be saved. Heating it later does not destroy mycotoxins (mould toxins) that have already formed, including ochratoxin A.
What to do: Dry cold, fast and thin. If you cannot keep wet hash cold until it is dry, do not start the wash.
9. Grading: The Melt Scale and Its Limits
Hash is commonly graded on a one-to-six star “melt” scale, based on how it behaves when heated. This scale is practitioner consensus. No standards body stands behind it.
| Rating | Common name | Behaviour under heat | Residue left |
|---|---|---|---|
| 1 star | Food grade | Does not melt; chars | Heavy, black |
| 2 star | Food grade | Softens, may bubble slightly | Heavy |
| 3 star | Half melt | Partial, uneven bubbling | Noticeable |
| 4 star | Half to full melt | Bubbles, but not completely | Light to moderate |
| 5 star | Full melt | Melts cleanly, bubbles steadily | Minimal |
| 6 star | Full melt, top tier | Melts fully and evenly | Almost none |
“Food grade” here means hash too impure to melt cleanly, traditionally kept for cooking. “Half melt” means it melts only partly.
What the scale really measures is how pure the population of intact resin heads is. Plant fragments, ruptured heads and cell debris do not melt, so they char and leave residue.
The Wikipedia article on hashish cited in Part 1 agrees. When a high level of pure THC is present, the product is almost transparent and starts to melt at the warmth of a human touch. The same article gives the smell test: good hash smells fragrant and aromatic, while poor material smells clearly mouldy or musty.
Other checks used on the bench:
- Colour: from near-white, through cream, to amber.
- Texture: dry and sandy.
- Clumping: the hash sticks together under gentle warmth.
9.1 How much of the scale is subjective
A great deal. Colour does go with quality, but it also depends on the cultivar, and a darker fraction can perform better than a paler one.
Melt is judged by eye, on equipment that is not standardised, and often by the person who made the hash. The scale is a shared vocabulary, not a measurement. Two competent people can rate the same material one grade apart.
What to do: When you compare batches, have the same person rate them on the same equipment, and record smell and residue alongside the star rating.
10. Yield Expectations
Always quote yield as a range and state what it is measured against. Fresh frozen material is mostly water, so the same percentage means very different things against fresh frozen weight and against dry weight.
For a related solventless method, the Wikipedia article on hash oil gives a comparison: 0.3 to 8 percent yield for rosin, against 10 to 20 percent for extraction with hydrocarbon solvents. Ranges for solventless methods are genuinely wide.
For ice water hash itself, the original article found no accessible published dataset. The figures that circulate are practitioner consensus:
- poor material: low single digits, as a percentage of fresh frozen weight;
- competently grown material: higher;
- material grown specifically for hash: higher still.
Only part of that total lands in the full melt grades. Total yield and full melt yield are two different numbers.
The same article notes that growing for hash is different from growing for flower, and that some cultivars give a misleading impression of how much they will yield.
What to do: Record every batch with its yield, the basis (fresh frozen or dry weight), and the full melt share separately, so you can compare cultivars honestly.
11. Faults and Their Causes
| Fault | Likely cause | Correction |
|---|---|---|
| Green colour | Chlorophyll from torn leaf; stirring too hard | Stir more gently; keep the slurry colder |
| Visible plant matter | Work bag too coarse; too many cycles | Better pre-filtering; stop earlier |
| Poor melt, heavy residue | Debris that does not melt; immature heads | Judge each fraction separately |
| Sour or fermented smell | Wet material kept warm; slow drying | Dry cold, thin and fast |
| Hay-like or musty smell | Contaminated input; mould | Reject the input |
| Dark colour but good aroma | Oxidation, late cycles, or the cultivar | Judge by melt and smell, not colour |
| Very low yield, but clean | Stirring too gentle, or a low-resin cultivar | Test with a cultivar you know |
| Gritty, will not press | Drying not finished; hard water | Finish drying; use softer water |
12. Storage, Ageing and Pressing
12.1 Storage
Store fully dried hash cold, dark and airtight. Let the container reach room temperature before opening it, so that moisture from the air does not condense on the cold hash.
The enemies are heat, light, oxygen and leftover moisture. Moisture is the most serious. Hash that was not fully dried can support microbial growth, not just lose quality.
Translator’s note: In a humid climate, opening a cold container straight away is the fastest way to add water to dry hash. Wait until the container has warmed up, and open it in the driest room available.
12.2 Temple balls
A temple ball is a traditional way to finish hash. Loose resin is worked by hand with gentle warmth and pressure until the individual heads burst and merge into one even mass. The Wikipedia article on hashish describes the same change: pure sieved resin becomes gooey and easy to shape under heat.
Pressing has three effects:
- It pushes out air, which slows oxidation.
- It mixes the whole batch into one even product.
- It changes the character. Bursting the heads releases aroma compounds, and pressed hash is often described as rounder and less bright than the same resin left loose.
So pressing is a trade-off, not an upgrade.
12.3 Hash rosin
Pressing dried ice water hash into hash rosin, including temperatures, pressure per bag area and realistic yields, is covered in Rosin: Solventless Extraction with Heat and Pressure.
13. What Actually Separates a Good Wash
- Cold that never breaks.
- Clean starting material.
- Stirring so gentle it feels unproductive.
- Each fraction judged on its own, not mixed together.
- Drying treated as the most important step, not the last one.
Almost every failed wash comes from pushing for yield with force, or from rushing the drying. These are problems of patience, not of technique.
What to do: On your next batch, write down the water temperature, the number of cycles, the drying method and the drying time, so you can trace any fault back to its step.
Discuss below: Which bag sizes are in your stack, and which bag gives your best fraction most often? If you work in a high-humidity area, how do you handle the wet hash between collection and drying? If you have washed the same cultivar both fresh frozen and dried, what changed? If you have measured data rather than impressions, post it here: this topic has far more folklore than measurement.
References
- Nishimura & Asahi (law firm). 2022 briefing on the Thai Ministry of Public Health notification of 9 February 2022 (extract exemption at 0.2 percent THC by weight).
- Tilleke & Gibbins (law firm, Bangkok). Commentary on the Notification of the Ministry of Public Health Re: Controlled Herbs (Cannabis), B.E. 2568.
- Gwinn, Leung, Stephens and Punja (2023). Fungal and mycotoxin contaminants in cannabis and hemp flowers: implications for consumer health and directions for further research. Frontiers in Microbiology.
- Montoya, Conroy, Vanden Heuvel, Pauli and Park (2020). Cannabis contaminants limit pharmacological use of cannabidiol. Frontiers in Pharmacology.
- Wikipedia, “Hashish” and “Hash oil” (articles quoted in this guide).
This article is part of the forum’s educational series on extraction and post-harvest processing. It is for education only, and the processing steps apply only where cannabis processing is legally permitted.