Wood pellets for smoking: what actually changes the smoke
A pellet grill is engineered to burn wood cleanly, and smoke flavour is a product of burning it badly. That contradiction decides everything in the bag, and species is not the variable people think it is.
A pellet grill is a machine built to burn wood well. An auger meters fuel a few grams at a time, a fan pushes air through a small burn pot, and the fire runs with air to spare. That is combustion engineering doing exactly what it was asked to do, and it is why the thing will hold 225 °F overnight with nobody standing in the yard.
Smoke flavour is what comes off wood that is burning badly. The compounds people are actually buying, the guaiacols and syringols that read as smoke on the tongue, are products of pyrolysis at the edge of a starved flame. Clean combustion finishes them into carbon dioxide and water, which taste of nothing. So the appliance's central virtue is the thing standing between the buyer and what the buyer came for.
Everything on this page follows from that one sentence, including the awkward part: the decision people agonise over in the aisle, apple against cherry against a competition blend, is close to the least important thing in the bag. What lands on the meat is decided by the set temperature, by how much water the pellets have picked up, and by how much of the bag is dust. Species comes fourth, and it is not a close fourth.
What is tested here, and what is not
Analyzed. Pellet manufacture, moisture specification and ash behaviour come from published fuel-pellet standards and wood science. The account of where smoke compounds come from is combustion chemistry. Auger behaviour is described from how the control loop is documented to work, not from a controller we have opened.
Tested. Nothing yet. The figures further down are blank until the readings exist, and what they will hold is a relationship between storage humidity and pellet integrity, not a ranking of brands. Method at How We Test.
The dial on the front is the smoke control
A pellet controller does not modulate a flame the way a gas valve does. It runs the auger on a duty cycle: feed for a couple of seconds, wait, feed again. The gap between feeds is what sets the fire.
At 180 to 225 °F the gaps are long. The burn pot cools between feeds, the newest pellets land on embers rather than flame, and they spend a while smouldering before they catch. Smouldering is incomplete combustion, and incomplete combustion is smoke. At 275 °F and above the feed is nearly continuous, the pot stays hot, pellets ignite on arrival, and the fire is clean enough that very little survives to reach the food. Same bag, same wood, same machine, opposite result.
This is what the smoke settings on newer units are doing. They are not adding anything. They are deliberately running the fire worse, usually by cycling the fan and stretching the feed interval, and the reason they only work below a temperature ceiling is that above it there is no way to keep the fire dirty and hold the setpoint at the same time.
The practical version: if a cook has produced no flavour, the first thing to look at is not the bag. It is the fact that the cook ran at 275 °F because that is what the recipe said. Run the first two or three hours low, let the fire misbehave while the surface is still cold and receptive, then raise the setpoint to finish. That sequence is worth more than any species on the shelf.
Moisture is the other half of the fire, and it is the half you control
Pellets leave the mill at roughly 6 to 8 percent moisture. Seasoned splits sit at 15 to 20 percent, and that gap is the whole reason a pellet burns hot and consistently: there is almost no water to boil off before the wood can pyrolyse. Consistency in a pellet grill is a moisture story before it is anything else.
Wood is hygroscopic, and a pellet has been ground, dried and compressed, which leaves it far more eager to take water back than the log it came from. Left open in humid air it does not simply get damp. It swells, the compression relaxes, and the pellet reverts toward the sawdust it was made from. A hopper of swollen pellets and loose dust is an auger jam, and an auger jam on an overnight cook is a temperature crash discovered at six in the morning.
Which reframes the storage advice everyone repeats without explaining. A sealed bucket is not about keeping pellets tidy. It is about protecting the one property that makes the fuel work. My honest position is that where a bag has been kept tells you more about how a cook will go than what is printed on the front of it.
Fines are the variable nobody prints on the bag
Every bag contains some dust. Pellets abrade against each other in the bag, in transit, and again every time the hopper is filled. That dust is called fines, and it does two things, both bad. It packs around the auger flights and changes how much fuel a given rotation actually delivers, so the controller's idea of the fire drifts away from the fire. And it burns fast and dirty in a way that has nothing to do with the useful kind of dirty, adding ash without adding much heat.
Ash content is the closest thing the industry has to a quality number, and it is rarely on the bag. Below about one percent is good. Above two and the burn pot needs clearing more often than most people clear it, which is its own temperature problem: a pot half full of ash is a pot with less air moving through it, and the controller compensates by feeding more fuel into a space that cannot burn it.
None of this is visible in the aisle. The one thing a buyer can do at home is tip the last third of a bag through a coarse sieve and look at what falls out, which takes a minute and says more about the product than the flavour name.
| Variable | What it changes | Range in normal use | Who controls it | What to do |
|---|---|---|---|---|
| Set temperature Dominant | How complete the combustion is, therefore how much smoke survives | 180 to 300 °F | You, on every cook | Run the first hours low, finish hot |
| Pellet moisture | Burn consistency, auger reliability, whether the pellet holds together at all | 6 to 8 percent, new | You, through storage | Sealed container, not the paper bag |
| Fines and ash | Fuel delivered per auger turn, air through the burn pot | Under 1 percent ash is good | The mill, then handling | Sieve the bag end, clear the pot |
| Species on the label | A mild shift in the aromatic fraction, on top of the base wood | Small | The blend, not you | Pick one you like and stop worrying |
Ranking is an argument from combustion behaviour, not from a taste panel. Ash and moisture figures are the published specifications for fuel-grade pellets. Nothing in this table has been measured by us yet, and the row that will be measured first is the second one.
No binders added is a fact about lignin sold as a feature
It is printed on nearly every bag, and it is true, and it means almost nothing. Wood already contains its own adhesive. Lignin is the polymer that glues cellulose fibres together in the standing tree, it softens somewhere around 140 °C, and a pellet mill works by forcing warm sawdust through a steel die under enough pressure to generate exactly that heat. The lignin flows, the pellet cools, the lignin sets. That is the entire process.
So a mill that added a binder would be a mill that had failed to run its die hot enough or its feedstock at the right moisture. Claiming no binders is like a bakery claiming no added gluten. The absence is not a virtue anyone worked for, and it is worth naming because it is the kind of statement that trains buyers to trust the front of a bag.
The genuinely useful consequence sits one step further on. Since the glue is the wood, woods with different lignin behaviour pelletise differently, and that is why the base of almost every bag is what it is.
100 percent hardwood is true, and it is not the claim you read
A bag labelled apple is very unlikely to be apple. In most cases it is a base of oak or alder carrying a minority fraction of the named wood, and the label is accurate because oak and alder are hardwoods. The claim on the front is about the botanical class of the fibre, not about the proportion of the tree in the picture.
This gets treated as a scandal, and it is not one. Oak and alder are chosen because they pelletise cleanly, burn at a predictable rate and leave little ash, and because fruit wood in the volumes a mill needs does not exist. A pure apple pellet would be a worse fuel and an unreliable one. The base wood is doing the job the machine needs done, and the flavour wood is riding on it.
What is fair to criticise is the packaging, which shows an orchard and says nothing about the blend, and a handful of mills that publish their ratio while most do not. What is not fair is the video-shaped version of this, the one that counts a bag as fake because it contains oak. Both sides of that argument are fighting over the fourth most important variable in the fuel.
The comparison worth making is with the same woods in solid form, where the proportion is whatever you put in the firebox. That is the real difference between a pellet grill and a stick burner, and it is the same trade the chips, chunks and splits question turns on: the smaller the fuel, the less control you have over the fire it makes.
| On the bag | Usual base | What the named wood adds here | Same wood as a split |
|---|---|---|---|
| Apple | Oak or alder | A slight sweetness, easily lost above 250 °F | Dense, slow, dilutes without shrinking the fire |
| Cherry | Oak or alder | Colour rather than flavour, and less of it than a split gives | Light, fast, needs oak underneath it |
| Hickory | Often mostly hickory | The one blend where the name usually dominates | Strong enough to overrun a long cook |
| Mesquite | Oak, with a small mesquite fraction | Enough to notice, far less than the solid wood | Aggressive, short cooks only |
| Competition blend | Oak base, several flavour woods | A hedge. Reasonable default if you are undecided | No equivalent |
Base woods are the industry norm rather than a rule, and a few mills do sell single-species pellets at a price that reflects the difficulty. Nothing in this table is a measurement. The split column links to pages where the same species is discussed as a solid fuel, where its properties actually get to act.
The October spike is not barbecue, and it matters
Search demand for smoking wood follows the grilling season and falls off a cliff in autumn. Demand for wood pellets does the opposite: it runs flat through summer and then roughly triples into October, with a second rise in December. Nothing about barbecue explains that shape. Heating season does.
A large share of the pellets sold in the world are heating pellets for stoves, and they are a different product wearing the same shape. Softwood is normal in them, bark is permitted in the lower grades, and neither is screened for anything to do with food. They are also, at a glance in a hardware aisle in November, indistinguishable from cooking pellets and usually much cheaper per kilo.
Do not put them in a food grill. Not because a stray softwood pellet is dangerous in any dramatic sense, but because the resin content makes sooty, acrid smoke that tastes the way a chimney smells, because bark carries whatever the tree was standing in, and because nothing in that supply chain was ever asked to be clean. A bag that does not say food grade or barbecue on it is a heating pellet until proven otherwise, and the price is usually the tell.
What we are measuring, and what is still blank
Everything above says storage decides more than species, which is easy to assert and worth putting a number on. The measurement below is the one that does it, and it needs a kitchen scale and an oven rather than a smoker, so it is not waiting on anything except a bag of pellets we can name.
It is also being run in the worst climate on earth for the question, which is the point. If a pellet survives a wet season here, the storage advice everyone repeats is more cautious than it needs to be. If it does not survive a week, the advice is not cautious enough.
Pellet mass against storage humidity: planned, not yet recorded
Protocol. Three samples of twenty pellets each from the same bag, weighed to 0.01 g. One sealed in a jar with desiccant as the control, one open on a shelf indoors, one open in an unconditioned room during the wet season. Ambient temperature and relative humidity logged hourly beside each sample. All three weighed daily at the same hour on the same scale, and photographed against a fixed rule so swelling can be read as length as well as mass. Starting moisture is fixed by oven drying a fourth sample to constant mass at 105 °C, which is the same gravimetric method used for splits on this site rather than a pin meter. Integrity is scored by dropping a pellet 300 mm onto a hard surface and recording whether it survives intact, which is a crude test and a repeatable one. What is measured here is the relationship between storage air and fuel condition, not a comparison of brands, and the brand and composition of the bag are recorded because a heating pellet would answer a different question.
The figures and the photograph are published once this protocol has been run on the bench. Until then this page carries the method, not the numbers.
The table stays empty until the readings exist. If the humid sample gains almost nothing over a month, this page will say so and the storage advice on it will get shorter.
Where it fits
A pellet grill is the easiest way to cook something for twelve hours and one of the harder ways to get a lot of smoke into it, and those two facts are the same fact. The machine trades flavour for consistency, which is a trade a great many people are right to make, and the mistake is trying to buy the flavour back at the shelf instead of taking it back at the controller.
If you want the fuel to carry more of the load, the answer is a fuel that burns badly on purpose, which means chunks or splits in something that lets a fire misbehave, or charcoal with wood on top of it. If you want the machine, keep the machine and stop optimising the bag. The species question is a real one, and a pellet hopper is the place where it matters least.
What I would take from this page is one habit rather than a purchase. Buy a bag, decant it into something that seals, and spend the attention you were going to spend comparing flavour names on the first three hours of the cook instead. That is where the smoke is, and it is free. More on how this site decides what counts as a measurement is on How We Test, and who is behind it is on About.
