Scope. This page provides a controlled reader-run protocol and does not claim site-run measurements. For cooking, cooling, storage and reheating, follow the current official guidance linked on the page.
Somewhere in the recipe it says to hold the pan at a low temperature and keep it there. You turn your hob to its lowest setting, and within a few minutes the pan is well past where it was supposed to be. There is nowhere lower to go. The instruction appears to require equipment you do not own.
You almost certainly do own it. The equipment is a second pan, or an oven, or in the simplest case just patience and something heavy. What none of these will give you is the steadiness of a controlled water bath, and this article is not going to pretend otherwise. What they will give you is a temperature that stays inside a usable band for long enough to do the job, which is what the recipe actually needed.
Why your lowest setting is still too hot
Domestic hobs are designed around the assumption that people mostly want to boil, fry and reduce. Delivering a small, precisely controlled amount of power is a different engineering problem and it is not the one they are optimised for.
Each hob type falls short in its own way. Gas hits a floor set by flame stability, below which the flame will not sustain. Radiant electric keeps radiating for a while after you turn the control down, and its residual heat sets a floor of its own. Induction can produce a low average, but many hobs get there by cycling full power on and off, so the pan base sees peaks rather than the gentle heat the display implies. We take all three apart in how the three hob types behave at low heat.
There is a second problem underneath the first, and it belongs to the pan rather than the hob. A thin pan has very little thermal mass, so it follows its heat source almost immediately. Whatever the hob does, the pan does, and whatever the pan does, the food touching it does. A thin pan is a faithful transmitter of every flaw in the appliance below it.
That observation is the key to three of the four methods below. All of them work by inserting something between the hob and the food that has more thermal inertia than the food does. Metal, water, air, or the pan itself. The mechanism is always the same: give the system enough mass that it cannot change temperature quickly, and the swings disappear into it.
How to measure stability
Stability is not a spot reading. A single probe check on a system that is oscillating tells you what the temperature was at the instant you happened to look, which is close to useless.
So the measure is drift and swing across a stated period. For each method, log liquid temperature continuously across a thirty-minute hold and record three things: the total drift from start to finish, the peak-to-trough range across the hold, and the temperature at the pan base as distinct from the temperature of the contents. That last one is where scorching lives; a sauce can be at the right temperature in the middle and catching at the metal.
Alongside that, a fourth figure worth having: time to settle. A method that eventually becomes very stable but takes twenty minutes to get there is not much use for a sauce you are holding for fifteen.
Instruments calibrated before and after each session. Same pan, same volume, same ambient, multiple replicates, spread recorded. Discarded runs logged with their reasons.
This page gives no stability figure for any method; every value is yours to record on your own equipment.
Method one: a diffuser between burner and pan
A diffuser is a plate that sits between the heat source and the pan. Heat goes into the plate, spreads through it, and reaches the pan base more evenly and less intensely than it would directly.
What it does. Two things at once, which is why it works. It lowers the effective floor, because some of the energy is lost from the plate to the room rather than reaching the pan. And it damps, because the plate's own mass has to change temperature before the pan can, which smooths out both hot spots and cycling.
Where it helps most. On gas, where the hot ring of flame contact becomes a hot spot on a thin pan base. On any hob where the problem is unevenness rather than raw output.
Its limits. It is a passive device with no feedback, so it moderates the heat without controlling it. If your hob's floor is far above your target, a diffuser may not close the whole gap. It also adds a lag in both directions, which is fine for holding and irritating if you need to change the temperature quickly.
Is it worth it? That is a measurable question, including whether a cheap improvised alternative does the same job, and what a diffuser actually does sets out how to answer it rather than asserting it here.
Record this method's drift, swing and time to settle on your own equipment.
Method two: a water bath on the hob
Put a pan inside a larger pan of water, and heat the water rather than the food. The classic bain-marie.
What it does. Water is an excellent thermal buffer. It takes a lot of energy to change its temperature, so it responds slowly and smoothly to whatever the hob is doing underneath, and the inner pan sees a gentle, even, all-round heat rather than a hot patch of metal at the bottom.
It also gives you a hard ceiling. Water cannot exceed its boiling point at ambient pressure, so as long as there is water in the outer pan, the inner pan cannot be driven above it no matter what the hob does. For anything that fails catastrophically above a certain point, that ceiling is genuinely valuable.
Where it helps most. Delicate sauces, custards, melting, and anything you need to hold rather than cook. It is the single best answer to induction cycling, because the water absorbs the peaks entirely.
Its limits. Setting your target below the ceiling still requires controlling the water temperature, which means a probe in the water and attention to the hob. It is slow to change. It takes two pans and a certain amount of faff. And water evaporates, so a long hold needs topping up, ideally with hot water so you do not crash the temperature.
Cost: nothing, if you own two pans that nest. This is the method to reach for first.
For getting more out of it, including how steady an improvised bath actually is and where the point of diminishing returns sits, see building a water bath from what you own.
Record this method's drift, swing and time to settle on your own equipment.
Method three: the oven as a holding box
An oven is a large, insulated, slow-moving thermal environment. That makes it a mediocre precision cooker and a rather good holder.
What it does. Surrounds the pan with warm air on all sides rather than applying heat to one face. There is no hot spot at the base, so there is nothing to scorch against, and the sheer volume of the cavity means its temperature moves slowly.
Where it helps most. Long holds. Braises. Anything you want to leave alone for hours. Also, resting and holding several components at once, which is the main way a home cook can copy the professional habit of finishing to order.
Its limits. Domestic ovens are frequently inaccurate at low settings, and the dial and the cavity can disagree substantially. Ovens also cycle their element, so the cavity temperature oscillates around the set point rather than sitting on it. And many ovens simply will not go low enough; their lowest setting is above the band you want.
The fix for the first two is a separate oven thermometer left in the cavity, so you calibrate the dial against reality once and know your own offset thereafter. The fix for the third is to prop the door slightly ajar, which lowers the cavity temperature at the cost of making it draught-dependent and much less predictable.
Cost: nothing, assuming you have an oven, though verifying it properly wants a thermometer you can leave inside.
Record this method's drift, swing and time to settle on your own equipment.
Method four: off-heat coasting with a heavy pan
The most primitive method, and the one that reveals the underlying principle most clearly.
What it does. A heavy pan stores a lot of heat. Bring it to temperature, take it off the burner, and it releases that stored heat into its contents slowly as it cools. You are using the pan as a battery. When the temperature falls below your band, you return it to the heat briefly, then take it off again.
This is where the thermal-mass argument becomes concrete. A thin pan coasts for almost no time at all, because it has stored almost nothing; take it off the heat and it is cool within a minute or two. A heavy pan can hold a workable temperature for a meaningful period unattended. The difference is not marginal, and it is the strongest practical argument for owning one heavy pan.
Where it helps most. Short holds. Finishing a sauce, where you deliberately want the heat falling rather than rising. Situations where the hob's floor is hopeless and you have nothing else to hand.
Its limits. It only goes one direction: down. It requires attention and repeated intervention. And it is entirely dependent on the pan; with a thin one, this method does not exist.
Cost: nothing at all. No purchase, no equipment, no setup. If you own one heavy pan you already have everything you need.
Record this method's drift, swing and time to settle on your own equipment.
Which method for which job
| Job | Best method | Why |
|---|---|---|
| Holding a butter emulsion through a meal | Water bath | Hard ceiling below the break point, no hot spot at the base |
| Cooking a custard gently | Water bath | Even, all-round heat and a ceiling that limits the damage of a lapse |
| Holding several components before serving | Oven | Volume, and it holds everything at once |
| A long braise below a simmer | Oven | Slow, unattended, no base hot spot |
| Fixing a hot spot on a thin pan over gas | Diffuser | Spreads the flame ring across the base |
| Finishing a sauce off the heat | Coasting with a heavy pan | Temperature falling by design |
| Induction cycling scorching your sauce | Water bath | Water absorbs the peaks completely |
The table ranks by mechanism. To rank by stability, fill in each method's drift and swing from your own runs and rank by those, not by price; if the free method wins, use it.
An honest limit, and a safety note
None of these four is as stable as a properly controlled water bath with active feedback. A controlled bath measures its own temperature continuously and adjusts its power to correct any deviation. Everything in this article is open-loop: it moderates the heat but has no way of knowing whether it has hit the target. The difference shows up over long holds, where a small drift accumulates.
That is not an argument for buying anything. For most jobs, a band is what you need rather than a set point, and a bain-marie will keep you inside a band for as long as you keep an eye on it. It is an argument against pretending the methods are equivalent when they are not.
Separately: holding food warm over a period is a food-safety matter as well as a texture matter, and the time-and-temperature rules that govern it come only from named food-safety authorities, not from us; check your own authority's current guidance before holding anything for long. Nothing in this article should be read as guidance on how long anything may be held.
In Plain English
Holding a low temperature means keeping food inside a narrow band for a period, usually because it is finished and waiting rather than still cooking. Most domestic hobs cannot do it directly, because their lowest setting is above the band or because they deliver low power in bursts.
The fix is to put something with thermal inertia between the hob and the food. Water in an outer pan is the best general answer and costs nothing if you own two pans. An oven is the best answer for long holds and for several things at once. A diffuser helps with hot spots on gas. A heavy pan taken off the heat will coast for a useful period with no equipment whatsoever.
This matters to anyone cooking sauces, custards, emulsions or anything gentle enough that the difference between holding and cooking is the difference between dinner and a wasted afternoon.
What to do next: before you start the dish, decide which of the four you are going to use, and set it up first. Improvising a holding method after the sauce is already made is how sauces get lost.
Where to go from here
Protocols for testing diffusers, pans and probes on your own equipment are in what a diffuser actually does, pan thermal behaviour and six thermometers, tested against ice and steam. Where a free workaround matches a purchase, the free one wins.
If you would rather work from method than from equipment, the technique library covers holding, water baths and low-temperature work as techniques in their own right, each with its working range and its failure mode.
Related technique: The Four Temperature Bands Every Cook Should Know.