Four reasons an oven stops holding temperature

Start with what the display means

The number on the front of the oven is what the control believes, derived from a resistance reading off one sensor in one spot in the cavity. It is not a measurement of your food’s environment and it was never meant to be.

So the first job on any temperature complaint is to stop arguing with the display and measure the cavity independently, over at least three full cycles, with the door shut and the oven loaded the way you actually use it. What comes back sorts the fault into one of four cases.

Case 1: uniform offset

The oven reads 350 and holds a steady 315, or a steady 380. The error is the same at every setpoint and stable across the whole cycle.

This is the sensor, nearly always. Oven temperature sensors are resistance devices with a known curve; a common type reads in the region of 1,080 ohms at room temperature. Measure it cold with the oven at ambient. A long way off the expected value and the control is being lied to consistently, which is exactly what a uniform offset looks like.

Small uniform errors can often be corrected with the oven’s own calibration offset. Large ones mean a sensor.

Case 2: overshoot and sag

The oven blows past setpoint on the way up, then falls short at the end of a long bake. The average might even look correct, which is why owners describe it as inconsistent rather than wrong.

This is a heat delivery problem, not a measurement problem. On an electric oven, look at the elements: one element open leaves the oven trying to hold temperature on the other, which changes the whole cycle shape. On a gas oven, look at the valve. A valve that opens late and then delivers full rate produces exactly this signature.

Case 3: slow preheat

Preheat that used to take twelve minutes now takes twenty-five, and everything else looks normal.

Check the bake element draw first. An element with a partially failed section still glows and still draws current, just less of it. On a double oven, check whether the second cavity is behaving the same way, because that separates an appliance-wide supply problem from a single element.

If the elements are fine, look at the door. A gasket that no longer seals at the top is a continuous heat leak, and preheat is where you notice it first.

Case 4: heat loss over a long bake

The oven reaches temperature normally and then cannot hold it for a two hour roast.

This is the door, most of the time. Take a strip of paper, shut it in the door at four points around the frame, and pull. Where it comes out with no resistance, the gasket is not sealing. Then check whether the door is square, because on a heavy professional door the usual cause of a bad seal is a hinge that has settled rather than a gasket that has worn.

Hinges go in as a pair. Replacing one twists the door and moves the problem to the other corner.

The self-clean question

Self-clean drives the cavity far above any cooking temperature and it finds every weak component in the oven. On a newer machine that is fine, it is what the cycle is for. On an oven over about fifteen years old it is a genuine risk to the thermal fuse, the thermal limit and the control board, and we see ovens that ran a clean cycle and never woke up.

If your oven is that age and it needs cleaning, clean it by hand. It is a bad afternoon rather than a dead oven.

What we do

We log the cavity with an independent probe over three cycles before quoting, measure sensor resistance cold, check element draw on both bake and broil, and put a level on the appliance. That sequence costs one diagnostic visit and it stops customers buying the wrong part, which is the expensive mistake in oven work.