Solenoid coil wire gauge and turns: the tradeoff, and the one point that never moves
Inside a fixed winding window, ampere-turns go with wire area and copper loss goes with its square, so the same window at the same power gives the same force whichever gauge you wind. A 0.20 mm coil at 24 V and a 0.40 mm coil at 6 V both land at 877 ampere-turns and 3.0 W. Gauge sets the supply and the driver; the window sets the force.
Why this happens
The winding window is a fixed area determined by the frame, the bobbin and the wall thickness. Into it you can put round wire with a fill factor of 0.45 to 0.55 for a random wind, or 0.60 to 0.70 for a layered or orthocyclic wind. What matters is that the copper cross-section available is set by the window and the fill, and not by which wire you choose to put there.
That single fact produces the result that surprises people. Write the winding down from first principles, with window area A, fill factor f, wire area a, mean turn length l and copper resistivity ρ:
- the number of turns is N = fA / a
- the resistance is R = ρ N l / a
- the current at voltage V is I = V / R
- therefore the ampere-turns are N I = V a / (ρ l)
Ampere-turns go with wire area, so doubling the diameter quadruples them. But resistance falls with the square of area, so the current rises by a factor of sixteen. Power rises with it, and the two effects cancel exactly where it counts.
A worked pair makes this concrete. Take a window of 40 mm by 10 mm, which is 400 mm², a fill factor of 0.55, giving 220 mm² of copper, and a mean turn length of 50 mm.
| 0.20 mm wire | 0.40 mm wire | |
|---|---|---|
| Conductor area | 0.0314 mm² | 0.1257 mm² |
| Turns | 7003 | 1751 |
| Resistance | 192 Ω | 12.0 Ω |
| Drive to reach 3.0 W | 24 V at 0.125 A | 6.0 V at 0.500 A |
| Ampere-turns | 877 | 877 |
| Current density | 3.98 A/mm² | 3.98 A/mm² |
Same ampere-turns. Same power. Same current density, so the same steady-state temperature. The two windings are interchangeable, and the only difference is whether the drive is 24 V low-current or 6 V high-current. Ampere-turns vary with the square root of power, so at a fixed window, doubling the power raises force by a factor of two while the circuit is unsaturated.
There is a more compact way to write the same result. Ampere-turns equal current density multiplied by the copper cross-section, and the copper cross-section is the window area times the fill factor. So:
NI = J × f × A
At the thermal limit, J is fixed by how fast the coil can shed heat. That makes ampere-turns a function of the window and the fill, and of nothing else. The wire gauge has left the equation entirely.
Put the worked example through the shorter form. At 3.98 A/mm², the 220 mm² of copper gives 876 ampere-turns, which is the 877 the two windings reached from opposite ends of the gauge range. Raising the sustainable current density to 6 A/mm² takes it to 1,320 ampere-turns, which is 50 % more, and force rises by 2.25 times while the circuit is unsaturated. The power needed rises with it, from 3.0 W to 6.8 W. Extra force is bought by removing more heat, never by choosing different wire.
The timing behaves the same way. Inductance is proportional to the square of turns, and resistance is proportional to turns, so the L/R time constant works out as:
L / R = μ N A_core / (ρ l_core l)
Substituting N a = fA, which is constant at a fixed window, removes both turns and wire area from the expression. The time constant depends on the core cross-section, the core length, the copper area and the mean turn length. Rewinding with a different gauge leaves it unchanged. In the worked pair, moving from 1,751 turns to 7,003 turns multiplies the inductance by 16 and the resistance by 16, so the ratio that sets the rise time is identical. The complete behaviour of the winding collapses into one product, voltage multiplied by wire area: that sets the steady ampere-turns, sets the rise shape against an unchanged time constant, and sets the copper loss you have to remove. Gauge only decides which voltage and current pair you use to reach it.
What gauge does change is the small stuff. Enamel is applied as an approximately fixed thickness regardless of conductor size, so it takes a larger share of a small conductor. Between AWG 30 at 0.255 mm bare and AWG 40 at 0.080 mm bare, both in grade 2, the copper fraction of the window falls from about 0.77 to about 0.72. That is roughly 7 % of the available copper, and finer wire packs into a random wind slightly better, which partly offsets it. The two effects are the same order and they pull against each other, so this is not the lever it is sometimes described as. The real constraint on going finer is handling: below about 0.10 mm, winding tension control becomes the limit, and tension is also what damages the enamel.
Check these in order
1. Establish the thermal ceiling before the winding. Estimate the current density you can sustain in the actual mounting, with the actual enclosure. Continuous coils in free air typically sit around 4 to 6 A/mm²; short-pulse duty can run far higher, tens of amperes per square millimetre for milliseconds. This number, not the wire, decides the ampere-turns.
2. Measure the window, and confirm the fill you can actually achieve. A quoted 0.65 is a layered wind on a clean bobbin. Production will deliver something less, and the difference shows up directly in force. Winding limits for temperature are covered in coil temperature limits.
3. Choose the gauge from the supply and the driver. Decide the voltage rail and the switching device first, then pick the wire that lands the current where the driver is comfortable. This is the only decision the gauge actually makes.
4. Confirm the current density against the duty. Compute the current from the chosen voltage and resistance, divide by conductor area, and compare with the figure from step one. In the worked example, 0.125 A through 0.0314 mm² is 3.98 A/mm², and 0.5 A through 0.1257 mm² is the same number, which is the invariance appearing as a check rather than as a coincidence. If the answer is over the limit, the fix is more copper or better cooling rather than a different gauge.
5. Check the voltage tolerance band. A 192 Ω winding on a 24 V rail with a plus or minus 10 % tolerance swings the power by about 21 %, because power varies with the square of voltage. Resistance itself also moves with temperature, and copper gains 0.393 % per kelvin. Both effects are covered from the force side in calculating holding force.
6. Check the response requirement against the drive. Since the L/R time constant is fixed by geometry, speed is bought with drive voltage. If the requirement is tight, the answer is a higher pull-in voltage or a capacitor-assisted drive rather than a rewound coil, as set out in solenoid response time.
7. Check that the end-of-stroke force still wins. The winding decides what the coil can produce. Whether that is enough depends on the gap and the return spring at the end of the stroke, which are separate questions from gauge.
What actually to change
| Symptom | What to change | Why not the obvious thing |
|---|---|---|
| Force short at the required power | Increase the window, or raise the sustainable current density | Rewinding to a different gauge at the same power changes nothing measurable |
| Coil too hot at the force you need | Improve the conduction path from winding to mounting | Reducing current density means reducing ampere-turns with it |
| Actuation too slow | Raise the pull-in voltage or add a capacitor-assisted drive | More turns raises inductance and slows it further |
| Driver cost too high | Move to finer wire and a higher rail to cut the switched current | The ampere-turns and the force stay the same |
| Window will not take the copper | Review the fill factor and the bobbin wall first | Changing gauge moves copper between enamel and conductor, not into the window |
| Force varies between units | Check gap and friction before winding tolerances | Winding repeatability is usually better than assembly repeatability |
When it IS the harder problem
The window is already full and the thermal ceiling is already reached. Then there is no winding left to change, and force has to come from somewhere else. A larger frame is the honest answer, and the choice between topologies at that point is covered in open frame versus tubular.
The working point is saturated. Past roughly 1.6 to 2.0 T in low-carbon steel, extra ampere-turns produce heat rather than flux, so the window argument stops applying and adding copper stops helping. The force-against-current curve bends, and the bend moves with temperature.
The driving constraint is speed rather than force. A laminar or eddy-current-limited magnetic circuit is then the problem, and a different core construction buys more than any winding change. The dynamic behaviour of inductance is covered in the response-time article linked above.
The force has to be held across a long stroke. Ampere-turns are then not the limiting term, because fringing and leakage grow with the gap and eat the flux before it reaches the work. Adding copper to a long-stroke design is the most expensive way to get a small improvement.
The supply is AC. Then the winding impedance depends on plunger position, the current is not set by resistance, and the whole framework above changes shape. That case is covered in AC versus DC solenoids.
The practical conclusion is worth stating plainly, because it saves a lot of experimentation. Gauge is a driver decision. The window and the current density are the force decisions. If a design is short of force, rewinding it with different wire will not fix it, and the time spent proving that is time that could have gone into the gap, the return spring or the thermal path.
solenoid coil wire gaugesolenoid coil turns calculationsolenoid winding designsolenoid coil resistance calculationcoil fill factor
Frequently asked
- Does a thicker wire give more force?
- Only if it lets you put more power into the same window. Ampere-turns go with wire area, but resistance falls faster, so at equal power the ampere-turns are identical. A 0.20 mm winding at 24 V and a 0.40 mm winding at 6 V both produce about 877 ampere-turns at 3.0 W. The thicker wire gives you the same force through a different voltage and current pair.
- What actually determines the force I can get from a given coil?
- Ampere-turns equal current density multiplied by the copper cross-section, and the copper cross-section is the window area multiplied by the fill factor. So at a given current density the force is set by the window, not by the wire. To get more force you need a larger window, a higher current density through better cooling, or a magnetic circuit that wastes less of what you already have.
- Why is there an optimum fill factor below 100 percent?
- Because round wires cannot tile a plane without voids, and the enamel coating takes up space without carrying current. A random wind typically reaches 0.45 to 0.55, and a layered or orthocyclic wind reaches 0.60 to 0.70. The rest of the window is air and insulation.
- How do I choose the wire gauge then?
- From the supply and the driver, not from the force. Fine wire means more turns, less current and a higher voltage, which suits a low-current drive but wastes a slightly larger share of the window on enamel and is harder to wind without damage. Thick wire means fewer turns and higher current, which needs a bigger switching device and heavier tracks. Both give the same ampere-turns at the same power.
- Does gauge change how fast the solenoid responds?
- Not the time constant. The L/R time constant works out to a function of core geometry, window copper area and mean turn length only, so it is the same for both windings. What changes is the drive voltage you need, because reaching a given ampere-turn level in a given time scales with voltage multiplied by wire area.