Silicone Hose: The Strongest Braid Sits at 54.7 Degrees

The braid is the pressure vessel, not the rubber

A silicone hose without reinforcement is a tube. It carries pressure until the wall stress exceeds the tensile strength of the rubber, which for silicone at 100 C is not much. The braid is what makes it a hose, because the fibres take the hoop and axial loads and the silicone only has to be a seal.

The fibres carry load in tension, and the tension depends on the angle at which they cross the axis. At one particular angle the geometry balances the hoop and axial loads so that the fibre is loaded purely in tension with no tendency to rotate or straighten. That angle is the arctangent of the square root of two, 54.7 degrees, and it is the same angle that appears in pressure vessels, in biological tissue and in the Clapeyron net.

Measured burst pressure against braid angle: 12 bar at 30 degrees, 17 at 40, 21 at 50, 22 at 55, 17 at 65 and 11 at 80. The curve is symmetric around the optimum in theory and slightly skewed in practice, because the braid also has to be manufacturable at the ends.

What going off the optimum actually does

A braid at 30 degrees is too close to the axis. Under pressure the hose tries to lengthen and the braid angle increases towards the optimum, so the hose grows in length and shrinks in diameter before it fails. That movement is visible as a hose that gets longer when it is pressurised, and it is a common complaint on low-angle braids.

A braid at 80 degrees is too close to the hoop. Under pressure the hose tries to shorten and swell, the angle decreases towards the optimum, and the hose grows in diameter. It fails by bursting along the fibre direction rather than across it, which produces a long split rather than a hole.

Both failures are progressive rather than instant. The hose moves towards the optimum angle as it is pressurised, so a braid that is off-angle does not fail at the rated pressure on the first cycle. It creeps towards the optimum over hundreds of cycles and fails later, which makes the defect look like a material fatigue problem.

Carriers set the strength, angle sets the efficiency

Once the angle is right, burst pressure scales with the amount of fibre. Measured against carrier count: 14 bar at 24 carriers, 20 at 36, 26 at 48 and 31 at 64. The relationship is close to linear because each carrier adds the same amount of fibre at the same angle.

That gives a designer two independent knobs, which is the useful part. Carrier count sets the strength at a fixed diameter, and angle sets how efficiently the fibre is used. A hose that needs 30 bar can be built with 64 carriers at the optimum angle, or with more carriers at a worse angle, and the first option is lighter and cheaper.

Table 1 — Braid angle against burst pressure, and the movement that gives it away

The mandrel decides the bore, and the bore decides the braid

A silicone hose is built on a mandrel, and the mandrel is what keeps the bore open while the rubber is applied, the braid is laid over it and the whole assembly is vulcanised. The mandrel diameter sets the finished bore, and the braid is laid over a diameter that is larger than the finished bore because the rubber underneath has thickness.

That means the braid angle is set on a diameter that the customer never sees, and the angle that results on the finished hose is slightly different from the angle the machine was set to. A hose with a thick inner liner and a thin cover has its braid at a larger diameter relative to the bore than a thin-lined hose, and its effective angle is different.

The mandrel also sets the length. Silicone shrinks during vulcanisation, and the shrinkage is restrained by the mandrel in the bore but not in the length. A hose pulled off a straight mandrel comes out slightly shorter than it went on, and a hose made on a curved mandrel has to account for the curve in its cut length.

Table 2 — Hose failures read back to the braid or the mandrel

How to test a hose you did not design

Burst testing is destructive and slow, so most hose is qualified by a sample and then trusted. The trust is misplaced if the braid angle drifts, and it drifts whenever the braider tension changes, which happens with wear on the carriers and with a change of yarn lot.

Two measurements catch most of it without a burst test. The first is the change in length under pressure: a hose at the optimum angle does not change length, and a hose that grows by 2 percent at rated pressure is a hose whose angle is wrong. The second is the braid angle itself, measured on a cut section with a protractor, which takes a minute and is almost never done.

A pressure-hold test at 1.5 times the rated pressure for one minute catches the gross failures. It does not catch the off-angle braid, because an off-angle hose holds pressure perfectly well until it has crept. That is why the length measurement matters more than the pressure test for this particular defect.

What to specify

A hose specification that can be enforced names the braid, not only the performance.

• Burst pressure and the test temperature, since silicone loses strength quickly above 120 C.
• Braid angle, with a tolerance of a few degrees, measured on a cut section.
• Carrier count and yarn specification, because burst pressure follows both.
• The change in length at rated pressure, which is the cheapest check on the angle.
• Mandrel diameter and the finished bore, since the two differ by the liner thickness.

Add the bend radius and the working temperature, because a hose that passes a straight burst test can still collapse on a tight bend at 130 C. The burst test measures the braid; the application measures the wall.