Why Doesn't the Extraction Arm 'Pick Up' Anything? Capture Velocity and the Invisible Limits of an Extraction Arm

A suction nozzle does not work like a magnet, and this single sentence explains half of the efficiency problems encountered in the field.

When you walk into a welding workshop, you’ll often see a familiar sight.

The extraction arms hang up high, left in standby mode. Meanwhile, the operator carries on welding, whilst smoke rises, passes a metre away from the arm and drifts up towards the ceiling. When asked, the answer is always the same: ‘The extraction’s on; the machine’s working, isn’t it?’

The unit is indeed working. The fan is turning, the filter is in place, the indicator is green. But at that moment, the system isn’t capturing anything.

The reason is neither as simple as the unit’s power nor the operator’s negligence. The cause is a physical fact that goes against most people’s intuition: extraction is not the mirror image of airflow.

Airflow travels far, but suction is not generated from a distance

When you stand in front of a fan, you feel the breeze metres away; the expelled air travels in a steady jet. With suction, however, the situation is exactly the opposite: the suction inlet draws in air from all directions (from the front, the sides, and partly from the rear) simultaneously. The air comes not from a single corridor, but from the entire spherical volume surrounding the inlet. For this reason, the suction velocity drops dramatically as you move away from the inlet.

As a rough estimate, at a distance equal to the inlet diameter, the velocity falls to less than one-tenth of the value at the inlet.

A 160-millimetre intake orifice with a velocity of 10 metres per second at the orifice will register a velocity of less than 1 metre per second just 16 centimetres away. At half a metre, the effect drops to a level indistinguishable from the natural air movements in the room. The intake orifice is not a magnet; it cannot ‘draw in’ smoke from a distance. The smoke must pass through that small area where the extraction is effective.

Capture velocity

So, what velocity is required in that area to capture the smoke?

This is known as the capture velocity, and the general consensus for welding fumes is approximately 0.3 to 0.5 metres per second at the point where the pollutant is generated. This velocity must compete with two factors.

1. Cross-drafts in the workshop. An open door, an axial fan, or even a moving forklift can generate air movement of this magnitude.

2. And more importantly, the smoke’s own thermal rise. The column of hot fumes above the welding arc accelerates upwards of its own accord, and if the arm is in the wrong position, this column will pass through the extraction zone without ever being captured.

A practical conclusion follows from this: the hood of the extraction arm must be positioned where the fume column is heading – that is, slightly above and in front of the welding point, typically at a distance of 30–40 centimetres.

The cost of incorrect positioning is severe. As the extraction’s effectiveness diminishes with the square of the distance, an operator who places the hood twice as far away will require roughly four times the airflow to achieve the same capture efficiency. No fan can compensate for this waste indefinitely. That is why there is a saying in the industry, half-joking, half-serious: the cheapest way to increase capacity is to move the arm even closer to the welding area.

Capture velocity

Where does the 650 m³/h threshold come from?

In terms of standards and regulations, a minimum airflow threshold of around 650 m³/h per arm is widely accepted for individual extraction arms. This figure is not arbitrary; it is the reverse of the calculation outlined above.

Given a typical hood geometry and a realistic working distance (the 30–40 centimetres we mentioned), the flow rate required to ensure capture efficiency against cross-drafts and thermal rise corresponds roughly to this level. Of course, ‘some’ air is extracted even at lower flow rates; however, capture that safeguards the operator’s breathing zone is only consistently achieved above this threshold.

Those who have read our previous article will undoubtedly have made the connection. This figure of 650 is not the unit’s catalogue flow rate, but the actual flow rate at the inlet under conditions of a dirty filter and a fully extended arm. A unit listed in the catalogue as 1,000 m³/h may fall below this threshold when system resistance is taken into account; this is precisely the practical significance of ISO 21904’s requirement to issue a warning in the event of a drop in flow rate.

Three practical habits that work on site

All this physics can be boiled down to three behaviours.

Reposition the fume extractor at every new welding position; the arm is not a piece of equipment that is set up once and forgotten, but a tool that is an integral part of the job, just like a welding clamp.

Position the fume extractor along the path the fumes will rise, not directly above the welding point, as the thermal column rises at an angle.

And use a flanged fume extractor. A simple flange fitted around the opening cuts off the ineffective air intake from behind, noticeably extending the effective extraction distance; the cost is virtually zero, whilst the benefit is measurable.