In high-frequency induction welding, the impeder decides how much of the generator’s power ends up at the weld. For decades it has been a ferrite core. Fluxtrol’s soft magnetic composite (SMC) impeder cores replace that ferrite with a material that does not saturate as early and does not fracture under thermal shock, and the published evidence now covers trials, simulation and an off-line validation method.
What an impeder does
In HFI welding, a steel strip is roll-formed into an open-seam tube, an inductor heats the seam edges and rolls squeeze them together. The coil induces current around the tube OD. When it reaches the open seam it has three paths: along the edges to the weld apex (most desirable), back along the edges toward the incoming strip (somewhat useful), or along the tube ID (undesirable leakage).
Leakage on the ID raises the current needed on the outside, which raises coil current and the reactive power (kVAR) the capacitors must supply, which increases losses in the matching components. The impeder is a magnetic core inside the tube that increases impedance on the ID path, so more current stays on the edges.
Why ferrite becomes the limit
Ferrites are the standard because they are cheap, highly permeable in weak fields and have low losses at high frequency. But their saturation flux density is low, about 0.4 to 0.5 T in welding conditions, and they are sensitive to thermal shock and temperature. In small tubes the core cross-section is small, so flux density is high, the ferrite saturates, and current flows more freely around the ID. The ASM Handbook reports ferrite life for small tubes of several hours to a week. Fluxtrol identifies the impeder as the limiting factor on many small-tube lines today.
Why an SMC impeder
SMCs are insulated iron-based particles in a polymer binder. Properly designed, an SMC core can carry more than double the 0.4 T saturation flux density of typical ferrite, and its magnetic properties are more stable over temperature and time. The result is more effective ID suppression at high loading. SMCs are also machinable, mechanically robust and suited to many geometries.
The comparison in Fluxtrol’s presentations shows the crossover clearly: at low fields, typical ferrite outperforms Fluxtrol A, though Fluxtrol A is still about 100 times better than air. At 0.7 T and above, ferrite is only about 2 times better than air, while Fluxtrol A stays about 100 times better.
Performance evidence
| Trial | Setup | Reported result |
|---|---|---|
| Prinz & Co. GmbH Stahlrohre | 15 mm OD, 1.2 mm wall, about 400 kHz | About 40% energy savings at the same speed; over 20 h life vs about 4 h typical for ferrite |
| Fahop | 21 to 27 mm OD, 2.6 mm wall, Fluxtrol A | 30 to 50% energy savings; about 5 times longer life |
| Kent Corporation | 12.7 mm OD, 1.25 mm wall, about 400 kHz, 300 kW | About 20% energy savings; ran over a full shift without power adjustment |
| Through-flow trial (Fluxtrol 75) | 19 mm OD, 3 mm wall | Over 30% lower current, nearly 40% lower power |
| Return-flow impeder | Smaller SMC core | Almost 20% savings with a core 3 mm smaller in diameter |
The Prinz tests in Germany were funded by the German government, with the impeder system made by TWTools and research partners from Chemnitz University of Technology and the University of Hanover. Fluxtrol had no active participation in the Fahop analysis, which used Serbian costs. Energy-cost and CO2 figures in the sources rest on stated assumptions of about 5,000 production hours per year, $0.08/kWh and 6.99 x 10^-4 tons CO2 per kWh. At Prinz that gives 25 kW saved, 87 tons of CO2 and about $10,000 a year. Savings from longer life and speed are additional. An external SMC magnetic bridge above the tube opening is a related option, reported to save 20 to 30% energy in some applications.
Where SMC impeders fit best
- Small diameter, heavy-walled steel tubes
- Small-tube lines where the welder is the bottleneck and the impeder is in saturation
- Lines where impeders fail from mechanical impact
- Impeders that contain metallic components
- Applications with very short impeder life
Costs and trade-offs
SMC cores cost 10 to 50 times more per unit than ferrite, have lower permeability at low fields and higher losses, which means better cooling and may rule out gas-cooled systems. Fluxtrol’s presentations state that annual impeder costs are often similar because lives are longer, while production savings tend to be much higher. Mill trials are expensive, and ferrites remain proven technology, which is why validation matters.
Off-line validation: test before the mill
Fluxtrol’s workflow validates impeders without stopping a production line, in three steps: 2D and 3D models for cross-section and cooling, a laboratory stand for physical simulation, and field testing with partners once the design is proven. A 2026 peer-reviewed study in Applied Sciences, with Leibniz University Hannover and Chemnitz University of Technology, formalizes the method: a 3D electromagnetic-thermal model of the welding process was checked against measurement, then reduced to a 2D model by removing the tube and applying a correction factor. For a representative Fluxtrol 50 impeder, it showed a hotspot under the inductor, the critical thermal limit.
On the lab stand, the impeder sits in a coil that mimics tube-welding flux, with coil voltage and current, water flow and inlet and outlet temperature measured for calorimetric loss. A 10 mm hollow core tested at 200 kHz and about 1.2 T showed power-loss deviation below 5% against simulation. In the 2026 study, simulated and measured outlet water temperature differed by about 12% in one comparison. Read the paper , the Fluxtrol announcement , or earlier work from IMAT 21 and HES 23 .
Cooling, geometry and sustainability
Cooling. Cooling needs both total heat removal (set by coolant flow) and local heat removal (set by water velocity, impeder shape and thermal conductivity). Fluted designs were tuned to keep the internal temperature below 300 degrees C at 0.8 T and 300 kHz, and CFD models water flow in return-flow cores, whose closed loop adds flow resistance. Modeled loss on one fluted core was 20 to 30% below measured. Details are in Cooling Requirements for SMC Impeders .
Geometry. Flux density varies along the core, with the high-loading zone under the coil winding, so cross-section and flutes can be optimized for magnetic loading and cooling together. See IMAT 23 on impeder geometry optimization .
Energy and emissions. Lower power at the same speed means lower kWh per ton. In the through-flow trial, over 100 kW less power for 20% higher line speed meant 45% lower kWh per ton. See Improved Sustainability of Induction Tube Welding Systems .
Learn more
Read the full presentation , the tube welding brochure , and the process overview at Tube & Seam Welding . To assess your line, contact Fluxtrol with tube size, wall thickness, frequency, power and impeder life, or request pricing through the SMC order form .
SMC grades used in impeder work
Real-world results
Prinz & Co. GmbH Stahlrohre (Germany)
15 mm OD steel tube, 1.2 mm wall, about 400 kHz, 50 to 150 kW. SMC impeders compared with ferrites at the same line speed.
- About 40% energy savings
- Over 20 hours of life with little wear, versus about 4 hours typical for ferrites
- 25 kW saved, equal to 87 tons CO2 and about $10,000 per year on stated assumptions
Fahop (Serbia)
21 to 27 mm OD tube, 2.6 mm wall, 13 mm impeders, about 400 to 500 kHz, 250 kW. Independent analysis at line speeds from 10 to 60 m/min.
- 30 to 50% energy savings with Fluxtrol A
- Impeders reported to last about 5 times longer
- Fluxtrol had no active participation in the study
Through-flow impeder trial
19 mm OD, 3 mm wall tube; ferrite replaced with Fluxtrol 75, with only the impeder changed.
- Over 30% lower current, almost 40% lower power at the same speed
- Ferrite heavily saturated at 0.4 to 0.5 T; the SMC stayed below its 1.4 T saturation