50W600 vs 50W470: Non-Oriented Silicon Steel

50W600 and 50W470 are both 0.50 mm non-oriented electrical steel, and they differ by one grade step on the loss ladder. The intuitive move is to buy the lower loss grade. That is often the wrong call, because in this band the lower loss grade also carries lower induction, more silicon and a harder, more expensive sheet. This page compares the two on guaranteed values, on punching behaviour, and on the annual energy figure that decides the case. It sits within the wider non-oriented electrical steel range.

The Guaranteed Values Side by Side

Both sets of limits come from the GB/T 2521.1 grade table, so they are ceilings a supplier must meet, not typical results:

Property 50W470 50W600 Difference
Nominal thickness 0.50 mm 0.50 mm Same
Conventional density 7.70 kg/dm³ 7.75 kg/dm³ 50W600 denser
P1.5/50 max 4.70 W/kg 6.00 W/kg +1.30 W/kg
P1.5/60 max 5.90 W/kg 7.55 W/kg +1.65 W/kg
J2500, reference 1.54 T 1.57 T +0.03 T
J5000, minimum 1.65 T 1.67 T +0.02 T
J10000, reference 1.74 T 1.76 T +0.02 T
Loss anisotropy ±10% ±10% Same
Minimum bend count 10 10 Same
Minimum stacking factor 0.97 0.97 Same

The headline is the loss gap: 50W600 carries a ceiling 1.30 W/kg higher, about 28% more loss at the same test point. The counterweight is induction, where 50W600 is consistently a step higher across all three polarisation levels.

Why the Higher Loss Grade Has the Higher Induction

The reason is silicon content. Density is the tell: 7.70 kg/dm³ against 7.75 kg/dm³. The 50W600 sheet carries less silicon, which is exactly what allows the higher flux density and what keeps the material softer and cheaper to produce.

The same axis drives the mechanical behaviour, and the code letters follow the same logic as the American M-series notation covered in how CRNGO grade codes read. Lower silicon means lower tensile strength and higher elongation, which is why the grades punch differently even though both are 0.50 mm and both carry the same stacking factor. The GB annex lists 380 MPa and 16% elongation for 50W470 against 300 MPa and 22% for 50W800; 50W600 falls between those two on both counts, and the exact values for a given lot come from the mill data sheet.

So the grades are not ranked on a single quality scale. One is cooler running; the other is more forgiving to form and carries flux better at high excitation. Which matters more depends on the machine.

Where 50W600 Wins

Three situations favour the higher loss grade:

  1. High working flux.In a design running close to saturation, the extra 0.02 to 0.03 T at each polarisation level buys more than the loss ceiling costs. Starting and overload conditions are where this shows up, since the high-excitation behaviour of the core decides the inrush current and the starting torque margin.
  2. Tight tooling and high punching volume.The softer sheet wears the die less and produces a cleaner sheared edge. Over a long production run, the tooling and scrapping difference can outweigh the energy figure.
  3. Cost-sensitive builds with loss headroom.Where the efficiency class is not binding, 50W600 is the cheaper material at the same gauge, and the design has room to absorb the higher no-load loss.

Where 50W470 Wins

50W470 earns its place when the core runs continuously and the loss figure is rated:

  • Efficiency-class-driven designs, where the no-load loss ceiling is the constraint rather than the flux level.
  • Continuously energised cores, such as transformers and reactors rather than intermittently run motors.
  • Long duty cycles, where the annual energy gap accumulates into a real figure.

The trade can be put in numbers. A 5 kg core at the two ceilings differs by 5 × 1.30 = 6.5 W of no-load loss. At 6,000 energised hours a year that is about 39 kWh per unit; across a thousand units it becomes a procurement-level argument, not a rounding error. On a short-duty machine the same gap is negligible.

50W600 and 50W470 non-oriented silicon steel coils compared

Punching, Annealing and Stacking

Both grades are 0.50 mm with a 0.97 minimum stacking factor, so the stacking side is common ground; the gauge itself is covered in 0.35 against 0.50 mm non-oriented steel. What differs:

  • Burr sensitivity.The harder 50W470 sheet holds a sharper sheared edge longer but loads the punch more; a dull die shows up as burr and interlaminar shorting, which erodes the loss advantage.
  • Strain recovery.Punching raises local loss near the cut edge on both grades. Stress-relief annealing recovers more of it on the higher-silicon grade, so a loss-critical 50W470 design usually specifies annealing after punching where a 50W600 design might skip it.
  • Edge quality at the same die life.Between the two, the 50W600 route usually runs longer between regrinds for a given burr limit.

Reading the Choice as a Decision

A practical order of questions:

Question If yes If no
Is the design flux-limited rather than loss-limited? 50W600 next question
Is the no-load loss a rated figure over long duty? 50W470 next question
Is punching volume high and tooling life critical? 50W600 next question
Is first cost the binding constraint with loss headroom? 50W600 50W470

If the answers conflict, the loss figure is usually the one that gets audited, because it appears on the efficiency label while tooling cost does not. Comparing the four M-series bands puts this choice in the context of the wider grade ladder.

FAQ

Is 50W470 always better than 50W600? No. 50W470 has the lower loss ceiling, but 50W600 has higher induction, lower density, lower tensile strength and a lower price. A high-flux or heavily punched design can do better on 50W600.

Are the loss figures directly comparable? Yes, both are quoted at 1.5 T, 50 Hz under the same standard. They are not comparable with oriented steel grades, which are measured at 1.7 T.

Does the density difference matter? It matters for weight and for silicon content. 7.70 against 7.75 kg/dm³ tells you 50W600 carries less silicon, which explains both its higher induction and its softer mechanical behaviour.

Can the two be mixed in one core? Not advisable. Mixing grades in a stack puts different loss and induction behaviour in the same magnetic circuit and makes the measured core loss hard to attribute.

What if the certificate shows values between the two bands? Then the lot is being sold against its actual measurements, not a specific grade. Match the offer to the design requirement rather than insisting on the code.

Deciding between 50W600 and 50W470 for a specific frame? Send the working flux, duty cycle and efficiency target to vera.liang@ztwysteel.com, and we will compare the two on the values that matter for your design. The silicon electrical steel sheet range lists the stocked 0.50 mm material, and choosing non-oriented steel for motors covers selection above the grade level.