What is a servo press?

A servo press uses servo motors for precise force and position control during forming operations.

How do I calculate servo press tonnage?

Use our free tonnage calculator at servopress.club/tonnage-calculator.html

Motion profile for HSS seat frame stamping

greg_engineer

Developing a motion profile for HSS seat frame stamping. The standard profile isn't cutting it — getting springback and cracking. What approach works?

**Motion Profile
**
- Approach: 400mm/s
- Pre-contact: 80mm/s at 20mm above material
- Forming: 50-100mm/s
- Dwell at BDC: 50-100ms (reduces springback variation 80%)
- Return: 500mm/s

**Blank Holder Force
**
- HSS 590MPa: 15-20% of forming force
- HSS 780MPa: 20-25%
- UHSS 980MPa: 25-30%

Increase if wrinkling. Decrease if splitting at punch radius.

**Lubrication
**
EP draw compound at 2-4 g/m2 both sides. Never dry stamp UHSS.

**Tonnage Limits**

- High: 110% nominal (die crash, double blank)
- Low: 85% nominal (missing blank, broken punch)

yuki_rd

The 50-100ms dwell recommendation is good for 590-780 MPa. For 980 MPa and above, we've found that 100-150ms gives better results. The extra dwell time allows stress relaxation in the material, reducing springback by an additional 10-15%. The cycle time impact is minimal —50ms extra on a 2-second cycle is only 2.5%.

pressroom_tom

I run these profiles daily. One thing to watch: the forming speed needs to be consistent across the entire forming zone, not just at the contact point. If the servo accelerates through the forming zone (which happens with some motion profiles), you get inconsistent springback. Use the constant-velocity segment, not the deceleration curve.

toolndye_dave

On dwell time and constant velocity mode — one thing I'd add from our testing - when you're running 980+ MPa UHSS for seat frames, monitor your servo motor torque curve closely during the forming zone. We saw cases where the torque spike at initial contact was 2-3x the steady-state forming load, and that transient was causing micro-cracks in the part edge. The fix was adding a 15mm pre-forming segment at 30mm/s before the main forming speed kicks in. Basically a soft-touch entry. Reduced edge cracking from ~4% to under 0.5% on DP980 seat side members. Also worth noting: if you're using a link-drive servo press (like the Komatsu H2F series), the velocity profile near BDC is inherently different from a direct-drive crank. The link mechanism naturally slows down near BDC, so your programmed dwell might be effectively longer than you think. We measured actual slide velocity with a laser displacement sensor and found the real dwell was 180ms when we programmed 100ms. Not a problem per se, but it explains why some shops get different springback results with the same nominal profile on different press types.

jim_apps

Coming back to this with an update — we've been running DP980 seat side frames for about 8 months now and have refined the profile significantly from where we started.

The biggest lesson: the "right" forming speed isn't a single number, it depends on where you are in the draw. We split our forming zone into three segments:

Segment 1 — Initial contact to 25% draw depth: 25mm/s. This is where the blank is being pulled into the die cavity and the flange is under maximum tension. Going faster here causes the material to neck at the draw radius. We lost a lot of parts to this before we slowed down the entry.

Segment 2 — 25% to 75% draw depth: 45mm/s. The material has work-hardened from the initial draw and can handle more speed. The flange area has shrunk so there's less friction. This is where you recover cycle time.

Segment 3 — 75% to BDC: 30mm/s, then 120ms dwell. Slow down again because this is where the final geometry is set. Any speed variation here shows up directly as springback variation. The dwell is critical — we tested 50ms, 80ms, 100ms, and 120ms. Springback standard deviation dropped with each increase up to 120ms, then plateaued. So 120ms is our sweet spot for DP980 at 1.4mm thickness.

Return speed: we actually backed off from 500mm/s to 350mm/s. At 500mm/s the stripper was leaving marks on the part surface. The seat frame is a visible interior component on this particular vehicle, so surface quality matters. 350mm/s eliminated the marks with only 0.15 seconds added to the cycle.

pressroom_tom's point about constant velocity is critical. On our Komatsu controller, we had to explicitly set the forming segments as "constant velocity" mode rather than "position interpolation" mode. In position mode, the controller overshoots the target speed at segment transitions and you get velocity spikes. Constant velocity mode limits acceleration and keeps the speed smooth through the forming zone.

One thing I haven't solved yet: the first 5-8 parts after a die change always have slightly different springback than steady-state. Die temperature hasn't stabilized yet. We're considering adding die preheaters but haven't pulled the trigger.