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

Force curve analysis and optimization tips

mchen_servo

Been spending a lot of time analyzing force curves lately and wanted to share some optimization techniques that have worked for us.

We just published a deep dive on servo press force curves —how to read them, optimize them, and use them for quality control. But force curves are one of those topics where practical experience diverges wildly from theory.

The biggest lesson I've learned working with force curves: the curve shape matters more than the peak value. Two parts can hit the same peak force but have completely different curve profiles —and only one of them is a good part. A sharp spike followed by a drop usually means the punch broke through too fast (common in blanking). A gradual ramp with a smooth plateau means the material is flowing properly (what you want in drawing).

Load cell placement is another thing that trips people up. I worked with a shop that had beautiful, consistent force curves —until we discovered the load cells were on the crown, not the slide. They were measuring frame deflection, not actual forming force. Moved the sensors to the slide adapter and suddenly saw 15-20% variation they'd been blind to. That variation was causing intermittent quality issues they'd been chasing for months.

Questions for discussion:

What sampling rate are you running for force monitoring? Is 1 kHz enough, or do you need 4 kHz for high-speed blanking?
How do you set your force curve envelope tolerances? Percentage-based or absolute values?
Anyone using force curve data for predictive die maintenance (tracking gradual curve drift)?

Force curves are the most underutilized diagnostic tool on servo presses. Let's share what works.

lisaQ_metrology

Force curve envelope monitoring is our primary quality gate. We set upper and lower force limits at 10 points through the stroke. Any part that falls outside the envelope gets automatically diverted. Our false reject rate is about 0.5% and our escape rate (bad parts that pass) is essentially zero. The setup takes about 2 hours per die but it's worth every minute.

anna_plc

The load cell placement issue Mike described is more common than you'd think. I recommend strain gauge load cells mounted on the slide adapter plate, as close to the die as possible. Crown-mounted sensors measure frame deflection plus forming force —you can't separate the two. Slide-mounted sensors give you the actual force on the part.

mchen_servo

Want to add some practical force curve tricks that I've picked up that aren't in any textbook.

Trick 1 — Use the force curve derivative (slope) for punch wear detection. A sharp punch produces a steep force rise at initial contact. As the punch wears, the cutting edge rounds over and the force rise becomes more gradual. We track the slope of the force curve at the contact point and set an alarm when it drops below 70% of the sharp-punch baseline. This catches punch wear about 5,000 hits before it starts affecting part quality. Way better than running until you see burrs.

Trick 2 — Area under the curve (integral) for draw quality. For drawing operations, the total energy (area under force-displacement curve) correlates strongly with material thinning. We calculate the integral in real-time and trend it. A gradual increase in energy means the material is getting harder to form — could be a material lot change, lube breakdown, or die wear increasing friction. A sudden spike means something went wrong (double blank, mislocated blank, broken pilot pin).

Trick 3 — Compare left-side vs right-side force curves for off-center loading. If you have a 4-channel tonnage monitor, subtract the left pair from the right pair. The difference curve should be near zero for a centered die. If it's consistently offset, your die is loaded off-center and you're wearing the gibs unevenly. We caught a die that was 8mm off-center this way — the operator had shimmed it wrong during setup.

Trick 4 — Force curve at different temperatures. Run the same part at startup (cold die) and after 2 hours (warm die). Overlay the curves. The difference tells you exactly how much thermal expansion is affecting your process. On our UHSS line, the warm-die curve peaks about 4% lower than cold-die because the material is slightly softer at elevated temperature. We use this data to set our tonnage monitor windows wide enough to accommodate the thermal transition without false alarms.

lisaQ_metrology's 10-point envelope is solid. We use 20 points on critical parts but honestly the extra 10 points only caught one additional failure mode in 2 years. 10 points is probably the sweet spot for most applications.