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

ABB ACS880 fault 7121 motor stall - parameter fix

toolndye_dave

ABB ACS880 fault 7121 motor stall on our servo press. Turns out it was a parameter issue, not a mechanical problem. Here's the fix.

Understanding the Fault Logic:
The ACS880 triggers 7121 when motor speed is below the stall speed threshold (31.24) while torque is above the stall torque threshold (31.25) for longer than the stall time (31.26). Default settings are too sensitive for press applications.

Common Causes:
1. Parameters too tight: At BDC, a servo press motor legitimately produces high torque at near-zero speed. Adjust 31.24 to 5% of rated speed and 31.25 to 120% of rated torque.
2. Mechanical binding: Check if the fault occurs at a consistent slide position.
3. Encoder feedback loss: If the drive loses encoder feedback, it sees zero speed with high torque demand - identical to a stall.
4. Undersized motor: If the press was recently retooled with heavier material, the motor may be genuinely overloaded.

Parameter Adjustments for Press Applications:
- 31.24 (Stall speed): reduce to 2% of rated speed
- 31.25 (Stall torque): increase to 150% of rated torque
- 31.26 (Stall time): increase to 60s

anna_plc

Fault 7121 on ACS880 in press applications: check parameter 30.01 (stall speed limit). The default is often too high for press forming where the slide moves very slowly under load. Reduce it to 2-3% of rated speed. Also verify that the motor model parameters in group 99 match the actual motor nameplate. Wrong motor data causes the stall detection to trigger falsely.

toolndye_dave

The stall speed threshold in the ACS880 is parameter 30.01. Default is usually 10% of rated speed. For press applications where the slide legitimately moves very slowly during forming (especially with dwell), you may need to reduce this to 2-3%. Otherwise the drive thinks the motor is stalled when it's actually just moving slowly under high load.

jim_apps

One more thing to check with 7121 on press applications ??the load curve during the forming stroke. If you have a sharp load spike at bottom dead center (common with coining or embossing features), the drive can interpret the sudden torque increase as a stall condition even if the motor is still moving.

We fixed this by adjusting parameter 31.26 (stall time) from the default 20s down to 5s, and also enabling the load curve function (group 37 on ACS880). The load curve lets you define expected torque vs speed profile so the drive knows high torque at low speed is normal during forming. Took about an hour to tune but eliminated all false 7121 faults.

mike_blanking

Had another 7121 situation last month that was a bit different — turned out to be the speed feedback filter. Parameter 90.41 (speed feedback filter time) was set to 50ms from a previous tuning session. At low forming speeds that much filtering was masking the actual speed signal, so the drive thought the motor was at zero speed when it was actually crawling at 2 rpm.

Dropped it to 8ms and the fault went away. The downside is you get a bit more noise in the speed signal but for press work at low speed that matters way less than false stall trips.

If anyone is chasing intermittent 7121s that only show up during the forming portion of the stroke, check your feedback filter settings before you start messing with the stall thresholds. Saved us a lot of headache.

carlos_mx

The speed feedback filter point mike_blanking raised is huge and I want to expand on it. Parameter 90.41 is one of those "set it and forget it" values that can bite you months later.

Here's the scenario I've seen twice now: someone tunes the drive for a high-speed blanking job (40+ SPM), sets the speed filter higher to smooth out the feedback noise at high velocity. Works great. Then the die changes to a slow forming job at 8 SPM with a 200ms dwell. Now the filter is masking the low-speed signal and you get phantom 7121s.

My rule: if you change 90.41 for a specific job, write the original value on tape on the cabinet door. Better yet, save it in the parameter set (group 61 on ACS880 lets you store 4 complete parameter sets). We use Set 1 for blanking, Set 2 for forming, Set 3 for deep draw. Operator selects via DI — no manual parameter changes needed.

Also — if you're getting 7121 only on the first 5-10 strokes after a cold start, check the motor thermistor reading (parameter 09.11). Cold motor winding resistance is higher, which throws off the current model. The ACS880 motor model adapts but it takes a few cycles. Extending 31.26 (stall time) by 2-3 seconds for the warm-up period is a reasonable workaround.

mchen_servo

We chased a 7121 on an ACS880-equipped 400T for almost a week before finding the root cause, and it was none of the usual suspects.

Turned out the motor thermistor wiring had a marginal connection — not bad enough to throw a temperature fault, but the intermittent resistance spike was confusing the drive's motor model. The drive thought the motor was stalling when it was actually just getting bad feedback data. Checked the thermistor connector, found one pin with a hairline crack in the solder. Reflowed it and the 7121 never came back.

Point being: 7121 says "motor stall" but the ACS880 infers stall from multiple inputs — speed feedback, current, and motor model. If any of those inputs are slightly wrong, the stall detection logic can false-trigger. Before you spend days tuning parameters, verify your feedback signals are clean. Scope the encoder output and check the thermistor circuit resistance. A $2 solder joint cost us 4 days of downtime.