Why Does a DL06 High-Speed Counter Reset Before Preset?

The DL06 punch sequence sometimes starts at 99 counts for a preset of 100, even though the encoder is configured for the high-speed counter. Do not add a count to the preset or change counter mode first: separate a display/scan observation from a genuine early trigger, then check the preset math and encoder signal at the point the counter resets.

Stop using scan time or a preset offset as the first fix

Two quick fixes are tempting, but neither proves why the punch triggers early:

• Shorten the PLC scan. At 400 Hz, pulses arrive every 2.5 ms. A scan longer than that can make ordinary ladder logic or a status display miss intermediate count values. The high-speed counter exists to count pulses independently of ordinary scan-by-scan observation; speeding up the scan does not remove an extra electrical edge or correct a wrong preset.
• Add one to the preset. Raising a preset of 100 to 101 can conceal a displayed 99/100 discrepancy, but it changes the commanded length and may create a different error. Do not use it as a production correction until you establish that the counter is truly one pulse short at the physical punch trigger.
• Switch to Mode 20 without checking the encoder. A mode change is relevant only if the encoder is quadrature and the wiring and configuration support that mode. The existing setup is described as Mode 10 with one input. Do not select a different mode just to suppress an intermittent count.

Keep the machine stopped or run a controlled, non-production test if an early punch could create a hazard or scrap. Record the observed raw count, preset, and output state before changing logic.

Separate a scan-time observation from a real early punch

The counter is self-resetting at its preset. If the 100th pulse arrives after the counter instruction executes in a scan, a value read over the programming connection can show the count at the end of that scan; the reset can occur on the next scan. The terminal value may therefore be too brief to observe reliably, even when the counter has counted correctly.

In the supplied ladder, rung 4 executes the UDC CT174 V2014 instruction, rung 5 sets Y17 when CT174 is true, and later logic uses Y17 with the selected punch and run state to energize an output. A status display of CTA174 is not, by itself, proof that the punch command occurred at that displayed count. Observe the physical punch event and the count/terminal state together.

The distinction matters because the reported production symptom is a punch at about 198 mm for a 200 mm target. If the punch really occurs two millimeters early, that is one encoder pulse early at the stated scale; a fleeting count display alone does not explain it. Capture or trend the count, CT174, Y17, and the relevant punch output during a slow test. Use the DL06 diagnostic tools to measure actual scan time rather than relying on the estimate that it is below 2 ms.

Compare the observed symptom with the likely fault path

These are diagnostic branches, not a reason to assume a particular cause. The decisive measurement is whether CT174 reaches its programmed terminal condition and sets Y17 at the same physical length on a repeatable test.

Verify the length calculation and the value loaded into CT174

The program reads desired length from V2000, multiplies it by 5, divides by 10, and writes the result to V2014. With the stated encoder scale of 100 pulses per revolution and 2 mm per pulse, a 200 mm target should produce a 100-count preset:

V2014 = (V2000 × 5) ÷ 10
For V2000 = 200 mm: V2014 = 100 counts

1. At a safe stopped condition, set the intended length at the Optimate 413 and read the actual value in V2000.
2. Check V2014 after the scaling rung executes and before counting begins. Confirm the value equals the intended number of pulses; do not rely only on the HMI display.
3. Turn the encoder by a known number of pulses or revolutions and compare the physical distance with the raw counter value. The stated scale predicts 100 pulses per revolution and 2 mm per pulse.
4. Check the exact arithmetic behavior for the values used in production, especially non-multiples of the scaling increment. Confirm the stored preset rather than assuming how integer division handles every value.

The program also calculates V2002 by doubling CTA174. Treat that as a derived value, not the raw pulse count, when comparing the HMI distance, the counter, and the preset. Ensure the counter is loaded with the intended V2014 at run time; an accurate equation is not useful if the loaded operand is stale or different from the displayed setting.

Check whether the encoder can create an extra edge

The recommendation to consider Mode 20 applies only if the encoder is quadrature. The described setup says “single input,” so first identify the encoder output type and the DL06 input wiring. A single-channel signal cannot provide the direction/phase information expected from a quadrature arrangement. Do not switch modes until the encoder, wiring, and counter configuration match.

1. With motion stopped, inspect the encoder signal and wiring for intermittent transitions, loose connections, or electrical interference. Verify the input changes cleanly once per intended pulse.
2. Repeat the test while slowly turning the shaft through the point where the punch would normally trigger. If the count changes more than once for one intended encoder pulse, resolve the signal or mechanical oscillation before changing the preset.
3. If the encoder is quadrature, consult the DL06 documentation for the supported mode and input arrangement. Test the correctly configured mode under controlled conditions and confirm the count direction and distance scale before returning to production.

A normal scan-time explanation does not rule out an extra input edge. The input signal itself must be checked when the physical length is short by one full encoder increment.

Restore operation with a controlled test, then repair the cause

For a temporary production restore, use only a verified encoder scale and preset, and run a guarded test at the target length before releasing the machine. Do not compensate with a one-count offset unless the machine’s process specification explicitly calls for that target correction and a responsible controls engineer approves it.

For the permanent repair, correct whichever measurement fails: length value transfer, preset arithmetic, counter configuration, or encoder signal quality. Keep the run conditions consistent during comparison, including the encoder direction and the machine’s stopping/punch sequence. Change one item at a time and record the resulting preset, raw terminal count, and physical length; otherwise an apparent improvement can hide a second fault.

Confirm the punch repeats at the commanded length

Verify the result over repeated cycles, not just one slow hand rotation. For a 200 mm target, check that V2014 is 100, that the counter terminal condition and Y17 occur at the intended point, and that the punch measures 200 mm rather than 198 mm. Also test another approved length to catch scaling errors that a single setpoint can hide.

Compare the PLC’s reported distance with a physical measurement. Since V2002 is derived as twice CTA174, validate both the raw counter and the scaled value; do not let a scaled display substitute for verifying the pulse count. Return to production only after the trigger count and measured length agree consistently.

FAQ