How do I get Tecweigh WP50 faults into the PLC?

This guide is the Tecweigh layer on top of getting belt weigher data into your PLC or SCADA. For what each message means, see what the messages on a Tecweigh WP50 mean. We’re an independent service company, not a Tecweigh dealer or agent. For bookings and parts, see Tecweigh belt weigher and WP-series servicing.

What does the plain serial port tell the control system?

That an alarm exists, not which one. The WP50’s RS485 auxiliary port can broadcast to a remote display, or be polled by a PLC, with up to 16 units on one multi drop line. Status comes as a single ASCII character. It separates a positive and negative rate, a zero, tracking or span in progress, a load error and a general alarm. That general alarm covers more than 30 conditions, so a lost speed signal and a high total look the same to the PLC.

There are two data formats. The legacy format lets a WP50 share a line with older WP20 and WP25 units and use their layout; the WP50 format adds decimal points and comma separated fields. A WP50 replacing an older unit on an existing line needs the legacy format, or the PLC gets numbers that look almost right.

What changes when the comms card is fitted?

Every alert gets its own flag. The CM100 runs Modbus TCP (the default), EtherNet/IP or PROFINET, fits either socket, and takes a DHCP or static address. Its fault value matches the front panel alert list one for one, from bad scale card comms and zero belt speed to the calibration results and lock alerts, and the same flags appear on all three protocols. So the control room can be told the speed signal has gone, not just that something is wrong.

Two catches. Not every flag applies in every mode, so one that never goes true may belong to batching or feed control. And one flag is batch complete, a success message: code that alarms on any flag in the value will alarm on a good batch. Read the flags that matter.

Can the PLC tell a calibration from a real stop?

Yes, and almost nobody wires it up. A separate scale status value flags a load error, an alarm present, an auto zero, auto span or tracking in progress, and a negative rate, positive rate or rate in deadband.

Without it, a calibration looks exactly like a stoppage: the rate sits at zero, the low flow alarms go off and the historian logs a dropout that never happened. With it, the plant can hold those alarms during a calibration. It’s free to set up at design time and awkward to add later. See what happens during a calibration visit. Give rate in deadband its own alarm text too: nothing is totalising, but the belt hasn’t stopped and the scale hasn’t failed. See why a belt weigher is wrong at low flow.

Why trend the scale comm errors count?

It’s one of the few predictive numbers a belt weigher gives you. The maintenance menu shows the errors in the data from the scale over the last minute, and it’s published on all three protocols. Sample it on a schedule and watch the baseline. In our experience a failing link shows a rising count weeks before it drops out: damp at a gland, a chafing cable, a radio path gone marginal. Where it ends is bad scale card comms.

What if there’s no comms card and no spare points?

Work on or near a conveyor, or inside an electrical enclosure, needs isolation under your site’s procedures and trained, authorised people. If that’s not you, stop and call us on 1300 101 666.

Set any WP50 relay to the error function. It operates on a scale data error, a load error or low power, the three faults that stop the scale weighing. It doesn’t say what’s wrong; it says don’t trust these tonnes. The relay list calls it scale data error even though the alert list says bad scale card comms.

Three outputs are dry contact form C; the fourth is a 0-24 V DC pulse output. Wire the fault contact so healthy is a closed circuit, so a broken wire or lost supply reads as a fault. For the rate and speed alarm relays, set realistic limits and use the delay: an alarm that’s always on gets ignored within a week.

Why does our PROFINET controller never raise an alarm from the weigher?

The PROFINET device description file we hold declares process alarms disabled, so the weigher doesn’t raise its faults as device alarms. It only reports link state on its network ports. The PLC program has to read the fault flags itself, or the controller shows a healthy device while a fault stands on the panel. Use the description file that matches your card’s firmware.

What are the lights on the comms card telling you?

• All four dark at boot: an invalid board configuration. Power down and reseat the card. It clicks when it’s fully home.
• All four red: the hardware isn’t recognised. That’s a replacement, not a setting.
• Duplicate address (EtherNet/IP): usually a spare card on a default address. Talk to whoever holds the site address list.
• Connection timeout (EtherNet/IP): usually the controller end, not the weigher.

What needs planning before a protocol change or firmware update?

Neither belongs in the last hour of a shutdown. A protocol change takes about two minutes to install. Don’t cycle the power or touch the keypad while it runs, or the flash memory can be corrupted.

Firmware loads from a FAT32 USB stick with one file in the root. The comms card shows nothing while it updates, so wait at least 15 minutes before disconnecting, then check the trace log written to the stick. If a connection keeps dropping, check the firmware revisions on the processor and card, and the mapping revision the PLC code was written to, before anyone swaps network switches.

What should be settled before anyone quotes the integration?

First, what the control room needs: one contact that says don’t trust the tonnes, a few values over serial, or every fault by name over the bus. Each is a different job. Then:

• Protocol: set by the site’s control system, not the weigher.
• Code: who will read the status and fault values, and what plant alarms do during a calibration.
• Read only or write: serial and fieldbus can both start a zero or span and reset totals. Don’t send commands while someone is at the unit. A remote zero or span is refused if it would move the reference by more than about 5%, and the keypad locks don’t cover remote paths.
• Records: what each output carries, the addressing, the mapping revision and the flags in use. We keep that with the scale’s history in AIMM, our asset management system.

If the tonnes are wrong rather than missing, start with common belt weigher faults: a perfect integration reports a bad number faithfully. For the older processors, see Tecweigh WP20 and WP25 fault codes and calibration.

We supply parts for Tecweigh and the other major belt weigher brands: a genuine part where it’s available, or a suitable alternative, with compatibility confirmed and the part identified on the quote.

Need a manual for your equipment?

This job can need the processor manual, the comms module supplement, the mapping file for your protocol and, for PROFINET, the device description file. Each is revised separately, so match the mapping revision to the firmware on the processor and the card before you decode a single value.

Talk to us about manuals for your equipment. Send the make, model, serial number and software version (a photo of the label is fine) and we’ll help you find the right document.

When should you call us?

Call before the integration is scoped, so the status and error flags are in the design from the start. Call when the PLC shows faults that don’t match the panel, a connection keeps dropping, the card comes up all red, or before a protocol change or firmware update in a shutdown. Call 1300 101 666 or email admin@accurateindustries.com.au with the processor and card details and a photo of the indicators. See booking a belt weigher service.

Frequently asked questions