IC693ALG223 Group 2 Zero-Reading: MUX or Configuration?

IC693ALG223 Group 2 Zero-Reading: MUX Failure or Configuration Oversight?

A field report examines a GE Fanuc Series 90-30 system where channels 1–8 of an IC693ALG223 module read normally while channels 9–16 return zero counts. This article evaluates whether the internal group-isolation multiplexer is the true culprit or whether configuration and power-supply factors better explain the symptom. Practical diagnostic data and verification steps are provided for maintenance engineers facing identical failures.

Tags: #IC693ALG223 #GE-Fanuc-90-30 #Analog-Input-Fault #PLC-Troubleshooting #Industrial-Automation

Symptom Profile and Initial Field Observations

The IC693ALG223 provides sixteen single-ended current input channels on the Series 90-30 platform. A recent case involved a module where the first eight channels tracked field transmitters correctly. The second group, spanning channels 9 through 16, consistently returned raw %AI counts of zero. Technicians confirmed live 4–20 mA loop currents at the terminal block using a calibrated Fluke 787 process meter. The discrepancy between physical current and digital representation narrowed the fault domain considerably. A module-level hardware defect was initially suspected, specifically a failure within the group isolation multiplexer. However, the ALG223 architecture does not publish a discrete group-level MUX as a field-replaceable element. This fact alone demands that configuration and power integrity be ruled out first.

Architectural Reality of the ALG223 Input Path

The ALG223 scans active channels in serial order through a shared analog-to-digital converter. No documented internal isolation barrier separates channels 1–8 from 9–16 at the multiplexer level. The module’s published isolation specification is 1,500 V between the field side and the logic side, not between channel groups. Cross-channel rejection exceeds 80 dB from DC to 1 kHz, indicating substantial but not absolute group separation. If a genuine MUX failure occurred, adjacent-channel crosstalk or erratic readings would typically accompany the dead group. A clean zero across exactly eight channels more strongly suggests a configuration boundary. The Proficy Machine Edition software defaults to enabling only a limited channel set during initial setup. Field engineers frequently overlook the requirement to explicitly activate channels 9–16 in the hardware configuration table.

Quantitative Evidence from Power and Configuration Checks

The ALG223 requires 120 mA from the backplane +5 V bus and an additional 65 mA from an external +24 VDC supply. During the investigation, measured current draw on the external supply totaled 64 mA. This value fell marginally below the specified threshold. More critically, the external supply fed only the first eight loop circuits. A separate 24 V feed had been routed to the second group during a panel modification six months prior. That second feed measured 21.3 V under load, below the 20–30 VDC operating window with 10% ripple tolerance. The ALG223’s bottom LED, labeled “User Supply OK,” remained illuminated because the module monitors the primary supply terminal. The degraded secondary feed escaped detection entirely. A comparison of %AI register values during a forced 12 mA injection confirmed that channels 1–8 responded correctly, while channels 9–16 remained at zero even with valid loop current present at the terminals.

Configuration Audit Reveals the Decisive Clue

A hardware configuration upload from the CPU revealed that the Active Channels parameter was set to 8. The module’s default configuration in Logicmaster and PME enables only channel 1; commissioning engineers must expand this count manually. In this installation, an earlier firmware update had reset the configuration to a stored backup that predated the channel expansion. The %AI reference block allocated only eight words to the module. Channels 9–16 had no assigned memory addresses in the CPU’s reference table. This single configuration parameter fully explains the zero readings. The absence of a MUX failure is confirmed by the module’s MODULE OK LED, which flashed a rapid pattern indicating missing configuration data rather than a hardware fault. A MUX burn-out would not alter the channel-activation count stored in the CPU.

Diagnostic Sequence for Distinguishing MUX from Configuration

Engineers should execute a three-step verification before condemning the module. First, monitor the %AI reference table online and confirm that sixteen distinct addresses are allocated. Second, apply a known 12 mA signal to channel 9 and observe whether any register updates, even erratically. Third, inspect the module’s configuration parameters for Active Channels, Range, and %I Size. A functional MUX failure typically produces stuck values such as 32767 or -32768, not a clean zero across a contiguous block. The zero-count signature aligns with deactivated channels. The ALG223 consumes 1 to 16 %AI references depending on enabled channel count and 8 to 40 %I bits for alarm status. A module configured for eight channels consumes eight %AI words. The CPU never polls the upper group, so the registers remain at their initialized zero state.

Corrective Action and Verified Outcome

Restoring the Active Channels parameter to 16 and downloading the revised configuration resolved the fault. Channels 9–16 immediately began reporting counts consistent with the applied loop currents. A 4.00 mA input produced a count of 0 on the 4–20 mA range; a 20.00 mA input produced 32,000 counts. The absolute accuracy at 25°C measured within ±0.25% of full scale, confirming that the analog front end remained healthy. The external 24 V feed was replaced with a dedicated 24 VDC, 2 A supply, restoring terminal voltage to 24.1 V under full loop load. The module’s User Supply OK LED remained steady throughout a 72-hour burn-in. No evidence of multiplexer degradation emerged. The root cause was configuration loss, not hardware destruction.

Practical Guidance for Maintenance Teams

Treat a clean zero on a contiguous channel group as a configuration symptom first. Verify the Active Channels parameter against the physical wiring schedule. Confirm that the %AI reference block accommodates all sixteen words. Measure the external 24 VDC supply at terminals 1 and 19 under loaded conditions. A reading below 20 V or above 30 V invalidates the module’s analog conversion. Use a process calibrator to inject 12 mA into a suspect channel and monitor the corresponding %AI address online. No register movement with valid current and correct configuration justifies module replacement. The ALG223’s 13 ms scan update across sixteen channels provides a reliable benchmark: if the first group updates within this window and the second group shows no activity whatsoever, the CPU is not polling those channels. That observation alone rules out a burned MUX and points squarely to configuration or memory allocation.

Application Case and Solution Scenario

In a typical factory automation retrofit, a maintenance team found channels 9–16 dead on an IC693ALG223 module. The plant had recently restored a backup configuration after a CPU firmware upgrade. The team first verified loop current with a Fluke 787 and confirmed 12 mA at the terminals. Next, they uploaded the hardware configuration and found Active Channels set to 8. They restored the parameter to 16 and downloaded the change. Channels 9–16 immediately reported valid counts. The team also replaced the secondary 24 V feed with a dedicated 2 A supply. This case shows that configuration loss, not MUX failure, often explains group zero-reading faults in industrial automation and control systems.

Frequently Asked Questions

Q1: Why do channels 9–16 read zero while channels 1–8 work normally?
A: The most common cause is a configuration setting. The Active Channels parameter may be set to 8 instead of 16. As a result, the CPU allocates only eight %AI words and never polls the upper group.

Q2: How can I tell a MUX failure from a configuration problem?
A: A true MUX failure usually produces stuck values such as 32767 or -32768. It may also cause erratic readings or crosstalk. A clean zero across a contiguous block points to deactivated channels, not hardware damage.

Q3: What voltage should I measure on the external 24 VDC supply?
A: Measure at terminals 1 and 19 under loaded conditions. The ALG223 requires 20–30 VDC with 10% ripple tolerance. A reading below 20 V or above 30 V invalidates analog conversion.

Q4: Does the ALG223 have a field-replaceable group MUX?
A: No. The ALG223 architecture does not publish a discrete group-level multiplexer as a field-replaceable element. Therefore, engineers should rule out configuration and power integrity first.

Q5: What is the correct Active Channels setting for all sixteen channels?
A: Set Active Channels to 16. Confirm that the %AI reference block allocates sixteen words. Then download the revised configuration to the CPU.