A RevPi Connect 3+ with a RevPi DIO can report a healthy process-image connection while a loopback test still fails. In the reported case, the fault was terminal identification: the jumper connected two outputs, not an output to an input. The DIO inputs are numbered from the bottom upward, so identifying a channel as the “second pin down” selected the wrong terminal group.
Confirmed Root Cause
The failed test did not demonstrate an input, bridge, or shared-power-supply failure. The supposed output-to-input jumper connected two output terminals. Because no DIO input received the output signal, every monitored input remained at 0.
Use the module terminal diagram and count the input terminals from bottom to top. Do not infer the input number by counting downward from the top of the connector.
Observed Electrical Behavior
| Observation | Engineering interpretation |
|---|---|
| The selected output measured 23.9 V when commanded on | The output circuit produced the expected nominal 24 V-class signal; this alone did not prove that the jumper reached an input. |
| No input changed to 1 | Consistent with the jumper terminating on another output rather than an input. |
| A breadboard button also failed to produce an input | Terminal selection remained the first issue to correct before diagnosing the external contact circuit. |
| Power indicators were green | Power was present, but indicators could not validate channel-to-terminal mapping. |
Device and Process-Image Evidence
piTest -d found two modules and marked both present:
Address 0: RevPi Core V1.2, module type 95 (0x5f)
input offset 113, length 6
output offset 119, length 5
Address 31: RevPi DIO V1.4, module type 96 (0x60)
input offset 0, length 70
output offset 70, length 18The kernel log also reported two devices, completed DIO communication initialization with piDIOComm_Init done 0, set the bridge state to running, and started data exchange. These records support a functioning module connection and process-image setup; they do not verify the field wiring on an individual terminal.
Why the Log Shows Different Intermediate Offsets
During initialization, the kernel log first displayed RevPi DIO input offset 11 and output offset 81. It then recorded an adjustment and showed the effective DIO offsets as input 0 and output 70, matching piTest -d. Use the final adjusted values reported by piTest -d when interpreting the active process image rather than treating the earlier initialization values as a second configuration.
Correct the Loopback Test
- De-energize the field wiring before moving the jumper.
- Identify the input terminal group from the module terminal diagram. Count its inputs from bottom to top.
- Identify the intended output independently; do not assume that an equal physical position on the input and output connectors represents the same channel number.
- Move the jumper so it connects one confirmed output terminal to one confirmed input terminal.
- Re-energize the circuit and command only the selected output on.
- Measure the signal at the actual input terminal, then read the corresponding input through the process image.
- Command the output off and confirm that both the terminal voltage and input state return low.
Assessing the Shared Power Supply
The evidence does not establish that using one supply for the Connect, DIO, and X2 connections caused the failure. The module was present, the bridge reached the running state, data exchange started, and the energized output measured 23.9 V. The confirmed wiring error is sufficient to explain why no input became 1.
After correcting the terminal mapping, verify supply reference wiring only if the input still fails. Measure at the input terminal relative to the reference used by that input circuit. The evidence does not provide the required input thresholds, connector pinout, or supply wiring specification, so determine acceptance limits from the applicable RevPi DIO documentation rather than inventing a voltage threshold.
Verification and Escalation Decision
A successful repair requires both transitions: the correctly mapped input must become 1 when the confirmed output is energized and return to 0 when it is de-energized. If the measured input signal changes but the process-image value does not, retain the piTest -d inventory and relevant initialization log lines, then investigate channel configuration and process-image access. If the signal never reaches the confirmed input terminal, continue with the field wiring, terminal selection, and supply-reference checks before treating the DIO as defective.
FAQ
Why does my RevPi DIO input stay at 0 when the output is on?
First verify the physical terminals. In this case, the jumper connected two outputs, so no input received the signal even though the energized output measured 23.9 V.
Which direction are RevPi DIO inputs numbered?
The evidence identifies the input order as bottom to top. Count from the bottom of the input terminal group and confirm the position against the module terminal diagram.
What RevPi DIO offsets should I use from piTest -d?
For the reported RevPi DIO V1.4 at address 31, the final adjusted values were input offset 0 with length 70 and output offset 70 with length 18. Earlier offsets 11 and 81 appeared during initialization before adjustment.
