Troubleshooting
One table per area. Find the symptom, read the cause, do the fix. When the fix does not help, see Report a problem.
Power and LEDs
| Symptom |
Cause |
What to do |
| 5V IN is dark with the charger in |
The charger has no power, or it is in the wrong port. |
Check the charger. Use the 5V 3A port (PWR ONLY in the software). |
| USB is dark with the host cable in |
The host cable is in 5V 3A. |
Move it to the PC port. |
| USB is lit but PC stays dark |
The host has not opened the unit. |
Start the software. On Linux, run aectest setup-linux and plug the unit again. |
| PC stays dark after a plug-in, and the software does not list the unit |
The computer did not register the unit on that port. Some hub ports do this. |
Unplug the unit. Wait 5 s. Plug the unit again, or use another port on the computer. A port direct on the computer works best. |
| POWER flickers |
The 5 V source is too weak. The rail restarts. |
Use a 5 V, 3 A charger on 5V 3A. Never a computer port alone for high power. |
| PRES or INT is lit with no cable |
Normal while the power switch (MOD PWR) is off. |
Ignore it unless POWER is lit. |
| LPMD is dark while on a charger alone |
The cable is in high power without a host. |
Not allowed. Plug the host and let the software return the cable to low power. |
Connection
| Symptom |
Cause |
What to do |
No unit on the host |
The PC port is not connected, or another program holds the unit. |
Plug the PC port. Close the other program. Click Refresh units. |
no FTDI device found on the command line |
The USB chip of the unit is not on the bus, or the rules are missing. |
Check the cable. On Linux, run aectest setup-linux. |
Unit disconnected or No such device |
The unit left the USB bus during a run. |
Unplug the host cable and the charger. Wait 30 s. Plug the host cable. Click Refresh units. |
Server not reachable |
The server process stopped. |
Start it again. The page reconnects on its own. |
Server restarted toast |
The server was started again. |
Nothing. The page reloads and keeps your form values. |
| A second program is refused with a pid |
The server owns the unit. |
Stop the server before a command-line run on the same unit. |
Reading a cable
| Symptom |
Cause |
What to do |
NACK from slave |
The cable did not answer. MODE is at MAN and the MSEL slide deselects the cable. |
Set MODE to AUTO. Click Read cable again. |
pins low: module present or rail off |
The cable rail is off, so the pins read low. |
Set the power switch (MOD PWR) to ON. Read the cable. |
interrupt latched since insert |
The cable raised its flags at insertion. |
Read the cable. The read clears them and names them. |
interrupt pending after a read |
The cable holds a flag after the read. |
Open the flags on the Read tab. A real alarm shows there. |
The clock badge says 100 kHz, fallback |
The cable did not answer cleanly at the faster clock. |
Nothing. The unit stays at the slower clock for this cable. |
| The plate shows a class above the budget |
The cable needs more than 6 W. |
The unit cannot test it in high power. Read-only commands still work. |
The guided test
| Symptom |
Cause |
What to do |
| Start the test stays grey |
The cable is not read yet. |
Click Read cable first. |
No criteria for this cable type yet |
No file matches the part number. |
Test without criteria, or derive a file on the Advanced page and ask engineering to confirm it. |
| The pre-check names a switch |
A switch is off its shipped position. |
Set it as the card says. Click Retry. |
Label serial differs |
The scanned label and the cable serial do not match. |
Scan once more. A real mismatch is a FAIL by design. |
Cable removed during the test |
An end came out of the cage. |
Seat it again. Start the test over. |
Connection lost during the run |
The server stopped mid-run. |
Wait for Live in the corner. Start the test again. |
Criteria proposed, not confirmed |
The criteria file is a proposal. |
Hand the record to engineering. Do not bin the cable. |
| INCONCLUSIVE with a code |
See the code table. |
Test a cable lists every code with its fix. |
High power and temperature
| Symptom |
Cause |
What to do |
| A yellow WARN line |
The cable reached its own warning level. |
Nothing. The test continues. Check the fan. |
| THROTTLE, then INCONCLUSIVE THERMAL |
The cable came within 5 °C of its alarm level, or reached 70 °C. |
Let it cool. Put air over the heat sink. Run again. |
| THERMAL repeats with the fan running |
The cable runs hot on its own. |
Hand the cable to engineering with the record. |
Budget refused |
The cable declares more power than the budget. |
Raise the budget with a reason on Advanced, or test a cable within the budget. |
Refused: confirmations |
A confirmation box is not ticked. |
Tick both boxes before a high-power command. |
Firmware
| Symptom |
Cause |
What to do |
Every firmware command answers status 46h |
The cable wants the host password. |
Type the password. Try again. |
| Plan and Download stay grey |
No image file is chosen. |
Choose the file. |
RUN REFUSED |
The cable refused to run the new image. |
The old image runs. The new one sits in its bank. Try Run again or use the vendor tool. |
MODULE SILENT after Run |
The cable did not answer after the reset. |
Power cycle the unit. The committed image boots. |
The firmware update engine is not in this build |
The engine is not installed. |
Firmware info still works. Ask for a build with the engine. |
More on each case: Firmware update.
Report a problem
Open an issue on GitHub. One issue per problem. Attach:
- the unit serial
CT2M-NN and the cable serial,
- the output of
aectest --version,
- the report from the verdict card, or the session folder as a zip (see Records),
- the terminal output of the run,
- for a wrong page, the page name and the sentence.