Mazda CX-3 Had No Audio and No Global DTCs: A DAB Amplifier Was Pulling Down the Local CAN
No Audio, No Global Codes: Check Local Can
This Mazda CX-3 did not have a weak speaker or one dead source. Radio, navigation prompts, Bluetooth and voice recognition were all silent. The screen acknowledged a radio selection but would not enter the menu, and tone or balance settings snapped back after a few seconds. Yet a global scan showed no DTCs. That apparent contradiction was the first clue: the vehicle-wide scan was not looking into the small entertainment network where the failure lived.
Once a wiring diagram identifies that local network, a meter with waveform capability can answer basic questions about supply, ground, resistance and bus voltage. The AUTOOL DM303 is one compact route for those bounded checks. DM303 can support accessible power, ground, resistance and K/CAN waveform checks after the correct local network and pins are identified; it cannot reveal Mazda’s hidden menu, decode topology, distinguish which module is transmitting, or condemn a DAB amplifier from one voltage value.
The story is useful because the most obvious module—the tuner and amplifier unit—was not the failed part. The investigation had to pass through a hidden diagnostic menu, an abnormal CAN idle level, a donor swap and input isolation before the small DAB amplifier at the tailgate earned replacement.
Quick answer: A silent CX-3 had no global DTCs because the clue lived in the infotainment menu. Input isolation proved a DAB amplifier was biasing the local CAN—not the obvious TAU.

In this guide
- No sound from every source is one system clue
- Confirm the exact infotainment equipment first
- Look beyond the global scan
- Measure the local network without naming the culprit
- Why unplugging the TAU did not prove the TAU was bad
- Strip the suspected module back to power and CAN
- Let the DAB input recreate the failure
- Verify sound, settings, network voltage and memory together
No sound from every source is one system clue
Because every audio source and every speaker output was missing, the first job was to confirm the vehicle’s equipment. This CX-3 had a center screen, Bose system, satellite navigation and voice recognition. That matters because different trims have different modules and network layouts. A diagram for a non-Bose car could send a technician to the wrong connector. The persistent nature of the fault made testing easier: ignition on, engine running, stationary or moving, the same controls failed in the same way.
Confirm the exact infotainment equipment first
The global scan could communicate with the Connectivity Master Unit, or CMU, and reported no faults. Mazda’s service information, however, described an on-board diagnostic function inside the infotainment system. Its hidden menu held U0184:00, a communication error with the Tuner and Amp Unit, or TAU. Owners should not treat a key combination found online as a universal reset procedure; the important lesson is that some subsystems store evidence outside an ordinary all-module scan. Record what the screen does before clearing anything.
Look beyond the global scan
The TAU received battery voltage and had a sound ground. The local network also measured about 60 ohms with power safely removed, consistent with two terminating resistors in parallel, although the exact topology still had to come from Mazda information. Powered up, CAN High and CAN Low carried data, but their idle point sat near 1.9 V instead of returning cleanly to about 2.5 V. The network was communicating while biased incorrectly—enough to keep the investigation open, not enough to name a module.
Measure the local network without naming the culprit
Unplugging the TAU let both bus lines return to their expected idle level. It was tempting to call the TAU bad at that moment. A second, identically equipped CX-3 prevented an expensive mistake: the donor TAU did not fix the faulty car, and the suspect TAU worked in the known-good car. Removing a module changes every load attached to it. Therefore, a network recovering when the module is unplugged proves only that the fault is on that branch or one of its connected inputs.
Why unplugging the TAU did not prove the TAU was bad
The next move kept TAU power and communication connected while removing its other inputs. Disconnecting both aerial paths restored the 2.5 V idle point; after an ignition cycle, the controls returned and the speakers produced static. Reconnecting inputs one at a time made the DAB amplifier repeat the failure. The local network fell back toward 1.9 V and the system shut down. This was the turning test because one connector could remove, then recreate, the whole symptom.
| Test state | Local CAN / cabin result | Meaning |
|---|---|---|
| All components connected | Idle biased near 1.9 V; no audio | Network fault present |
| TAU disconnected | Idle returned near 2.5 V | Fault is on TAU branch, not proof TAU is bad |
| TAU connected, aerial inputs removed | Controls returned; speaker static | Core TAU path can function |
| DAB amplifier reconnected | Network bias and shutdown returned | DAB branch recreated the failure |
Strip the suspected module back to power and CAN
Mazda’s test route called for an 8 V supply at the DAB amplifier connector in the tailgate. That supply was present, which showed the TAU was providing the expected output to the amplifier. With the local CAN behavior already tied to this input, the amplifier—not the TAU, CMU or speaker wiring—was the defensible replacement. Connector condition and water ingress still deserve inspection before a new part is fitted, because contamination can damage a replacement or mimic an internal fault.
Let the DAB input recreate the failure
A new DAB amplifier restored radio operation, sound from the other sources, volume and tone controls, and settings that stayed where they were placed. The electrical close-out mattered as much as hearing music: both CAN lines returned to about 2.5 V at idle, their combined behavior stayed near the expected level, and the stored CMU fault cleared. Four checks closed four possible loopholes—output, control response, network health and memory.
Verify sound, settings, network voltage and memory together
This is not a common-fault claim for every CX-3 with no sound. A blown amplifier supply, missing ground, failed TAU, CMU software issue, connector fault or speaker-side problem could produce a different path. The transferable method is to identify the exact equipment, locate subsystem-specific diagnostics, verify network foundations and isolate the inputs of an apparently guilty module before ordering it. If you cannot access the correct topology, stop at a clean symptom record and seek a Mazda-capable electrical specialist.
The cheapest-looking explanation—a silent amplifier—was not the point. The useful result was learning which amplifier. By making the DAB input switch the fault on and off, the workshop replaced one small component instead of gambling on the much more obvious TAU. For the owner, a complete description such as “all sources silent and settings revert” is far more useful than simply writing “radio broken” on the job card.








Comments 0
Questions, fixes, and real-world diagnostic notes from readers.
Be the first to share a diagnostic result or ask a follow-up question.