My MAX7.2 has suddenly exposed a failure mode and is presenting fault code F52 on the LCD display.

The internet and available documentation describe this fault as “Bus Voltage Too Low.” I have also tried to gather as much information as possible from this forum and others. However, based on what I could find, it appears to be a fairly random fault, and I could not locate any posts that led to a definitive solution.

I have not opened the machine yet, as I want to gain as much insight into the problem as possible beforehand.

So far, I have run the following functional tests, and in my case the failure is deterministic, not random at all.

---

Test 1
Conditions: Battery available (52.7 VDC), NO PV, NO GRID

When I switch the inverter on, it enters its diagnostic phase and then outputs 220 VAC on the load output. It also asserts a relay, most probably the load (inverter) relay. After less than one second, the 220 VAC collapses and the inverter shuts down, displaying F52 on the LCD.

---

Test 2
Conditions: Battery available (52.7 VDC), PV ON, NO GRID (PV available under sunny conditions)

When I switch the inverter on, it enters its diagnostic phase and outputs 220 VAC on the load output, again asserting the load relay. The output remains stable and no fault occurs.

However, the PV does not charge the battery; it only powers the AC load output. The settings that allow PV battery charging are correctly configured.

---

Test 3
Conditions: Battery available (52.7 VDC), PV OFF, GRID ON

When I switch the inverter on, it enters its diagnostic phase and outputs 220 VAC on the load output, asserting the load relay. The output remains stable and no fault occurs.

However, the grid does not charge the battery, even though the settings to allow grid charging are correctly configured.

---

Initial Assessment

Based on the above conditions and observations, the F52 fault strongly suggests that the DC-DC converter, which is responsible for converting battery voltage to the internal bus voltage (and vice versa), is not functioning.

This explains nearly all observations. The only apparent contradiction is why bus voltage exists initially (for about one second) when running in battery-only mode. After some thought, I realized that a bus soft-start circuit is present. The machine uses this low-power soft-start path to pre-charge the bus capacitors and verify bus integrity (checking for shorts, etc.). This accounts for the brief energy available to produce AC for about one second. The bus then collapses because energy is not replenished via the DC-DC converter.

I am confident that the high-side IGBTs and low-side MOSFETs of the DC-DC converter are not blown; otherwise, the soft-start process would fail. My attention is therefore focused on the SG3525 PWM generator, which may not be enabling DC-DC operation.

If the SG3525 itself is functional, then the initial transistor amplifier stages that drive the A and B phase outputs could be faulty. The ±12 V rails are another possibility, but I have largely ruled them out because these rails also power the measurement op-amps, and a fault there would significantly distort the readings.

Of course, the DSP control board could also be at fault by not enabling the SG3525 PWM, or the opto-coupler driver stage may have failed.

Another possible fault could be a failed buck IGBT, but I have tentatively ruled this out because the body diode should still allow some discharge path from the battery to the bus. I hope this assumption is correct.

At this stage, my money is on the SG3525 or its associated circuitry. I certainly hope so, as I am not in the mood to replace power silicon given the amount of mechanical work required to rework this machine’s heatsinks.

I will provide an update when I open the unit during the week, and hopefully this will help someone else in the future who has to deal with the same F52 fault.

On 2026/02/15 at 7:59 PM, BritishRacingGreen said:

I could not locate any posts that led to a definitive solution.

? The AEVA repair page documents two solutions: replacement of the battery-side electrolytic capacitors, and replacement of the SG3525 chip itself. I note that the Axperts can be cranky about the brand of SG3525; best to replace with the same brand as existing, where possible.

On 2026/02/15 at 7:59 PM, BritishRacingGreen said:

Another possible fault could be a failed buck IGBT, but I have tentatively ruled this out because the body diode should still allow some discharge path from the battery to the bus. I hope this assumption is correct.

That logic sounds good to me. The buck stage only actively bucks when charging the battery (in these models, from either solar or utility).

Good luck, and let us know how it goes.