EDM 030 Dead, no beep, no light

[Repairman]

EDM 030

Symptoms: Dead, no beep, no light. Current draw is 0.006–0.007 A and remains steady.

SB seems dead: TX/RX are at 0.2 V.
I replace the SB with a new one, but the exact same problem remains.

I check the board and find that 1R0, instead of outputting 5 V, only outputs 0.30 V.

I remove F7001 and F7002 to isolate the circuit.
I inject 3 V / 2 A at 1R0, and the TPS53219 heats up. I replace it, but nothing changes.

I replace the high-side MOSFET as well, but still no change. Same exact problem.

I then check the DA9065 and find that all rails are missing:

• Where I should have 5 V, I only have 0.30 V
• Where I should have 3.3 V, I only have 0.12 V

I replace the DA9065, but the exact same problem remains.

I also check the Wi-Fi section: under the capacitors, there is no 3.3 V rail.

[Repairman]

@Calvin My friend, this is a tough one!

[Repairman]

Update:

I replaced the TPS53219 again, this time with one from a donor board, and the 5V rail is back.

DA9065: only the 5V rail is present. All the other rails are missing, and in the yellow circled area I only get 0.09V.

I replaced the DA9065 again with one from a donor board, but no luck. Same problem.

I also tried injecting 3V / 2A into the Wi-Fi module rail (the 3.3V rail generated in the F7002 area is also missing), but nothing heats up.

[Calvin]

If you inject voltage please do not exceed 1V. Some signal and power lines are not designed to handel more than 1V. If due to a short in a chip power rail and low volt signal lines are touching it could do more damage.

I would focus first on the southbridge and if solved move over to the WiFi pmics:
The SB pmic (DA9065) needs 2.45V at pin 29 to enable the standby voltages.

This 2.45V is the PG (power good signal) from the TPS62823. (orange marking)

If 5V reaches at the TPS62823 pin 7 (Vin) and pin 1 (EN) there should be an output 1.95V at pin 6 (SW). If output is missing I would change the dcdc converter chip.

[Repairman]

Thanks to your suggestion, I’ve made some progress.

TPS62823:

5.15V on pin 7

1.45V on pin 1

As a result, SW is not outputting anything.

I checked the resistor before pin 1: on one side it correctly measures 5.15V, while on the other side I only get 1.45V.

Could the TPS62823 be pulling the voltage down, or could the resistor itself be faulty?

[Calvin]

The resistor should have 10k Ohm.

This are my readings at a working board:

[Repairman]

Okay, I’ll check it right away.

If I wanted to replace it, can I find the same IC on an EDM-010 donor board? I might be mistaken, but I think the one on the EDM-010 has 12 pins instead of 8.

[Calvin]

The dcdc converters (TLV62090 qfn16) used on the edm-010 are not compatible. Needs to be at least a edm-020 or upwards.

[Repairman]

I removed the TPS62823.

Now the voltage across the 10k resistor has returned to almost normal:

5.16 V on one side

around 4.96 V on the other side, although it keeps fluctuating between approximately 4.4–4.9 V

[Calvin]

Now we need a replacement for the old TPS62823

[Repairman]

The replacement part has arrived

A3 replaced and incredible… Same problem. But what could pull down the rail?

[Calvin]

That’s strange. The PG from the mosfet controller TPS53219A is the EN for the the TPS62823.

You removed the TPS62823 and the 5V came back. So I’m surprised that with a new TPS62823 the voltage drops again.

I would change the mosfet controller.

[Repairman]

TPS53219A:

PG is indeed only 1.45V, so the problem seems to originate from this circuit.

I found another anomaly:

TPS53219A: On pin 14, across the 5R6 resistor, I measure a fluctuating voltage between approximately 4V and 6V instead of the expected 12V.

I removed the 5R6 resistor to rule out anything downstream pulling the voltage down, but the pad directly coming from the TPS53219A still shows the same abnormal fluctuating voltage.

The small capacitor above the 6R8 resistor also shows the same fluctuating voltage.

I have already replaced the TPS53219A, but the problem remains unchanged.

The capacitors connected in parallel all have a stable 12V.

The two MOSFETs above the controller also have the expected 12V and 5V supplies, and the 1R0 inductor outputs a stable 5.16V as expected.

[Calvin]

Here are my readings from the TPS53219 at the 5V_ACPI line

Is it possible that the output at the highside (left) mosfet is fluctuating?

[Repairman]

On the left high-side MOSFET I have 12.14V stable on the input and 5.15V stable on the output.
However, I found some significant differences compared with your measurements:
RF (pin 4): 0.03V
VREG/VDRV: 5.14V

[Calvin]

RF is the selector for the switching frequence and open (not connected to ground = 500 kHz are selected)
VREG is a 6.2V LDO output. At the PS5 ciruit connected to the VDRV input for the gate drive.

I don’t get my head around why PG and VREG are lower than they should be while the mosfet is generating propper 5.1V…

[Repairman]

I finally fixed it!

I had tried replacing the TPS53219A, both MOSFETs, and several other components, but the fault remained unchanged.

Then I noticed something strange by chance: whenever I touched the bootstrap area with my multimeter probe, the current draw dropped to zero. As soon as I lifted the probe, the current draw returned.

I started probing and inspecting different areas of the PCB and eventually noticed that a tiny 1 µF capacitor was missing from the underside of the board, near F7001. Interestingly, touching that exact area with the probe also caused the current draw to drop to zero.

My theory is that the previous technician accidentally knocked the capacitor off while opening or disassembling the console, either with a screwdriver or by hitting it with the metal APU retention bracket. On the EDM-030, that bracket sits extremely close to this capacitor.

I suspect the missing capacitor was affecting the stability of the bootstrap circuit and consequently the high-side MOSFET drive.

I soldered a replacement capacitor in place and BOOM! IT WORKS!

The previously unstable voltage reading at VBST is now stable at around 12 V, and the 2 V rail supplying the Southbridge has been restored.

All that troubleshooting, and the culprit turned out to be one tiny missing capacitor!

[jkyoho]

you mean this guy?

[Repairman]

Here he is, the culprit in the photo! This little guy was the problem! Apparently, its absence causes the bootstrap voltage of the TPS53219A to drop.

[Calvin]

Nice find and an intersting one. So the cap was missing and that caused the VREG and PG voltage to drop.