Board-Level Microsoldering | Austin, TX Lab | From $500 | No Data, No Fee
Samsung Galaxy Won't Turn On Data Recovery
A Galaxy that shows nothing on a charger has usually lost its power delivery, not its storage. The files are still in the UFS or eMMC package while a shorted charging path, a failed power IC, or corrosion under the chips keeps the board from starting. We measure the board before we quote it, repair the fault, and hand the phone back for you to unlock.
Does a Galaxy That Won't Turn On Still Have Your Data?
Dead is the wrong word for what most of these boards are doing. Voltage isn't reaching the parts that would put light on the screen, & the storage package sits off to one side of that problem holding the same ciphertext it held the second before the phone went quiet.
Flash cells don't need a working board to keep what is in them. A phone that has been in a drawer for a month holds what it held the day it stopped. Waiting isn't what moves the odds; the next thing somebody plugs into the port is.
Generation changes the bench work rather than the ending. Budget boards such as the A05 & A06 still ship eMMC 5.1, the flagships moved to UFS back at the Galaxy S6 & Note 5 generation, and from the S24 generation the monolithic storage package exposes no usable test points. That whole matrix sits on the Android & Samsung recovery hub. None of it moves the credential; the phone still needs yours.
Stop Charging It Before Anything Else
The instinct after a phone dies is to keep trying things, and that is where the avoidable damage comes from. A short is a low-resistance path, & every attempt to energize the board sends current down it again.
- Take it off the charger. Current going into a shorted board is dissipated as heat in whatever is shorted, which is the same part we need intact enough to identify under a thermal camera.
- Quit cycling the power button. Holding it for another thirty seconds runs a test you already ran. Nothing in the boot chain gets a second opinion from repetition.
- Leave a wet phone off for good. Liquid across points at different potentials turns the board into a galvanic cell, dissolving metal at the anode & plating it back out as conductive dendrites at the cathode. That reaction runs fast under applied power & slowly with the phone off, which is the entire reason the charger matters.
- Refuse a firmware flash. Odin writes Samsung-signed firmware & has no documented path for reading your files out, and the standard CSC binary in a firmware package wipes userdata by design.
- Don't buy parts on a guess. A battery or a charging port ordered off a symptom is a coin flip on a board nobody has measured yet, & the swap itself adds heat cycles to joints that may already be fractured.
- Get it measured while the board is still clean. There is no diagnostic charge here, so a board can be read on the bench before you commit to anything. Mail-in runs from any state & the work stays at the Austin bench.
How Do You Diagnose a Board That Draws No Current?
- Inspect before anything is energized. Under the microscope: bridged or bent pins in the USB-C port, corrosion residue at the connectors, impact marks, & the flux and reflowed solder that say somebody already worked on this board. A phone that has been opened before gets triaged differently from one that has not.
- Clean a corroded board first. Readings taken through a dendrite bridge describe the residue rather than the circuit, so a liquid board is stripped, its shields come off, & it goes through the ultrasonic cleaner before a probe ever touches it.
- Read the current draw at the port. A current-limited DC supply stands in for the charger, which means the board can be fed a known voltage with a ceiling on how much current the fault is allowed to pull.
- Interpret what the draw does. Near-zero draw from a known-good input sends the diagnosis toward the USB & charging group first, & a fault living there never reached the processor or the storage. A draw that comes up and then collapses says converters are starting & something is pulling a rail back down. Neither behavior is a diagnosis by itself, because no current reading separates a failed charging IC from a shorted core rail.
- Map the rails in diode mode. Each rail group is read to ground and compared against known-good values from a working board of the same model. Hard shorts announce themselves at or near 0.000 V; a line that is doing its job answers with a few hundred millivolts.
- Localize the short, then replace. Voltage goes back into the shorted rail with the current ceiling held low, & the FLIR camera finds the part turning that current into heat. The failed component comes off with a Hakko FM-2032 on an FM-203 or FX-951 base, or under Atten 862 hot air when it's a package, & reballs run on a Zhuo Mao station.
- Boot it, then hand it back to you. Once the board completes verified boot and reaches the lock screen, the remaining half of the job is yours: your credential unwraps the credential-encrypted storage on the original hardware, and the files come off the phone normally.
Cleaning after probing wastes the probing, & powering before inspecting can turn a two-component repair into a much longer one.
Which Samsung Power IC Failed?
Samsung builds its own power silicon, and on the Exynos-architecture boards with standard PMIC layouts it arrives as two part families doing two different jobs. Which family is implicated is most of the prognosis.
- S2MPS and S2MPB main PMIC
- Every core rail on the board comes out of this family: the supplies the processor, the RAM, & the storage package all run on. Nothing lights up without them, which is why a failure here looks the same from outside as a failure on the charging side.
- S2MU and S2DOS sub-PMIC
- This family owns the input side: charge control, AFC & Power Delivery negotiation, & the USB interface itself. It's the first suspect when a known-good charger produces no draw or a PC never sees the phone, & a failure in it leaves the processor and the storage package electrically untouched.
There is a third answer that is neither chip. A converter can be healthy while the load it feeds is shorted, so the rail reads dead and the PMIC that drives it is innocent. That is why the probe goes to the rail and not to the part somebody expects to blame.
Diode-Mode Rail Readings and Data Prognosis
Where the short sits decides what the job is and what the answer will be, so the map below is the part of the diagnosis a customer should actually care about.
| Where the probe sits | Diode-mode reading | What a short there implicates | What it means for the data |
|---|---|---|---|
| USB input and charging path | A few hundred millivolts to ground when healthy, at or near 0.000 V when shorted | The S2MU or S2DOS sub-PMIC, the USB-C assembly, or the charging IC on the input side of the board | Best position on this table. The processor and the storage never saw the fault, so the payload waits behind a component replacement |
| Main PMIC outputs feeding the board | Same split: a few hundred millivolts on a healthy line, at or near 0.000 V on a shorted one | The S2MPS or S2MPB main PMIC itself, or a shorted load sitting on the rail it drives | Repairable either way. Which side of the rail is shorted decides whether the PMIC comes off or the load does |
| VDD_CORE | At or near 0.000 V, the reading nobody wants to find on this rail | The processor itself rather than anything feeding it | Worst branch on the board. File-based encryption keeps the keys in the processor's secure hardware, so a destroyed processor is a destroyed key set |
| Joints under the processor or the storage package | A short that reads through the package instead of at any discrete component | Fractured or bridged BGA joints, the class the repair community documents on the S24 Ultra | Mechanical work rather than a parts swap: underfill removal and a reball, or a paired-set transplant when the board is past that |
Diode mode measures a voltage drop from a test point to ground, which is why the numbers only mean something next to a reference. A rail that reads low on one model reads normally on another, so the comparison is against a known-good board of the same model rather than against a number in a table somebody published.
A shorted charging path and a shorted core rail are separated by measurement rather than by how the phone presents. The black screen and the dead port read identically in both cases, but the phone call that follows isn't the same phone call.
A Failed Charging Path Mimics a Dead Motherboard
Two very different phones present identically across a counter. One has a shorted charging IC and an otherwise healthy processor; the other has a processor that is electrically gone. Both are black, silent, cold, and invisible to a computer.
On the Galaxy S24 Ultra the USB-C assembly and its charging IC sit on a sub-board joined to the main board, so a failure in that assembly cuts power flow to everything downstream while the processor and the UFS package sit there untouched. The phone reads as a dead motherboard and isn't one.
A shop reads a phone with no power as a phone with dead storage & quotes to desolder the storage chip, and on an encryption-era Galaxy that read comes back as ciphertext, because the per-file keys are wrapped by the secure environment on the original processor and the package on its own decrypts nothing in a socket or a programmer.
The exception is genuinely old hardware. Galaxy phones that shipped before forced encryption never had a key to lose, so a storage read on those boards can return real files. That is a narrow class of legacy devices, not a technique for a modern phone, and it never applies to a board whose processor is still the one that wrote the data.
A Galaxy that does power up & then loops at the splash screen is a different diagnosis with a different table. Samsung boot loop recovery covers that half of the fault catalog.
Galaxy S24 Ultra Sudden-Death Class
The presentation is specific enough that the board-repair community treats it as its own class. Somebody sets the phone down mid-charge or partway through a video, picks it back up, and it's inert. The screen stays black, the phone doesn't vibrate, & a long press on the power button does nothing. On a bench supply it draws zero or close to it, stays cold to the touch, and never brings up Download Mode.
A board that stays cold has no rail doing enough work to make heat, & a Download Mode that never appears isn't the bootloader passing judgment on anything, since Download Mode sits downstream of power and can't be presented by a processor that never came up. Reading that refusal as a software verdict sends people hunting firmware for the wrong half of the phone.
Three directions account for what the community writes up on this class: shorted BGA joints beneath the processor or the storage package, a failed USB-C sub-board with its charging IC, & the main PMIC. Two of those are component work with the data waiting behind them. The joints are mechanical, and that is the branch where a short under the processor puts the keys in question.
None of this comes from Samsung. It's what people doing board-level work see on the bench and write up, and no failure rate has been published for it, so this page doesn't attach one. What it does mean at intake is that a cold, unresponsive S24 Ultra is a power-delivery presentation and not a verdict on the storage.
So the sequence is measure, then quote. A number that comes before a probe is a guess wearing a price tag.
What Can Recovery Software Do for a Phone With No Power?
That isn't a knock on the software. Smart Switch, Samsung Cloud, and Google's backup are the right answer for a phone that still boots and can still be unlocked, and if that describes your phone we'll tell you so and charge you nothing for the sentence. Copy the files today and the rest of this page stops being your problem.
The interfaces people are pointed at when the phone is dead are worth naming individually, because each one fails for its own reason.
- Odin and Download Mode
- A firmware writer with no documented user-data readout path. The standard CSC binary wipes userdata and cache, and on 2025-and-later flagships the A/B partition structure makes even the userdata-preserving file a live soft-brick risk. It also needs a board that powers, which is the thing this page is about.
- Qualcomm EDL
- A firmware-programming interface, not a decryption tool. A raw read of a file-based-encryption device returns ciphertext because the credential-derived keys never materialize in that session, and many loaders refuse memory reads outright when Secure Boot or the OEM lock is on.
- ADB over USB
- USB debugging has to have been switched on before the phone died, and the host's RSA key has to be approved in a dialog on an unlocked screen. From One UI 6.1.1 Auto Blocker ships enabled by default and blocks USB commands on a locked phone.
- USB deep scan utilities
- They enumerate files, never blocks, so a scan lists what still exists on a working, unlocked phone and nothing more. The demo finding something is the tool listing your live files back to you.
So the order is fixed. Power comes back first & software second, because a phone that can't be switched on isn't a software problem waiting for the right download.
Which Board States End the Conversation?
Every dead board gets adjudicated into one of two piles at intake, and we would rather hand you the bad answer on day one than three weeks and an invoice later. The line between the piles is physics and cryptography, not effort or price.
| What the bench finds | Where it lands | Why |
|---|---|---|
| Shorted charging IC or failed USB-C sub-board | Repairable | Component-level replacement on the input side; the processor and storage were never in the fault path |
| Shorted main PMIC or a shorted load on a core rail | Repairable | The failed part comes off and the rails come back; the encrypted payload never moved |
| Corrosion under the packages, caught before more power went in | Repairable, and time-sensitive | Ultrasonic cleaning halts the reaction, then damaged components get replaced. The damage is progressive, so the arrival date matters |
| Fractured joints under the processor or storage package | Repairable, mechanically | Underfill removal and a reball, or a paired-set transplant moving processor, RAM, and storage together, plus the secure element on the Galaxy S21 and newer flagships that carry one |
| Cracked or electrically destroyed processor die | Terminal | The hardware-held keys died with it, so moving the storage to a donor board recovers nothing readable |
| Dead UFS controller on an encrypted phone | Terminal | The host cannot query the device descriptor and the payload behind that controller is hardware encrypted, so no fallback of the kind legacy eMMC sometimes allows |
| Factory reset already completed | Terminal | The key material for every file was destroyed at once, and no scan reverses that |
| Tripped Knox warranty fuse, or a cloud backup with no recovery code | Terminal for that data | A tripped fuse withholds the Secure Folder container keys permanently, and an end-to-end-encrypted Samsung Cloud backup needs the owner's recovery code |
Your credential is the other boundary here, and it isn't one we pretend to move. The keys for credential-encrypted storage are derived from your PIN, pattern, or password combined with secrets only the phone's secure hardware releases, & wrong attempts are throttled inside that hardware by Gatekeeper and Weaver rather than by anything a lab could hurry along.
Our half of the job ends at a phone that boots. Yours starts at the lock screen. A screen lock, Factory Reset Protection, or the Knox warranty fuse is not something this bench removes or works around, & unlock jobs get declined at intake.
The claim that a raw storage read returns only ciphertext holds on Exynos and Qualcomm Galaxy phones and on any model carrying a discrete secure element. Older MediaTek budget parts without a discrete secure element sit inside a documented BootROM exception class that forensic operators reach on a live board; it stops well short of flagships and Knox Vault models, & it is not something this lab performs, prices, or offers.
What Does Recovery From a Dead Galaxy Cost?
No data, no fee. Free evaluation. No diagnostic charges. The number is quoted firm after the board has been measured, so nothing gets authorized on an estimate, and a paired-set transplant doesn't carry a surcharge on top of it.
All of it happens in-house at 2410 San Antonio Street in Austin, Texas, in one shop that has been doing board-level work since 2008. Nothing is franchised & nothing goes out to a partner lab. A board that turns out to be terminal costs you nothing.
Drop it off during business hours, or put it in the mail from any state; the bench is the same either way. A free evaluation comes first, then a firm number, then your decision.
Data Recovery Standards & Verification
Our Austin lab operates on a transparency-first model. We use industry-standard recovery tools, including PC-3000 and DeepSpar, combined with strict environmental controls to maintain drive integrity. This approach allows us to serve clients nationwide with consistent technical standards.
Open-drive work is performed in a ULPA-filtered laminar-flow bench, validated to 0.02 µm particle count, verified using TSI P-Trak instrumentation.
Transparent History
Serving clients nationwide via mail-in service since 2008. Our lead engineer holds PC-3000 and HEX Akademia certifications for hard drive firmware repair and mechanical recovery.
Media Coverage
Our repair work has been covered by The Wall Street Journal and Business Insider, with CBC News reporting on our pricing transparency. Louis Rossmann has testified in Right to Repair hearings in multiple states and founded the Repair Preservation Group.
Aligned Incentives
Our "No Data, No Charge" policy means we assume the risk of the recovery attempt, not the client.
Technical Oversight
Louis Rossmann
Our engineers review all lab protocols to maintain technical accuracy and honest service. Since 2008, his focus has been on clear technical communication and accurate diagnostics rather than sales-driven explanations.
We believe in proving standards rather than just stating them. We use TSI P-Trak instrumentation to verify that clean-air benchmarks are met before any drive is opened.
See our clean bench validation data and particle test videoQuestions About Galaxy Phones That Draw No Power
Can any software get photos off a Samsung that won't turn on?
Will a new battery fix a phone that suddenly went dead?
Does a dead Galaxy mean the storage chip has to be desoldered?
A shop offered to replace the whole motherboard. Does my data come with it?
It got wet weeks ago and I kept charging it. Is it too late?
The phone was locked when it died. Does that change the job?
What does dead-board Samsung data recovery cost?
Related services
Related Samsung Recovery Pages
Generations, encryption reality, and the rest of the fault catalog
When the phone does power up but never finishes booting
Controller death and what sits behind it on legacy boards
How to ship a phone to the Austin lab safely
What happens when a recovery does not produce data
Free evaluation, no diagnostic charges
Galaxy dead on the charger?
Send it before more current goes through the board. Evaluation is free, the quote is firm, and a recovery that produces nothing is not billed.