Secure digital card recovery is steady work here, due to the fact that cards are used so heavily up north and treated poorly. We receive wedding images, stills from surveys, aerial drone footage of the coastline, dash cam cards that sat outside through a winter at minus five, and a fortnight's holiday photos stored all on one SDXC card without a backup anywhere. Nothing anyone owns is less protected. The cards we recover come from professional working photographers, from offshore contractors and from families who have lost their summer, and it is usually those same images that we send back home.
All memory card jobs are diagnosed at no charge. You get a firm quote for your recovery in writing prior to any work being done on your card.
In most cases you only pay if we recover your data. The exceptions are electronic/mechanical failures, chip level work, DVR and forensic jobs. All pricing information can be found on the data recovery cost page.
The first step with all memory cards is to determine what the problem with your device actually is. Twenty-odd years in, these twenty-five faults cover almost every possible one that will ever make it to our workbench.
You have tried three host devices and got the same answer, but the problem isn't with the memory. The problem lies in the controller or firmware itself. This means none of your pictures have been removed from the NAND memory, even though nothing can ask for them at the moment.
The problem is that the file table is broken, not the images themselves. Say no; agreeing may cause what was once an inexpensive job to be turned into an expensive one, faster than anything else on this page. Remove the card, and do nothing further.
This is the fault we see most often, and this is also the fault that owners will be wrong about the most. Typically on an Extreme PRO SDSDXXD, an Extreme SDSDXVE, an Ultra SDSDUNC or an SDSQUAB what fails is either the FAT or exFAT structure defining the card, or the controller's own mapping layer underneath. The images themselves have never been touched. The host will tell you the card is corrupted, unreadable or empty and offer to format. If at this point in time you decide to format again in-camera, you destroy your shoot. Stop here, take it out of the camera, get a quote first and then send it to us.
Frames remain in the flash unit until there has been new or "fresh" data recorded to them. Put the card in a drawer, not a camera case, and for sure do not use it for another shoot. Doing this will bring most of your files from a lost folder back intact.
Quick formats are used by cameras (and only those), which means no deep ones, and therefore the shoot underneath will be preserved and lifted intact. The only way this recovery can be spoiled is when you continue to take pictures after the event.
Wallets, slots, and low-quality adapters all damage cards. As long as the crack did miss the inner tracks, the die will be probed and read at that point; therefore, a card broken in two is worth getting a quote for and posting instead of just binning.
That number does not measure anything. That is the fallback ID of a controller after it crashes. The firmware state; that is about all, and the rest is at the chip level, and as such, is completely routine to work with.
There are two types of problems that may appear to a computer as if they have occurred due to the same reason: a controller locks up while it fails, or someone moves the little slider on a full-size SD card across. Of these issues, one takes approximately ten seconds to check and thus should be examined first.
You would be surprised how much cards can stand. Cleaning, stabilising, reading them at chip level, and normally a laundered card returns all its stored data.
Counterfeits wrap new writes back over one very small genuine section of the card as soon as it is full. What was written prior to the wrap gets restored. Whatever was written afterwards has been lost because nothing was ever saved, therefore there's no way for any lab to recreate it.
The files were only partially moved and part of the directory was cut off. We do not trust the remaining information for the file system. Therefore, we carve frames from the raw capture and rebuild every frame from its respective header.
Within a couple of years continuous video captures wear out the internal storage (flash) on this unit. Cold starts due to an Aberdeen winter don't help either. When carving whatever blocks still contain readable information the captured footage is restored. The clip which caused you to make the call is usually included in those blocks.
All components (controller and memory) are contained within a single solid slab, therefore no component packages can be removed. Test points are identified against pinout records, and we address the memory directly. More work than usual; same results as usual.
Type B cards in mirrorless cameras and survey drones will fail hard at high data transfer rates for film, and will fail completely without warning. They internally function like a PCIe device; therefore, our team dealing with NVMe has to deal with these units.
A sealed SDSQXCD type card used in a waterproof enclosure, used continuously throughout the summer along a coastline and up rivers will age significantly faster. The heat combined with the vibration accelerate the read error rate by approximately one year, and typically, the controller fails before the memory does.
Kingston's SDCS2 budget line can drop to a dead state and stop responding to the host completely. Once you get past the dead controller state, there are no problems reading the memory underneath.
Fake Samsungs will corrupt right where the write wraps around past the real limit of the part. The genuine span carves out cleanly. No one here is convinced by the stickers on the fronts.
Any long uninterrupted sequence can hang both the 633x and 1066x cards and scramble everything that was being written at the point that they hung. The card is rebuilt using a raw read, since what the card claims about its own condition after hanging is unreliable.
The power went off before the index could be written; therefore none of the file contains any information telling a player where to begin. However all of the video data exists. The structure simply needs to be rebuilt around the data that does exist, which comes directly from the stream.
Corrupting wear tables creates cards containing "zero-byte" ghost images wearing the correct name and date. The actual contents of the card remain on the underlying flash memory, but are currently inaccessible due to a mapping table that has been corrupted and requires rebuilding.
Unmarked generic readers will cause the cards to exceed their voltage limits and then release them during a write operation if they lose contact. The card will then pay for the saving on the reader, and a chip-level read will have to recover the shoot.
Fill up a card in 2016, store it in a drawer, and it loses charge over the next ten years. The card may well still mount, but you'll find that your files are now riddled with errors. Repeatedly reading from the media at a slow pace, giving ECC time to correct errors, yields far more than a single hopeful copy ever will.
Sustained 4K and high bit-rate video demands a write speed that will hold continuously, not in bursts. If the card can't keep up it overruns the buffer, clipping your files, and the directory says they stopped early. What landed gets carved out. What never arrived is gone.
Half a format, then dead battery. The card now has half of its old file structure on board plus half of the new one. Most hosts refuse to accept it after this happens. A better situation than it looks, because since the format didn't finish, it didn't finish erasing the card either.
Applications which offer to fix a damaged card or save what they find back onto the same damaged card are writing over blocks being read. We see this every week. Stop work on a damaged card as soon as possible. Get a quote before moving forward. Then post the card untouched.
A memory card consists of a piece of NAND (the actual memory) and a small controller inside a plastic housing. Nothing is shock mounted, there are no backup systems and there is nothing kept in reserve. For that reason, the process for recovering data from memory cards follows the "image first and then inspect" model. The process requires a copy of every sector of the memory card be made prior to anyone even attempting to view any one individual file. This keeps the original from taking a second knock, and it is why a card that has been through a botched attempt somewhere else is usually still salvageable when it lands here. Format type does not matter. It could be a SanDisk Extreme, Extreme PRO or Ultra SDXC, a Lexar Professional, a Kingston Canvas Go Plus, a Samsung PRO Endurance (from either a dashcam or doorbell), a Sony TOUGH or ProGrade Digital, older CompactFlash from your professional camera body, a Sony XQD, or CFexpress Type A and Type B from current mirrorless kit. Regardless of whether you have a UHS-I or UHS-II interface, regardless of whether the card is rated V30 or V90, regardless of whether it is labelled A1 or A2: it all works exactly the same way. The only thing different is how long it takes.
This is an important explanation, since the phrasing causes people to panic, and do the exact thing which destroys the data. When a camera indicates that a SanDisk SD card is corrupted, or when a laptop states that a SanDisk SD card is corrupted, it is almost never telling you that your photos are damaged. On a memory card two different types of items exist. First, there are the actual images themselves. These images are written into NAND pages. Second, there are a few structures to help describe where these images reside. Underneath these items resides another item that you can't view: the controller's translation table. The translation table maps logical block numbers to physical page numbers. Generally, it is either the images' description (file allocation tables, FAT32/exFAT boot sector, directory entries) or the translation table that has broken, never the images themselves. Pulling a card in mid-write. Dropping the battery while the card is flushing its buffers. Removing the card from a reader while copying files. Allowing a camera to brown out on a cold morning. Any of these will cause the descriptions (structures) to get out of sync with the physical locations on the card. The camera now reads the boot sector and does not recognise it. Therefore, it declares the card corrupted. However, the photographs are intact in their respective pages. If the translation table was destroyed instead, then the card displays 0MB, reports an incorrect capacity, or cannot be seen at all, and this problem should be sent directly to the chip bench and not to the software bench. The next part is important. The worst thing that could happen in reaction to the error would be allowing the camera to format the card. The prompt appears to be offering you the fix for the error. It is actually the reverse. Formatting creates a brand-new blank file system over the top of the damaged structures that need to be rebuilt. Depending upon the camera body, a full format may go even further and overwrite portions of the data area itself. Taking new shots onto this same card afterwards is the other way users end up losing everything. What we mean here by SanDisk SD card repair at this bench involves making an entire sector-level image first, building the file system from that image, and where directory entries no longer exist, using file headers with EXIF to carve images back together so as to restore shot order and original dates. SanDisk corrupted SD card recovery of this nature is relatively common here at our bench, so here is a quick summary: a card labelled "corrupt" normally still contains every single picture. Take it out. Slide the write-protect tab across if present. Get a quote first, then send us your card exactly as it is.
Physically broken cards are taken to the chip bench. Since a microSD has no physical separation (the controller, memory and contacts are moulded together into a single piece), the method for access is via the test points located on the body. We probe those to pull a raw dump and rebuild the dump into files which will open without argument. The determination of success is dependent upon whether the physical break caused any damage to internal tracks, and that is what the free assessment determines instead of guessing. Full size SD cards are more forgiving. The controller and memory are located on a small board as separate components; therefore, it is common for the crack in the shell to not cause any damage to the electronics. Salt air will gradually corrode contacts near coastal areas and harbours, and contact corrosion is often cleanable.
Card work is micro-surgical surgery performed directly on bare flash packages using the rigs below, identical rigs used for handling sticks and single piece devices:
This software talks directly with memory packages and replicates all arithmetic operations that were previously performed by a dead controller. When a stick or card doesn't respond, work starts here again.
The infrared and hot air profiles are tuned such that a memory package can be removed from a scrapped board with its stored charge undisturbed. The temperature control is the entire skill and rushing it will cost you your data.
Many cards store all of the data in one chip (slab) with no individual packages to remove. The test points have to be individually identified against pinouts and manually wired out. This process is slow. In fact, this is the only way to access and read the information contained within these cards.
Until an actual raw dump returns as identifiable folders with their original date stamps, the XOR, ECC, and wear-levelling schemes are unpacked. Once this has been done ordinary arithmetic can be applied stubbornly until they turn into recognisable files.
Tracks torn off a board; pads lifted from a board; plugs sheared from a board; all can be rebuilt under magnification; also, if necessary, ROM chips can be transplanted into a new board if the existing board is beyond repair. Many dead sticks require only this service. However, this type of service does not extend to storage soldered onto a motherboard, as we do not take that type of work on.
Cards/sticks are accessed through a blocked interface. When your original memory leaves here it carries exactly the same data it had when it arrived.
While the tray holds many cards, SanDisk takes up most of the space, and not by a narrow margin. More specifically, the Extreme PRO (SDSDXXD), Extreme (SDSDXVE) and Ultra (SDSDUNC), plus the microSD SDSQUAB and SDSQXCD, account for most of what comes through the door. Most of these cards are still physically fine when they come into the lab. When a card is deemed "corrupted", it generally means that there is a problem with either the file system structure or the controller's translation table above the memory. Therefore, when a card is said to be "corrupted" it is actually typically good news for the owner. What an owner should never do at that point is allow the camera to reformat it. There are some other brands around the SanDisks as well. Lexar's Professional 633x and 1066x both fall short on long bursts and stall. Kingston's Canvas Select Plus falls silent at the controller. Sony TOUGH cards are very durable physically but will eventually fail electronically. It appears that Samsung EVO Select and PRO Plus cards turn up counterfeit about as often as they turn up genuine, and they corrupt wherever writing wraps beyond the true capacity of the media. CFexpress Type B continues to grow as a category, because wildlife and landscape photographers shooting along the coast, wedding shooters covering venues throughout Aberdeenshire and Moray, and survey operators flying in support of the energy industry all use them with essentially no margin. Dashcam and doorbell cards get their own line, as loop recording wears out the flash but does little else to the card. The footage is retrieved by carving whatever blocks survived.
Get a quote first by telephone or use our free online diagnostic. Most of what you see on this bench will then be delivered via tracked, insured post. This is one of the safest ways to move a failing drive and packages handed in at a North-East post office are normally delivered here the next working day.
Still attached to a laptop, desktop, MacBook, iMac, server or CCTV/DVR unit? Remove the hard drive or SSD first. The bare drive travels on its own; when removing disks from a server remove all the disks and label each disk by its bay. This lab does not take machines apart. Any storage that has been soldered onto a motherboard (as in T2 and Apple Silicon Macs and a couple of thin laptops) will not be worked on: if it will not unbolt, we cannot work on it. If you're not sure what type of machine you have, ring us first. Phones and tablets are also not taken on.
↓ Print the shipping & booking-in form (PDF)
Your parcel should be addressed to Edinburgh Data Recovery, approximately two and a half hours (and 130 miles) from Aberdeen via the A90; alternatively it can reach us the next working day by tracked post if this journey time / distance is prohibitive to travel. As soon as your drive has been scheduled in we will contact you.
Still unsure about what should be placed inside of the packaging? Please call us first on 0800 689 0668, or use our free online diagnostic tool to assist you in deciding which items belong inside of the packaging before sealing.
Free diagnosis, one fixed price quoted in writing, and in most cases you only pay if we recover your data. Start online or phone it in.