logo elektroda
logo elektroda
X
logo elektroda
Advanced Search
logo elektroda
3D printer for home or business?
3D printer for home or business? Here are the perfect 3D printers to start with »
ADVERTISEMENT

Can adding RAM mess up the BIOS? A simple laptop upgrade that got complicated

p.kaczmarek2 2187 16

TL;DR

  • A Lenovo B5400 laptop stopped booting after a RAM upgrade from 4 GB to 8 GB modules, even though the SSD and USB boot options were still detected.
  • Standard fixes like swapping SATA drives, changing BIOS legacy/fastboot settings, and clearing CMOS failed, so the fault pointed beyond the memory sticks themselves.
  • A CH341 programmer and a reflash of the Winbond 25Q64FVSIG BIOS chip restored booting; the chip holds 64 megabits, or 8 megabytes.
  • After the BIOS swap, Windows booted immediately and USB installation worked again, but 8+8 GB still blue-screened, so 8+4 GB was the practical RAM setup.
Generated by the language model.
ADVERTISEMENT
This content has been translated flag-pl » flag-en View the original version here
Report a violation of the law
Reply Cool? Ranking DIY | New topic
Notify about new articles
📢 Listen (AI):
  • Laptop internals with RAM modules, a Seagate 2.5-inch drive, and a disconnected SSH 120GB SSD
    An interesting service story today. Can swapping 4 GB RAM for 8 GB damage a laptop so that it can no longer even boot or install the operating system? I have personally found out that this is possible - here I will show what happened and present the solution to the problem. Does your laptop not boot after an upgrade attempt, does not see the USB drive with the system, does not boot the system even though it can see the drive? This is the topic for you.
    BIOS/UEFI information screen showing version details, Intel Pentium 3550M CPU, and 4096 MB installed memory.
    The patient is a Lenovo B5400 laptop. On the one hand it's already quite old, several years old, still from the Windows 8 era, based on a dual-core Intel Pentium 3550M, but on the other hand it's well-preserved, with no scratches, a full set of hardware and a holding battery. It could still be suitable for social media, browsing the internet and making payments.
    The only problem is that lukewarm SSHD drive and the measly 4 GB of RAM. Especially as that 4 GB is in the form of a single bone, and the replacement SSD I already have in stock from liquidation....
    BIOS boot menu showing ATAPI CD, ATA HDD0, and PCI LAN options
    I started with this RAM. Something tempted me to try with two 8GB bones each. I guess that's silly, as RAM prices are knocking down, but everyone has different hobbies. I matched the two matching dice. Of course I disconnected the battery before running. With the 2x8 GB configuration the laptop wouldn't get up, and after running 4 + 8 GB there was an unusual problem. The laptop stopped booting Windows even though the drive was seen. I checked the standard stuff:
    - second SATA media, second SATA port (I have a drive bay for the CD/DVD slot)
    - booting from USB (Linux)
    - bIOS settings, legacy, fastboot
    - cMOS reset (battery)
    Nothing wanted to work, so I decided to revert to the initial RAM configuration (the memory that came from the factory) and then I found that the problem still remains and booting does not work despite repeating the above described operations.
    Open laptop RAM bay with SO-DIMM slots and a DDR3L 4GB module lying below
    A quick web search quickly made me realise that I'm not the only one having this problem on this particular model of laptop:
    Screenshot of a Reddit post about a Lenovo B5400 that cannot boot after a RAM replacement.
    It's exactly the same as mine! The laptop, colloquially speaking, has been "stoned" and now needs to be rescued. I decided on the "atomic" option, which is to upload a new BIOS with the CH341 programmer. This requires getting into the motherboard, i.e. you have to take the whole laptop apart.
    Laptop motherboard with fan, copper heatsink, and RAM slots on a wooden work surface
    The BIOS is in a tiny flash memory bone, here Winbond 25Q64FVSIG. the 64 in the name indicates the size - 64 megabits, or 8 megabytes. It needs to be programmed, but first also make a copy of it .
    Close-up of a motherboard section with a Winbond 25Q64FVSIG chip and surrounding SMD components
    Such a memory is supposedly programmable on the chip, but in my experience this doesn't always work well, so I soldered the whole thing out with hot air and programmed it on a board-transfer from CH341. Of course the uploaded BIOS has to be preferably under the same version and revision of the motherboard as we have, plus it also contains the keys/OEM, so you either have to patch them yourself afterwards or resign yourself to losing them.... i was able to find a fit on the web for the upload candidate. Internet forums are a treasure mine though.
    CH341 USB programmer with SOP8 clip and adapter board on a workbench
    The laptop has started! The BIOS shows an older version. You can see the differences:
    Lenovo UEFI BIOS screen showing BIOS version, Intel Pentium 3550M, and 4096 MB installed memory
    Windows from the old drive boots immediately, so it is however it was indeed the BIOS that was at fault:
    Lenovo laptop with partially opened interior; Windows logo on screen with “Przygotowywanie urządzeń” text.
    However, the 8+8 version gives a blue screen of death, so in the end I opted for 8+4. 8+4 works in both combinations, so it's not like one of the new bones is faulty either.
    Windows blue screen error with restart message and QR code shown on a Lenovo laptop screen.
    The 4+8 version works fine. Also booting from USB now works. You can install your preferred operating system. There are not the problems that appeared before.

    In summary , the standard procedure for adding RAM to the laptop ended with an unexpected problem. The BIOS had to be reloaded. In my case this went without too much trouble, but I realise that not every person adding a RAM die has the equipment and skills to solder, so it could be an unpleasant surprise for someone else. Anyway, even a quick web search shows that the same thing has happened to other people - and on the same laptop.
    Now the laptop works and performs well as a station for browsing the internet, receiving emails and reading social media. It also does well with document editing and a printer.
    The process shown worked, although I wouldn't say it was fully complete as I didn't patch the new BIOS, the ME region was also left unchanged, but I have a backup of the original batch in case I need to. I may try to address these issues in a separate topic. Here the laptop has passed the tests and is running stable.
    Have you also already had such a "surprise" with the standard procedure for adding memory?

    Cool? Ranking DIY
    Helpful post? Buy me a coffee.
    About Author
    p.kaczmarek2
    Moderator Smart Home  
    Offline 
    p.kaczmarek2 wrote 14332 posts with rating 12234, helped 648 times. Been with us since 2014 year.
  • ADVERTISEMENT
  • #2 21867848
    p.kaczmarek2
    Moderator Smart Home  
    This surprised me too, as I've been swapping out some of these memories, and hadn't come across this before. So much good that flashing the BIOS is nothing new to me, so I was able to salvage quickly. But what would a person who doesn't know how to solder do? Especially if she doesn't own the laptop....
    Helpful post? Buy me a coffee.
  • #3 21867854
    E8600
    Level 41  
    Here quite old hardware so bios without various security features like serial number, Windows key and other things that need to be transferred from old bios to new clean bios otherwise hardware won't boot. Observing repairs of modern laptops, where a significant part of laptop repairs consists of the secret art of crafting and winning the bios, one is discouraged from buying. In the past, you had to make an effort to damage the bios, but today bios can fail without user intervention.

    Nowadays it is no longer enough to be able to solder and use a programmer.
  • #4 21867877
    Mateusz_konstruktor
    Level 37  
    Let's take a closer look at the state of the connector of this RAM module (tick marks mine):
    Close-up of a SO-DIMM RAM module edge connector with gold contacts and a center notch.
  • #5 21867881
    p.kaczmarek2
    Moderator Smart Home  
    That's its own way, except that it's the old 4 GB RAM module what was there, not the new one, so that neither explains the "brick" nor the blue screens with the 8+8 combination.
    Helpful post? Buy me a coffee.
  • ADVERTISEMENT
  • #6 21867891
    oscil1
    Level 23  
    And the "FN+R" key combination didn't do anything - because that's what it's there for.

    p.kaczmarek2 wrote:
    The 8+8 version, however, gives a blue screen of death, so in the end I opted for 8+4. 8+4 works in both combinations, so it's not like one of the new bones is defective either.


    That's not true - it's just that the faulty part may now be where Windows stores nothing, and that's why it hasn't died.
  • ADVERTISEMENT
  • #7 21867898
    p.kaczmarek2
    Moderator Smart Home  
    Fn+R didn't work for me, by the way, you have a screenshot from Reddit (first post) about that too. I was surprised myself, and at first I already thought I had damaged something with electrostatics (first time ever), but a quick web search showed that I wasn't the only one who had this situation.

    And if you have doubts about the RAM itself, the most reliable way to do it is with a memtest, although I tend not to run it on new ones from the shop.
    Helpful post? Buy me a coffee.
  • #8 21867918
    gregor124
    Level 28  
    p.kaczmarek2 wrote:
    Today's interesting service story. Can swapping 4GB of RAM for 8GB damage a laptop so that it can no longer even boot or install the operating system? I have personally found out that it is possible

    No, it can't, as long as it's functional of course, a bug in the BIOS software can do it, for example.
    That's why I always recommend reading what the manufacturer has to say before making any changes, and certainly at least checking for BIOS updates.

    And what we have here. BIOS version 2.08 is available on the manufacturer's website as of 2018, which, among other things, introduces the fix
    VU#631788.
    In order not to bore you, I will write that this patch solves a problem with memory management by System Management Mode, this is a special mode used by, among other things, BIOS software, UEFI, etc.
    And the problem was that in some cases it could try to call routines in an area outside the physical RAM.
    Presumably if the latest version of the BIOS chip software, i.e. j7et63ww, 2.08, had been uploaded earlier, the problem would not have occurred.
    And, to prevent similar problems in the future, the latest available version, i.e. j7et63ww, 2.08, would have been uploaded

    https://pcsupport.lenovo.com/gb/en/products/l...-bit-64-bit-lenovo-b5400?category=BIOS%2FUEFI
    Helpful post? Buy me a coffee.
  • #9 21867930
    MichalMwp
    Level 11  
    The BIOS on laptops is modified every time the board is booted up so if the RAM doesn't work stably, you can crash the laptop even by firing up memtest on a flash drive.
    p.kaczmarek2 wrote:
    I was able to find a suitable upload candidate on the web

    This is where I would advise against this approach. As my colleague E8600 pointed out earlier, on new laptops this won't work. There's a lot of security, such as me-region (that's one of its primary functions), DMI, configurations in NVRAM, different bios for the same boards (HP and Acer excel at this). In addition, you can get into trouble and cause e.g. an End Point Security Error (in HP) which will write itself to KBC, or upload a bios with a Windows Autopilot licence, or upload a bios with an unconfigured me-region which will block the PCH or hybrid CPU permanently. On new boards, there are usually several memory cubes, plus a program in the KBC and signatures in the TPM. Really, it's not easy.
  • #10 21868080
    Mateusz_konstruktor
    Level 37  
    p.kaczmarek2 wrote:
    This is by the way, except that this is the old 4 GB RAM module that was there, not the new one, so this neither explains the "brick" nor the blue screens with the 8+8 combination.

    A. Looking at this photo, I thought of the laptop getting wet from underneath and expected a similar condition of the memory module slot.
    What is it like in reality?

    B. Isn't there the possibility of some sort of error when trying to operate the RAM modules in dual-mode?
  • ADVERTISEMENT
  • #11 21868609
    MARCIN.SLASK
    Home appliances specialist  
    Isn't there by any chance a burn at the bridge itself ( from the BIOS bone upwards in the photo)?
  • #12 21868680
    Dydelmax
    Level 39  
    @MARCIN.SLASK This is not burn-in - these are traces of the black glue used on BGA chips by Lenovo.
    The contacts on the RAM dice are indeed very dirty. I always clean the contacts with isopropyl alcohol in such cases, although some people recommend for this purpose.... a pencil eraser. :)
    A suitably crafted batch from the manufacturer's website + a data block from the original batch - this is what a BIOS repair should look like. Uploading random batches from the net, without verification of e.g. ME Region version may permanently "ugly" the hardware. It is not so bad if the PCH is a separate chip. Worse if the locked PCH is part of the CPU...
    Here, BIOS version 2.06 has been replaced by version 1.04, i.e. a downgrade has taken place.
    With this type of hardware it is worth replacing the CR2016/25/32 battery with a new one. After so many years this should be a normal maintenance activity.
    p.kaczmarek2 wrote:
    Have you also had such a "surprise" with the standard procedure for adding memory?

    There have been all sorts of strange situations with seemingly trivial and simple activities. The level of abstraction of the situation grew in line with the age of the expanded equipment. If one were to dig so well into ... memory :D there would be enough material for a sloppy book.

    EDITION (referring to the post below):
    I dedicate the following to people who are surpassed in their ability to read with understanding and reveal displays of hypocrisy: https://sjp.pwn.pl/poradnia/haslo/Co-robimy-gumka;21487.html
  • #13 21869117
    LEDówki
    Level 43  
    Eraser or eraser? Are you going to explain this riddle, or does everyone have to google or guess?
    I've had RAM failures when replacing modules on 2 occasions. It was as if I had not pressed the module into the socket. I always check the correct operation of the memory with memtest - it will run through the entire ram and there will be no surprises. Once I found a damaged module, but the owner put it in, because there is not enough RAM. The result - after some time the memory socket and the RAM board burnt out. I would never have expected BIOS damage without trying to update. It is good to know that such faults also happen.
  • #14 21869655
    Frog_Qmak
    Level 25  
    MichalMwp wrote:
    BIOS in laptops is modified every time the board is booted so if the RAM does not perform stably, you can crash the laptop even by firing up memtest on a flash drive.


    Can you write something more about this? \
  • #15 21873652
    Edward321
    Level 26  
    I was adding RAM and changed the HDD to an SSD in a hp laptop and it stopped booting. The light when I pressed power went out and it wouldn't turn on. After several attempts to turn it on and combining, I left it and it stood up after 2 days, it still works today. I still don't know what the problem was, cold solders? The second "new" RAM die is identical to the old one. Memtest ok. Apparently I am not the only one, because I also had such a surprise.
  • #16 21874195
    oscil1
    Level 23  
    >>21873652 You probably did not have the patience and it deleted the nvram. When it has stopped, it has already deleted itself.
Reply Cool? Ranking DIY | New topic
Notify about new articles
📢 Listen (AI):
Report a violation of the law
Today's popular

FAQ

TL;DR: A Lenovo B5400 that moved from 4 GB RAM to 2x8 GB can lose boot ability when a BIOS bug is triggered; one expert summary in the thread was: "the BIOS was at fault." This FAQ is for laptop upgraders who face a system that sees drives but will not boot Windows or USB installers after a RAM change. [#21867776]

Why it matters: A simple memory upgrade can turn into a no-boot BIOS recovery job if an older laptop firmware reacts badly to certain RAM configurations.

Configuration / method Observed result on this Lenovo B5400 Practical takeaway
Factory 4 GB Booted originally, then still failed after the bad upgrade event Returning old RAM alone did not clear the fault
2x8 GB Laptop would not start; later produced blue screens Highest-risk configuration in this thread
8+4 GB Booted normally after BIOS reflash Mixed capacity worked better than 16 GB total
Clean BIOS from web dump Restored booting, but lost original BIOS state Useful for recovery, risky without board-specific data

Key insight: The RAM swap did not physically "damage the BIOS chip." A BIOS software or compatibility fault was triggered by the new memory setup, and recovery required reflashing the Winbond SPI flash with a programmer.

Quick Facts

  • The affected laptop was a Lenovo B5400 from the Windows 8 era with an Intel Pentium 3550M, factory 4 GB RAM, and a Winbond 25Q64FVSIG BIOS chip. [#21867776]
  • The BIOS flash chip size was stated directly: 64 megabits, which equals 8 megabytes of firmware storage. [#21867776]
  • After recovery, the machine booted again, but the BIOS version shown changed from 2.06 to 1.04, which means the repair used an older dump. [#21868680]
  • User gregor124 notes that BIOS version 2.08 from 2018 adds fix VU#631788, described there as addressing a System Management Mode memory-management issue. [#21867918]
  • On this unit, 8+8 GB caused blue screens, while 8+4 GB worked in both slot combinations after the BIOS reflash. [#21867776]

Why can a Lenovo B5400 stop booting Windows or USB installers after a RAM upgrade from 4 GB to 8 GB or 2x8 GB?

It can stop booting because the RAM change can trigger a BIOS or firmware fault on this model, not because RAM directly damages the board. In the thread, the Lenovo B5400 still saw the drive, but it would not boot Windows or a Linux USB installer after a failed 2x8 GB attempt, and the problem remained even after the original 4 GB module was reinstalled. User gregor124 points to BIOS version 2.08 and fix VU#631788 as evidence of a firmware-side memory-management issue. [#21867918]

How do you recover a Lenovo B5400 with a corrupted BIOS after a RAM swap using a CH341 programmer and a Winbond 25Q64FVSIG chip?

You recover it by reflashing the BIOS flash chip with a known-good image. 1. Disassemble the Lenovo B5400 and locate the Winbond 25Q64FVSIG SPI chip. 2. Read and save a backup first, then program a matching replacement dump with a CH341 programmer. 3. Resolder the chip, reassemble, and test booting with a stable RAM setup such as 8+4 GB. The author desoldered the chip before programming because in-circuit flashing was less reliable in his experience. [#21867776]

What causes blue screens on a Lenovo B5400 with an 8+8 GB RAM configuration when 8+4 GB works normally?

The most likely causes in this thread are a marginal RAM module, a compatibility issue at 16 GB, or a firmware problem exposed only by the 8+8 GB layout. The author reports that 8+8 GB produced blue screens, while 8+4 GB worked in both slot orders after the BIOS repair. Another user argued that a faulty area of RAM may simply go unused in one layout, so mixed capacity can appear stable while the full dual-stick setup still fails. [#21867891]

How should you verify whether a new RAM module is actually faulty before blaming the BIOS, for example with MemTest?

You should test the RAM first, because unstable memory can mimic a BIOS fault. Run a memory test such as MemTest with each stick alone, then test both sticks together in the target configuration. The thread explicitly recommends memtest as the most reliable check, and another user says he always tests replacement memory this way because one bad module once led to a burnt socket and RAM board. [#21869117]

What is the Fn+R recovery key combination on Lenovo laptops, and why might it fail to restore booting after a bad RAM upgrade?

Fn+R is a Lenovo emergency BIOS recovery shortcut, but it may fail if the corruption state or hardware condition blocks the recovery path. In this case, the user tried Fn+R and it did nothing, even though another post says that is what the key combination is for. The same thread also mentions a similar no-recovery case shown in a Reddit screenshot, so this shortcut was not enough once the laptop entered its no-boot state. [#21867898]

Which BIOS version for the Lenovo B5400 addresses the memory-management issue mentioned as VU#631788, and how does that fix help?

BIOS version 2.08, identified as j7et63ww from 2018, is the version named in the thread. User gregor124 says it includes fix VU#631788, which addresses a System Management Mode memory-management issue where firmware routines could access memory outside physical RAM. That matters here because a RAM upgrade can expose exactly this class of address-handling bug during early boot or installer startup. [#21867918]

What is the ME region in a laptop BIOS, and why can using a random BIOS dump from the internet be risky?

"ME region" is a firmware area that stores Intel Management Engine code and platform settings, a key characteristic being that bad or mismatched data can break board startup or lock hardware. Using a random BIOS dump is risky because it may carry the wrong ME region, wrong board revision data, or other foreign settings. The thread warns that an unconfigured or mismatched ME region can block the PCH or even permanently damage recovery options on newer machines. [#21867930]

What is NVRAM in a laptop BIOS, and what kinds of board-specific data can be lost when flashing a clean BIOS image?

"NVRAM" is non-volatile firmware storage that keeps board-specific settings across power loss, a key characteristic being that it can hold machine identity and configuration values that a clean image does not restore automatically. The thread says a clean BIOS can lose OEM keys and other board data, and newer laptops may also depend on DMI, stored configuration, and security-linked settings. That is why a raw internet dump can boot a machine yet still leave hidden problems behind. [#21867930]

How do BIOS repair methods differ between older laptops like the Lenovo B5400 and newer models with TPM, DMI, Autopilot, and KBC security data?

Older laptops are easier to revive because they rely on fewer tied security records. In this thread, users say the Lenovo B5400 could boot from a reflashed dump even after a downgrade from BIOS 2.06 to 1.04. They contrast that with newer models, where TPM data, DMI records, Windows Autopilot information, KBC code, and endpoint security states can make a generic reflash fail or create new lockouts. [#21868680]

CH341 in-circuit programming vs desoldering the BIOS chip first: which approach is safer for repairing a Lenovo laptop BIOS?

Desoldering first was treated as the safer method in this repair. The author says the Winbond 25Q64FVSIG can supposedly be programmed on-board, but that this does not always work well in practice, so he removed the chip with hot air and programmed it on an adapter for the CH341. That approach adds soldering work, but it reduces read or write errors caused by the rest of the board loading the SPI lines. [#21867776]

What steps should you take before upgrading laptop RAM to avoid boot problems, BIOS corruption symptoms, or compatibility issues?

You should prepare the firmware and the memory path before inserting new RAM. 1. Check the vendor BIOS version and update first if a newer release exists, such as 2.08 for the B5400. 2. Power down fully, disconnect the battery, and inspect or clean RAM contacts and slots. 3. After installation, boot-test each stick separately and run memtest before assuming the upgrade is stable. The thread shows that skipping firmware checks can leave an old BIOS exposed to known memory-handling bugs. [#21867918]

How can dirty or damaged RAM contacts and slots affect boot stability, and what is the safest way to clean them?

Dirty or oxidized contacts can cause bad seating, intermittent reads, and false no-boot symptoms. In the thread, one user points out visibly dirty RAM contacts, and another says he routinely cleans such contacts with isopropyl alcohol, while some people use a pencil eraser. Isopropyl alcohol is the safer choice mentioned there because it cleans residue without adding abrasive wear to the contact plating. [#21868680]

What could explain a laptop failing after a memory upgrade if the RAM slot area has moisture damage, corrosion, or contamination?

Moisture damage or contamination around the slot can create unstable contact resistance or leakage paths that look like RAM incompatibility. One participant specifically suspected the photo resembled a laptop wet from underneath and asked whether the slot area had similar damage. That matters because corrosion in the slot can break training or detection long before the RAM itself proves defective in a standalone test. [#21868080]

How does dual-channel or mixed-capacity RAM operation like 4+8 GB versus 8+8 GB affect stability on older Intel Pentium laptop platforms?

Mixed-capacity RAM can remain stable even when matched-capacity RAM fails, because the controller may map memory differently. On this Lenovo B5400 with an Intel Pentium 3550M, the author reports that 8+4 GB worked in both combinations after reflashing, while 8+8 GB still caused blue screens. The thread does not prove full dual-channel behavior, but it clearly shows that the 12 GB layout was more stable on this older platform than the 16 GB layout. [#21867776]

What maintenance checks are worth doing on an older laptop during a RAM or SSD upgrade, such as replacing the CMOS battery or updating the BIOS first?

Update the BIOS first, replace aging support parts, and verify storage and memory after reassembly. One user says older hardware like this should get a fresh CR2016, CR2025, or CR2032 CMOS battery as routine maintenance. The same thread also recommends checking the manufacturer BIOS release before upgrades, because the B5400 had a later 2.08 firmware that was said to address a memory-management defect. [#21868680]
Generated by the language model.
ADVERTISEMENT