[Guide] Using CH341A-based programmer to flash SPI EEPROM

"SKU hack" is the expression used in CoffeeTime 0.7p. I do not know what the difference is, but upto 0.7hf CoffeeTime used "Z370 mod" and if one does open an image with "SKU hack" applied in 0.7hf, it is reported as "Z370 mod: bad", so something different is applied. Perhaps @svarmod can explain this and the "region lock" reports as well.

CPU used is a 9900K which does not work without SKU mod, but does work with the dump provided above. So the image is fine, only CoffeeTime reports are unclear.

OK, yes, I guess he will have to explain why it says bad and what SKU Hack means (Mod SKU, or not, FD changed or not, etc)
Yes, your dump has FD unlocked, I did not check what SKU ME or what FD changes were made in that regard either, I assume as you mentioned SKU was changed and FD edited and flashed in per your dump since that CPU is working.
So yes, mod: bad is what we need to know why it says that, maybe itā€™s simply due to the EC/PDR still being locked from Read Access in the FD?

@Lost_N_BIOS , @PCGH_Torsten

SKU-hack is maximum simplified analogue of SKU changing mod. Instead of many changes in FD and ME regions, the SKU-hack fixes 1 byte in the FD (307h offset), which still allows to run Coffee CPU. SKU will kept as default and ME at their factory state. So with SKU-hack, if your board already has compatible ME you can flash only FD and BIOS regions - mod will works.
Less changes - more stable. :wink:
This optimization is not absolutly necessary, thatā€™s why Iā€™m no hurry to release version of CT implemented this.

You right about regions unlocks - it checks all regions, include EC and PDR.

There are a lot of mods for testing and test results for systemizing at the moment. I hope new build will be released soon.

@svarmod - thanks for info on your tool terminology. Why apply the FD change, if SKU not modified at all? In those cases user does not need to re-write FD, since no SKU change, so you could also not change FD and let the tool tell user programmer re-write not needed no SKU/FD change made.
Do you mean, this FD change is required no matter what, even without SKU change? If so, how to some people get around it, and not have to re-write FD and CPU still work as expected, without SKU Change (And sometimes with too I think)?

Also, thanks for answer on the region lock check, that explains my thoughts about why. Good to know you are working on new releases.
Can you please make some way to tell user at the end, depending on BIOS/changes made, they either need pinmod or use programmer to rewrite all BIOS regions, or they do not need pinmod or programmer when itā€™s not required.
I think that will help a lot of users know right away the issue theyā€™re having or know in advance they need to buy a programmer.

@Lost_N_BIOS
Changes in FD need for start U0/P0 coffee cpu anyway. SKU-hack does all Z370 SKUā€™s work we need, but not affect on ME data.

Info about required flashing methods is already writes at finalizing mod. Will try make it a bit clear and add more info about changes.

Thanks @svarmod - I was not sure which or if all needed the FD edit. Good to hear you will add info on changes and if needed for user to rewrite FD etc to next version

Hi everyone,

After reading all posts and managed to get my board running again, I need to make a small contribution.
First of all, Iā€™m very well impressed of how much information is displayed on this discussion. Just about 20 pages of very useful data.

So let me share my experience:

- Gigabyte GA-H61M-S1 board got bricked after changing HD settings on CMOS setup. After boot, Windows 10 showed a blue screen and after that, the board wasnā€™t able to boot again.
- No network lights, no keyboard light, no video
- CPU fan ok. If memory taken off, there were some beeps.
- BIOS chip is MX25L3206E.
- Order CH341A (black edition) + clip, as I initially was trying to recover the BIOS, and was hopping the chip was ok.
- Tried CH341 Programmer Software v1.13, 1.18, 1.29, 1.30, 1.34; ASProgramer v1.4. Anyone was able to identify the chip. Tried rotate the chip 180Ā°, due to the different position/version of the software, without success. In some cases, the CH341A got very hot fast, showing of course wrong connections. Tried also to power the motherboard. No luck.
- The only software able to identify was Flashrom. But only works if power is supplied to the board. No need to desolder chip from the board.
- I was able to extract the BIOS file first (as board ID and MAC address network might be needed).
- Flashed the BIOS file from Gigabyte website, and have no issues with IDs or MAC address.

Thanks everyone for the great forum.

@Lost_N_BIOS Hello, I have a Gigabyte Z87X-UD3H with BIOS chip of MX25L12873F but I keep getting the error message of ā€œChip Main Memory with the contents are in disagreementā€. I am using Windows 10 64 Bit, ran CH341A 1.34 as Admin, and using the CH341A black programmer. I used the drivers inside the zip of the latest one that was uploaded on 13/12/2018.

EDIT: Managed to solve problem by using this version: https://www.up-4.net/4abbqsap4o71

@rootuser123 - Please see my comments here for MX25L12873F (only 1.31Free/1.4 or ASProgrammer works) - [GUIDE] The Beginners Guide to Using a CH341A SPI Programmer/Flasher (With Pictures!) (2)
I am unsure about Win10, but I know it works only with CH341A 1.31Free/1.4 version software (that is same version for anyone unsure) or ASProgrammer 1.40, but using that on Win10 I canā€™t comment on, only Win7
I see you solved, leaving in case someone else reads later looking for help








Sorry for the late feedback, but my CH341A just arrived today after 2 failed import attempts. ;-(

I managed to make backups (multiple times) of the two chips close to the BIOS Battery.

Iā€™m kinda confused because they are 2 MB and 8 MB in size. But the CAP file from Lenovo is just 4 MB in size. Did I read the wrong chips?

I was expecting the Binary to be exactly as big as the one Lenovo makes available for download.

Iā€™m confused. I was prepared, read alot and watched many youtube videos while waiting for the programmer to arrive. I was thrilled that the chips were recognized correctly. But I did not expect the binary readouts to be different in size than the downloaded BIOS. What schould I do now? Is the BIOS on a different Chip of the Board?

I opend the files in a Hex editor and the 2MB file seems to be the Recovery BIOS, as it contaiones some Text hinting at that.

The 8 MB file seems to be the Intel AMT BIOS, as I could find some references in that file.

Please advice what I can do. WHich chip does contain the 4 MB Bios that I need to flash in order to unbrick the board?

The BIOS is from a IS7XM M92p Board with a Q77 Chip.

I replied in the new thread you created @Spam00r

Hello Guys,
i am new to this great forumā€¦
i am trying to flash bios my Motherboard Gigabyte GA-B85M-D3H, 8 PIN CHIP on board = mxic 25l6473e
i am purchase CH341A black edition & clipper & 8 Pin chip 150ML
when i am start to detect from clipper its undetected i am trying many of versions V1.17, V1.18, V1.30 all is not detected too much tries but still not successful :frowning:

please guide me anyone can help Appreciate :slight_smile:

& one thing more i am downloading bios file from gigabyte official site but its not .bin file how to get .bin file downloaded file extension is B85MD3H.F15

https://www.gigabyte.com/Motherboard/GA-ā€¦support-dl-bios


Waiting for your helpful reply guys

Thank you

Hello rlsturk,

welcome in this forum!

First: Is the yellow jumper set to [1 2] near middle of the flasher (look backside)?
2nd: Be aware the 0 from the clipper is @ the side with the * point on the chip.
3rd: Is the other side connected to the 8 contacts of the 25 SPI-side near middle of the flasher?
4rth: If that not work, draw the clipper inside the 25-SPI-mounting by 180Ā°.

Thereā€™s a guide with pictures here in the forum for using CH341A.

Best regards, MiMo

P.S.

For MXIC MX25L6473E you could use CH341A programmer v1.34 and manually choose Type, Manu(facturer), Name.
The CH341A programmer software works best on WinXP, Vista or Win7. In Win10 thereā€™ra sometimes trouble.

Btw. the BIOS file B85MD3H.F15 is a binary file, but only without that extension .bin.
The format means BOARDNAME.FW-version.

!!! But plz before you flash a new bios, be aware to have a working copy stored on your harddrive !!!
Not only a copy from the manufacturer, but from your individual device -> because of often the MAC-address & so on is stored there.

And last, but not least: Proof the revision of your mainboard - do you really have the rev. 1.0?

Hi,

Having difficulty trying to backup the bios or read/detect the bios chip using a CH341A usb adapter and ver 1.18 of the software using a soic-8 clip.
The motherboard is a z270-ws with a Winbond 25Q128FVSQ bios chip. When I press the detect button it fails to detect the chip and furthermore pressing read reads in a lot of jiberish and garbage.

I have my clip setup like this:


What am I doing wrong?

Thanks

Note: Tested the clip on a netbox motherboard and the ch341a was able to read the chip but not in this case with the Z270-WS.

Also note: Forgot to take cmos battery out first time.

Thanks

Iā€™m not able to work with 128Mbit chips with software 1.29, verify is never OK. I suppose earlier versions wonā€™t work better. ASprogrammer 1.40 worked for me (I think 1.3.14 did work, too)
https://github.com/nofeletru/UsbAsp-flash/releases

@davidm71 - As @lfb6 mentioned for W25Q128FV use ASProgrammer 1.40, already tested many versions CH341A software here, all fail - Bricked Asus Z170-AR

I read that account and tried 1.40 of Asprogramer but got a file full of 'FFā€™s. Think perhaps in testing with other versions I may have blanked the spi rom. Is this possible?

@davidm71 - you shouldnā€™t have blanked the chip by reading attempts, but itā€™s possible if you accidentally hit auto on any of the attempts, by default ā€œAutoā€ erases first (Erase, blank check, write, verify)

Did the board run at all, even if not did it run somewhat and just not boot? If yes, start it and see, if it still acts 100% the same, then not likely itā€™s blank but something in the loop if off maybe loose connection, try again.
Put CMOS battery back in if you have it out, then if same still, connect power (24+8 or 4) and try again, if same, then yes, maybe you did blank it somehow



I donā€™t think I pressed ā€˜Autoā€™. I might have hit verify but pretty sure I did not press anything else besides ā€˜Readā€™. This board right now does not have a cpu anymore. My plans are to prepare it for a coffelake mod and buy an 8350K or a 9350KF cpu for it. I replaced it with an Apex IX that I got for $60 more or less on sale after rebate for the benefit of overclocking ram. The only problem the Z270 WS board ever had was that it would go on and off before posting after a cold boot so I wanted to rma it anyhow. Was a minor inconviencience.

Anyhow only way to know for sure would be to get my hands on a cheap celeron if I could find one used on ebay I would buy it. But what confuses me about flashing a mod bios onto Asus boards I have seen others short pads on the motherboard for some reason. Need to do more research. Perhaps that has the effect of unlocking certain regions I am guessing.

Thanks

@Lost_N_BIOS

Are you familiar with the function of these pins? I think they are direct access to the spi using a 9 pin cable by Asus. Am I right?