HELP! - Dead Intel VS440FX motherboard with BIOS EEPROM overheating - SOLVED

Discussion about old PC hardware.

Re: HELP! - Dead Intel VS440FX motherboard with BIOS EEPROM overheating

Postby SSTV2 » 2019-7-29 @ 22:55

The pinout is probably going to be that of a 28F002 DIP40, but still, you should reverse engineer the pinout of it just to be totally sure.
User avatar
SSTV2
Member
 
Posts: 342
Joined: 2014-11-19 @ 20:30

Re: HELP! - Dead Intel VS440FX motherboard with BIOS EEPROM overheating

Postby feipoa » 2019-7-30 @ 07:50

I have ordered the TSOP adapters from China today and will order two of the Micron chips and two of the Intel BV chips from utsource. So it will probably be a few weeks before I can provide an update. I'm gambling that the BIOS is the only issue at play.
User avatar
feipoa
l33t++
 
Posts: 6267
Joined: 2011-3-07 @ 13:54
Location: Canada

Re: HELP! - Dead Intel VS440FX motherboard with BIOS EEPROM overheating

Postby feipoa » 2019-8-08 @ 10:51

I extracted the contents of BIOS.EXE for the v18 BIOS for the Intel VS440FX motherboard. Screenshot attached. Which file is the BIOS to be programmed into the EEPROM? Is it,

1018CS1_.BIO
1018CS1_.BI1
1018CS1_.BI2
or
BIOSCS1.REC

Each of them are 65K. 1018CS1_.BI3 is 41K, so I assume this is not it.

And any chance this BIOS can only be programmed using the iflash.exe program, that is, does iflash do some piece-wise selection or sumation of the various files from BIOS.EXE package?

VS440FX_BIOS.jpg


On my Intel Batman's Revenge socket 4 motherboard, when I go to save a BIOS file using the Intel BIOS utility, it saves two separate files:
1010AF2T.BIO at 64.1 KB
and
1010AF2T.BI1 at 48.1 KB

When I use NSSI to make a copy of the BIOS, the file size is 64.0 KB. But when I use the Intel BIOS utility (FMUP.exe) to verify that the NSSI BIOS matches that of what is currently programmed in my EEPROM on this board, I receive an error "The loaded file is not a valid Flash Data File.

Why is that? Is anyone willing to use a program like NSSI or preferably AMI Flash (version that works w/PPRO?) to create a single BIOS file for the Intel VS440FX? Are there programs besides NSSI which will also write the EEPROM chip while in board? It uses an American Megatrends BIOS, perhaps some version of their software?
User avatar
feipoa
l33t++
 
Posts: 6267
Joined: 2011-3-07 @ 13:54
Location: Canada

Re: HELP! - Dead Intel VS440FX motherboard with BIOS EEPROM overheating

Postby SSTV2 » 2019-8-08 @ 21:05

Oh boy, the faulty BIOS chip is nothing, compared with the obscure BIOS updating practices of Intel. I had downloaded the same BIOS update utility and inspected main files with a HEX editor.

feipoa wrote:Which file is the BIOS to be programmed into the EEPROM? Is it,

1018CS1_.BIO
1018CS1_.BI1
1018CS1_.BI2
or
BIOSCS1.REC

None of the above, the BIOS image is divided into 4 parts, starting with *.BIO and ending with *.BI3. Each part has a 128 byte long header, which can be read with "SHOWHDR.EXE" program.

feipoa wrote:And any chance this BIOS can only be programmed using the iflash.exe program, that is, does iflash do some piece-wise selection or sumation of the various files from BIOS.EXE package?

Definitely, at least as long as boot block is intact and not corrupted, you could just insert the floppy disk with recovery files and after boot-up, BIOS recovery would then start automatically (there usually is a BIOS recovery JMP/SW somewhere on a Intel MB for that matter).

feipoa wrote:On my Intel Batman's Revenge socket 4 motherboard, when I go to save a BIOS file using the Intel BIOS utility, it saves two separate files:
1010AF2T.BIO at 64.1 KB
and
1010AF2T.BI1 at 48.1 KB

When I use NSSI to make a copy of the BIOS, the file size is 64.0 KB. But when I use the Intel BIOS utility (FMUP.exe) to verify that the NSSI BIOS matches that of what is currently programmed in my EEPROM on this board, I receive an error "The loaded file is not a valid Flash Data File.

Why is that?

64.0KB file definitely did not have a header when it was written with the help of NSSI, that's why "FMUP" could not recognise it. It's also possible that the whole BIOS image is compressed, because there is basically no legible ASCII text when *.BIx images are read with a HEX editor (in 440FX MB BIOS case). If that's the case, then you'd be comparing some random 64K data of decompressed BIOS image with an original 64.1K image, besides, shouldn't Intel S4 MB have a 128KB BIOS image?

According to "SHOWHDR.EXE" the lenght of all *.BIx files is 3A000 HEX, which translates to 237568 DEC (232KB), 24KB is missing to fill the whole chip, so I guess boot block is not included into *.BIx files, but... "1018CS1_.RCV" file is exactly 16KB long, which probably means that it's a boot block image. Not sure where the rest of 8KB should be, maybe BIOS update utility automatically fills those missing bytes with FF's on update?

Either way, if 16KB file is actually a boot block image, then you are saved :)
User avatar
SSTV2
Member
 
Posts: 342
Joined: 2014-11-19 @ 20:30

Re: HELP! - Dead Intel VS440FX motherboard with BIOS EEPROM overheating

Postby feipoa » 2019-8-08 @ 22:02

Wow, thanks a lot for looking into these files. I'm not familiar enough to know what to do without doing research.

BIOS Recovery? You are referring to this:

1.13.5 BIOS Recovery (J6D1)
This jumper lets you recover the BIOS data from a diskette in the event of a catastrophic failure. The default setting is pins 4-6 (normal operation). To recover the BIOS, turn off the computer, move the jumper to pins 2-4, then turn on the computer to perform BIOS recovery. After recovery, turn off the computer and return the jumper to pins 4-6 to restore normal operation. Refer to Section 3.14 for more details.


But using this method assumes that the existing BIOS chip is still usable and that just the EEPROM data went bad and needs to be preprogrammed.

If this is not the case, then I would need to re-program off motherboard the EEPROM chips that I have ordered. Is this even possible in an external programmer? I've already ordered the TSOP adapter. From what you have written, I am I to use the suspected 1018CS1_.RCV boot block file with my external programmer and write this to the new EEPROM out of motherboard? Is there a means to tell the programmer where this boot block data should be stored? And assuming I write this information in the appropriate location, then I need to solder the EEPROM back onto the motherboard, set the recovery jumper, and the system will glue together the pieces of BIOS from the diskette?

Or is there also a means to get these files properly programmed to the EEPROM out of motherboard?

EDIT: showhdr.exe wasn't in my package. It was in yours?

EDIT2: And do you think if someone else who owns this board made a BIOS image, perhaps using an appropriate version of AMI Flash, that I would be able to use the external programmer to program this BIOS and have it work?
Attachments
Intel_VS440FX_BIOS_18.00_8-14-2000.exe
(374.44 KiB) Downloaded 3 times
User avatar
feipoa
l33t++
 
Posts: 6267
Joined: 2011-3-07 @ 13:54
Location: Canada

Re: HELP! - Dead Intel VS440FX motherboard with BIOS EEPROM overheating

Postby SSTV2 » 2019-8-08 @ 23:11

feipoa wrote:BIOS Recovery? You are referring to this:

1.13.5 BIOS Recovery (J6D1)
This jumper lets you recover the BIOS data from a diskette in the event of a catastrophic failure. The default setting is pins 4-6 (normal operation). To recover the BIOS, turn off the computer, move the jumper to pins 2-4, then turn on the computer to perform BIOS recovery. After recovery, turn off the computer and return the jumper to pins 4-6 to restore normal operation. Refer to Section 3.14 for more details.

But using this method assumes that the existing BIOS chip is still usable and that just the EEPROM data went bad and needs to be preprogrammed.

Exactly

feipoa wrote:I am I to use the suspected 1018CS1_.RCV boot block file with my external programmer and write this to the new EEPROM out of motherboard?

Right.

feipoa wrote: Is there a means to tell the programmer where this boot block data should be stored?

Programmer's software should allow you to reallocate loaded image data in a buffer, but if by some reasons it won't have that function, then any HEX editor can be used.

feipoa wrote:And assuming I write this information in the appropriate location, then I need to solder the EEPROM back onto the motherboard, set the recovery jumper, and the system will glue together the pieces of BIOS from the diskette?

That's the whole idea :D

feipoa wrote:Or is there also a means to get these files properly programmed to the EEPROM out of motherboard?

There are 8KB of data missing to form a whole 256KB BIOS image, I wouldn't risk wasting time trying to put images in a right order, It's doable, but such method would be just not reliable enough.

According to the datasheet of a 28F002 FLASH chip, the boot block is located at the very top of address space (3C000 to 3FFFF HEX), this is where you have to put the alleged "1018CS1_.RCV" boot block.

feipoa wrote:showhdr.exe wasn't in my package. It was in yours?

No, I had to download it from elsewhere, but BIOS files attached here are identical to those I have.

feipoa wrote:And do you think if someone else who owns this board made a BIOS image, perhaps using an appropriate version of AMI Flash, that I would be able to use the external programmer to program this BIOS and have it work?

Doubt it, but I know that at least one forumer has this motherboard, maybe he could help you?
User avatar
SSTV2
Member
 
Posts: 342
Joined: 2014-11-19 @ 20:30

Re: HELP! - Dead Intel VS440FX motherboard with BIOS EEPROM overheating

Postby feipoa » 2019-8-09 @ 00:41

Yes, kixs has this board. I sent him a PM earlier about it, but I don't know which version of the AMI flash software would be most appropriate for this oddball BIOS. Do you know? I suspect NSSI may not be sophisticated enough. I started a thread asking if anyone knows of a about an AMI flash repository, but no responses yet.

The EEPROM chips I have on order are E28F002BV at 60ns. And because of logistics, I did end up ordering the TL866II plus, which came with all the adapters pre-packaged. So, the new programmer arrived today. The TL866II plus has specific support for the E28F002BV, but not the *BC for some reason. Hmm... Are you familiar with this programmer? It came with some 25 adapters. I essentially bet the farm on getting this damn board working rather than sourcing a new one. https://www.ebay.com/itm/223265765989 According to the seller, for E28F002BC, I would use TSOP32/40/48 adapter with TSOP40A adapter, then put it into the programmer. The programmer arrived today, but the E28F002BV chips have not. They are from China utsource.net The Micron chips, MT28F002C1, have arrived at my US-shipping service location, but I'm not ready to ship all my junk up to Canada just yet.

I installed the TL866II software, but it doesn't let me use it if it doesn't detect the hardware attached to the USB port.

I've sent TL866II a tech support question about burning these multi-parted BIOSes, but I don't anticipate much help there.

Unfortunately, I didn't know about the recovery option on this motherboard until I had already removed the EEPROM chip. But considering that the EEPROM is heating up to 60 C, is it likely that the recovery method would even work?
User avatar
feipoa
l33t++
 
Posts: 6267
Joined: 2011-3-07 @ 13:54
Location: Canada

Re: HELP! - Dead Intel VS440FX motherboard with BIOS EEPROM overheating

Postby feipoa » 2019-8-09 @ 10:21

Are specific recovery disks needed for the corresponding motherboards? For example, I see one for the socket 4 Batman's Revenge board here, http://files.mpoli.fi/hardware/ROM/OTHER/ , called BATREVRD.EXE. I am guessing I run this and it creates a bootable diskette?

Unfortunately, this site doesn't have one for the Intel VS440FX motherboard. The manual states:

3.14 Recovering BIOS Data
Some types of failure can destroy the BIOS data. For example, the data could be lost if a power outage occurs while your are updating the BIOS in Flash memory. You can recover the BIOS data from a diskette by changing the setting of the BIOS Recover jumper (refer to Section 1.13.5). To recover the BIOS: 1. Turn off the computer and move the jumper to the BIOS recovery setting. 2. In drive A, insert the supplied boot disk containing the BIOS files and Recovery file. 3. Boot the computer. The system should beep once and drive activity should start. When recovery is successful, the system beeps twice and drive activity stops. After recovery, turn off the computer and return the jumper to the original pins to restore normal operation.


Anyone have the VS440FX recovery disk?
User avatar
feipoa
l33t++
 
Posts: 6267
Joined: 2011-3-07 @ 13:54
Location: Canada

Re: HELP! - Dead Intel VS440FX motherboard with BIOS EEPROM overheating

Postby The Serpent Rider » 2019-8-09 @ 11:02

Anyone have the VS440FX recovery disk?

Apparently you can use this method: https://www.intel.com/content/www/us/en ... -kits.html
User avatar
The Serpent Rider
Oldbie
 
Posts: 1488
Joined: 2017-3-25 @ 19:07
Location: Stagnant Demesne

Re: HELP! - Dead Intel VS440FX motherboard with BIOS EEPROM overheating

Postby feipoa » 2019-8-09 @ 11:29

Those instructions were last reviewed in 2017, so I wonder if they are still relevant? The instructions ask that just a formatted diskette be used (needn't be a bootable disk) and the disk needs to contain the *.BIO "recovery file". But what about the other parts of the BIOS?

For the Batman's Revenge Recovery disk, I went through the process of creating the Recovery Disk, which intel has some DOS software that creates the disk with the relevant files on it.
Batmans_Revenge_socket4_recovery_disk_with _single_binary_file.jpg

This is a bootable diskette and contains a single binary BIOS file, OLDBIOS.BIN.

This is odd because if you look at the non-recovery disk, but just the ordinary BIOS update disk, the BIOS is broken into parts. For the Batman's Revenge board, see below:
Batmans_Revenge_socket4_BIOS_files_not_via_recovery_disk.jpg


So the official recovery disk is glueing these BIOS pieces together for us? And if that is the case, would obtain a VS440FX Recovery Disk allow us to write the glued BIOS with the programmer and be done with it?

I tried to use the MK_DISK feature in the VS440FX BIOS update v18. All it does is create a feature limited DOS bootable disk, solely licensed for doing flash updates. I then coppied over the VS440FX BIOS update files and it looks like this:
VS440FX_used_MK_BOOTZ_to_create_boot_disk_then_coppied_over_all_BIOS_update_files.jpg


Also, if this motherboard is a Gateway board, will an Intel BIOS Recovery Disk even work? Or must we procure a Gateway BIOS Recovery Disk?

EDIT: According to this post concerning Intel BIOSes on Gateway boards, you can just extract BIOS.exe to a boot floppy and set the Recovery Mode jumper. viewtopic.php?f=46&t=65168

EDIT: Slight variation in what was noted in the above POST for getting Intel BIOSes onto Gateway boards, this comes from google groups,
Code: Select all
Rodger MacKendrick    
11/12/1998
Yes this Gateway motherboard has been upgraded with success. But you have to use the bios recovery method. I got my instructions from this forum some time ago but here they are. You make a bootable disk BIOS disk as per instructions that come with the BIOS file download. After the disk is finished remove the READ ONLY attribute from the AUTOEXEC.BAT file on the disk. Open it up with Notepad or Edit find the line "iflash" and change it to read "iflash /p <filename>.bio" <filename> should be replaced with the file on the disk with the .BIO extension. shut down your computer. Put your motherboard BIOS jumper in recovery mode insert your disk and powerup the machine. YOU WILL HAVE NO VIDEO OUTPUT!! your machine will beep once then run the floppy for SEVERAL MINUTES. Let it run until you hear two beeps wait an additional two minutes. Power off return the jumper to normal position and restart you will now have the intel POST screen and updated BIOS. Good Luck


I wonder if this would work on my Gateway Batman's Revenge board to get 10.13 running rather than 10.10. But perhaps the above scheme doesn't work on this board because the Batman's Revenge BIOS on the Recovery Disk is not pin 2 parts, rather, it is 1 part. And the 10.13 update is in 2 parts.
User avatar
feipoa
l33t++
 
Posts: 6267
Joined: 2011-3-07 @ 13:54
Location: Canada

Re: HELP! - Dead Intel VS440FX motherboard with BIOS EEPROM overheating

Postby 640K!enough » 2019-8-09 @ 12:15

This is just a guess, but I know of a few instances where it was helpful. Since you appear to know someone who has the same board, it may be worth asking them to provide a BIOS dump created with UniFlash (check the documentation to be sure that the chipset is supported first). Those are usually complete images in a single file, that can be written with a standard programmer; no need to assemble the sub-images in the right order. Once the system is modestly workable, you can use one of the other methods to flash the latest version using Intel's software.

If my memory is correct, there was a specific option to include the recovery/boot block in the image. It may be worth ensuring that that option is used.
640K!enough
Member
 
Posts: 491
Joined: 2017-5-11 @ 17:52
Location: Canada

Re: HELP! - Dead Intel VS440FX motherboard with BIOS EEPROM overheating

Postby Sedrosken » 2019-8-09 @ 13:33

feipoa wrote:
Code: Select all
Rodger MacKendrick    
11/12/1998
Yes this Gateway motherboard has been upgraded with success. But you have to use the bios recovery method. I got my instructions from this forum some time ago but here they are. You make a bootable disk BIOS disk as per instructions that come with the BIOS file download. After the disk is finished remove the READ ONLY attribute from the AUTOEXEC.BAT file on the disk. Open it up with Notepad or Edit find the line "iflash" and change it to read "iflash /p <filename>.bio" <filename> should be replaced with the file on the disk with the .BIO extension. shut down your computer. Put your motherboard BIOS jumper in recovery mode insert your disk and powerup the machine. YOU WILL HAVE NO VIDEO OUTPUT!! your machine will beep once then run the floppy for SEVERAL MINUTES. Let it run until you hear two beeps wait an additional two minutes. Power off return the jumper to normal position and restart you will now have the intel POST screen and updated BIOS. Good Luck


I can vouch for this procedure's success on my (VS440FX, not Batman's Revenge -- see Al-Jalima) board. It was a sphincter-clencher moment for sure, but when I put the jumper back and turned it back on it still roared to life, and didn't have the Gateway logo anymore, though certain utilities like Everest could still spill the beans on this board's cow-spotted roots. As for whether it'll help you... I don't know. I still had a functional BIOS, but it was just out of date and I wanted to see if maybe a newer BIOS would fix the LBA issues that the IDE controller had. If you need someone to make you a complete image, and I can do that in software, let me know and I'll do it.
The following are Pastebin links to see the specs of each named machine.
Arwic, Al-Jalima and Glenden Wood.
User avatar
Sedrosken
Member
 
Posts: 145
Joined: 2013-5-13 @ 13:53
Location: The Sticks

Re: HELP! - Dead Intel VS440FX motherboard with BIOS EEPROM overheating

Postby SSTV2 » 2019-8-09 @ 14:55

feipoa wrote:The TL866II plus has specific support for the E28F002BV, but not the *BC for some reason. Hmm... Are you familiar with this programmer?

I don't have that programmer, unfortunately, but if it doesn't support the BC variant of that FLASH chip, then it simply can't supply 12V to the RP# pin.

feipoa wrote:Unfortunately, I didn't know about the recovery option on this motherboard until I had already removed the EEPROM chip. But considering that the EEPROM is heating up to 60 C, is it likely that the recovery method would even work?

No chance, the FLASH chip is damaged one way or another, but it's not yet known if it suffered a catastrophic failure, it might still be possible to partially read it, so there is a probability that the boot block is still intact (supply 5V to the chip to see whether it heats up). I had success of reading an EPROM that was fried by reversed power supply polarities, it had heated up so much, that plastic sticker on top of it had charred (but in normal operation it remained cool) :)

feipoa wrote:Are specific recovery disks needed for the corresponding motherboards?

I think BIOS update utility for the VS440FX MB already contains all files required for the BIOS recovery.

feipoa wrote:EDIT: According to this post concerning Intel BIOSes on Gateway boards, you can just extract BIOS.exe to a boot floppy and set the Recovery Mode jumper. viewtopic.php?f=46&t=65168

EDIT: Slight variation in what was noted in the above POST for getting Intel BIOSes onto Gateway boards, this comes from google groups,
Code: Select all
Rodger MacKendrick    
11/12/1998
Yes this Gateway motherboard has been upgraded with success. But you have to use the bios recovery method. I got my instructions from this forum some time ago but here they are. You make a bootable disk BIOS disk as per instructions that come with the BIOS file download. After the disk is finished remove the READ ONLY attribute from the AUTOEXEC.BAT file on the disk. Open it up with Notepad or Edit find the line "iflash" and change it to read "iflash /p <filename>.bio" <filename> should be replaced with the file on the disk with the .BIO extension. shut down your computer. Put your motherboard BIOS jumper in recovery mode insert your disk and powerup the machine. YOU WILL HAVE NO VIDEO OUTPUT!! your machine will beep once then run the floppy for SEVERAL MINUTES. Let it run until you hear two beeps wait an additional two minutes. Power off return the jumper to normal position and restart you will now have the intel POST screen and updated BIOS. Good Luck

This is what you will have to do, apparently, these old Intel motherboards don't automatically start BIOS recovery by looking for a file with *.BIO extension, like more recent motherboards would do.

You must have a bootable floppy disk with all the BIOS update files and manually start "IFLASH" utility, then type in the name for the first part of a split BIOS image (*.BIO) or edit the "AUTOEXEC", like it was suggested in the comment above. Type IFLASH /h for all command line options.

My suggestion to you would be to not overthink this whole BIOS recovery process, it's very simple. Your main concern at this moment is getting the boot block written into your new FLASH IC or the whole image if someone will be able to supply it.

I'm attaching the alleged boot block, that is already positioned at the very top of the address space for a 256KB FLASH chip just in case.

BLCK256K.zip
(10.61 KiB) Downloaded 2 times
User avatar
SSTV2
Member
 
Posts: 342
Joined: 2014-11-19 @ 20:30

Re: HELP! - Dead Intel VS440FX motherboard with BIOS EEPROM overheating

Postby feipoa » 2019-8-10 @ 08:15

640K!enough wrote:This is just a guess, but I know of a few instances where it was helpful. Since you appear to know someone who has the same board, it may be worth asking them to provide a BIOS dump created with UniFlash (check the documentation to be sure that the chipset is supported first). Those are usually complete images in a single file, that can be written with a standard programmer; no need to assemble the sub-images in the right order. Once the system is modestly workable, you can use one of the other methods to flash the latest version using Intel's software.

If my memory is correct, there was a specific option to include the recovery/boot block in the image. It may be worth ensuring that that option is used.


I recall trying Uniflash v1.40 (the last version) on my Batman's Revenge socket 4 board to save the BIOS and it returned some kind of error. Looking at the documentation for Uniflash 1.4, it mentions that the 430LX (Mercury - socket 4) and the 440FX (Natoma - socket 8) chipsets are untested. I also tried AMIBIOS8 from AMI and various AMI Flash tools versions, 4.6, 4.65, 5.20, 5.30, 6.20, and 6.34. Neither of them wanted to make a copy of the Batman's Revenge BIOS except for NSSI. I'm not sure if these results can be generalised to include the VS440FX motherboard though. Both boards use this oddball split-up BIOS.

The NSSI documentation doesn't mention if it coppies the bootblock, but looking at the Uniflash documentation, I see this statement,
Since v1.33, flashing is done including bootblock by default. This is less dangerous that flashing a new BIOS and leaving incompatible bootblock from previous BIOS there. And more, the bootblock is not exactly defined anywhere - some BIOSes might use bigger bootblock than other, which can cause problems.



Sedrosken wrote:
feipoa wrote:
Code: Select all
Rodger MacKendrick    
11/12/1998
Yes this Gateway motherboard has been upgraded with success. But you have to use the bios recovery method. I got my instructions from this forum some time ago but here they are. You make a bootable disk BIOS disk as per instructions that come with the BIOS file download. After the disk is finished remove the READ ONLY attribute from the AUTOEXEC.BAT file on the disk. Open it up with Notepad or Edit find the line "iflash" and change it to read "iflash /p <filename>.bio" <filename> should be replaced with the file on the disk with the .BIO extension. shut down your computer. Put your motherboard BIOS jumper in recovery mode insert your disk and powerup the machine. YOU WILL HAVE NO VIDEO OUTPUT!! your machine will beep once then run the floppy for SEVERAL MINUTES. Let it run until you hear two beeps wait an additional two minutes. Power off return the jumper to normal position and restart you will now have the intel POST screen and updated BIOS. Good Luck


I can vouch for this procedure's success on my (VS440FX, not Batman's Revenge -- see Al-Jalima) board. It was a sphincter-clencher moment for sure, but when I put the jumper back and turned it back on it still roared to life, and didn't have the Gateway logo anymore, though certain utilities like Everest could still spill the beans on this board's cow-spotted roots. As for whether it'll help you... I don't know. I still had a functional BIOS, but it was just out of date and I wanted to see if maybe a newer BIOS would fix the LBA issues that the IDE controller had. If you need someone to make you a complete image, and I can do that in software, let me know and I'll do it.

I may or may not try the recovery procedure to get the .13 BIOS on my Batman's Revenge board. It currently has .10. I'm not having the best luck recently, so this might be pushing it. What cautions me is that the recovery disk for this board contains a single part 128K BIOS, while the .13 update is two part.


SSTV2 wrote:
feipoa wrote:The TL866II plus has specific support for the E28F002BV, but not the *BC for some reason. Hmm... Are you familiar with this programmer?

I don't have that programmer, unfortunately, but if it doesn't support the BC variant of that FLASH chip, then it simply can't supply 12V to the RP# pin.

It may simply be the case that the *BC chip was added to the list because nobody got around to it. There are so many EEPROM chips past and present. How would it be possible to support the *BV chip if it cannot provide 12 V to RP#? How else would it program the boot block? Same for its support of the Micron counterpart, MT28F002C1. It also uses RP#. The support list is here: http://www.autoelectric.cn/MiniPro/TL866II_List.txt I don't understand how the TL866IIplus doesn't use an external power supply. It seems to rely on power from the USB cord. My Wellon programmer uses a 12 V wall wart for its 12 V needs. The TL866IIplus would need to have some step-up transformer or other IC regulator to get 12 V from 5 V. The programmer itself is tiny, so makes me wonder a bit. EDIT: Looking again at the datasheets, it seems that E28F002BV can get away with 5 V to erase/program the boot block, while the *BC cannot. But then I see in the TL866II support list, for example, that P28F001BX-T and P28F010 are supported, and don't these EEPROMs require 12 V for Vpp? How does it work around this? [And I think the Micron TSOP-40 I have on order, MT28F002C1 also need 12 V, and it is...EDIT: not supported; the *B1 is supported, which is 5 V or 12 V for Vpp. If I program the *C1, then I need to connect 12 V to Vpp and RP#?]

Intel E28F002BV
Code: Select all
Intel SmartVoltage Technology
5 V or 12 V Program/Erase
5 V program/erase operation has been added. For program and erase operations, 5 V VPP operation eliminates the need for in system voltage converters, while 12 V VPP operation provides faster program and erase for situations where 12 V is available, such as manufacturing or designs where 12 V is in-system. For design simplicity, however, just hook up VCC and VPP to the same 5 V ± 10% source.


Micron MT28F002C1
Code: Select all
RP#
Reset/Power-Down: When LOW, RP# clears the status register, sets the internal state machine (ISM) to the array read mode and places the device in deep power-down mode. All inputs, including CE#, are “Don’t Care,” and all outputs are High-Z. RP# unlocks the boot block and overrides the condition of WP# when at VHH (12V); RP# must be held at VIH during all other modes of operation.

VPP
Write/Erase Supply Voltage: From a WRITE or ERASE CONFIRM until completion of the WRITE or ERASE, VPP must be at VPPH (12V). VPP is “Don’t Care” during all other operations
.

SSTV2 wrote:My suggestion to you would be to not overthink this whole BIOS recovery process, it's very simple. Your main concern at this moment is getting the boot block written into your new FLASH IC or the whole image if someone will be able to supply it.

I'm attaching the alleged boot block, that is already positioned at the very top of the address space for a 256KB FLASH chip just in case.

Alright, I will follow Mr. MacKendrick's advice from the google groups post. User kixs mentioned that he was willing to make a soft copy of the BIOS from his VS440FX board. Unfortunately, he's on holiday at the moment. According to tracking information, the *BV EEPROMs arrived in Canada on the 9th, so I should have them by Monday.

Much appreciated! So you've attached the boot block from the v.18 for the VS440FX BIOS? I would then use the TL866IIplus to program this, solder the EEPROM to the motherboard, then follow the BIOS recovery procedure? Is the ideal order of operation to use the BIOS from kixs, or the boot block recovery approach? Due to soldering and desoldering and small delicates, the method with the highest probability of success would go first.

Here's a screenshot of the TL866IIplus software. Showing instructions of which adapters to use with the E28F002BV chip.
TL866IIplus_TSOP_adater_instructions_for_E28F002BVT.jpg


And the pop-up config page that I receive after trying to import your boot block binary. I left everything at default.
TL866IIplus_loading_your_boot_block_binary.jpg


EDIT: Comment from the primary eBay distributor of TL866IIPlus,
Hello,Sir
E28F002BC or E28F002BV ,It's the same programming voltage, VCC/VPP:5V/12V.... , Ignore "check ID"
Last edited by feipoa on 2019-8-10 @ 20:23, edited 9 times in total.
User avatar
feipoa
l33t++
 
Posts: 6267
Joined: 2011-3-07 @ 13:54
Location: Canada

Re: HELP! - Dead Intel VS440FX motherboard with BIOS EEPROM overheating

Postby feipoa » 2019-8-10 @ 09:37

I was considering again the recovery method for updating the BIOS on my Batman's Revenge board as this board was already on my testbed. It contains BIOS 1.00.10.AF2T, the T being OEM for Gateway. I run FMUP.EXE on the BIOS update disk and get the following error:
Batmans_Revenge_1013_BIOS_Error.jpg


For the VS440FX board, autoexec.bat contains iflash /p 1018CS1_.BIO , which should automatically program the BIOS. As I understand it, this is needed because there won't be any display on the monitor to manually select BIOSes. Now the BIOS update disk for the Batman's Revenge doesn't use iflash.exe, it uses fmup.exe. And when I compare the command flags of iflash to that of fmup, I see:

IFLASH.jpg

FMUP.jpg


/p has a different function in fmup and it doesn't look like there is a program flag for fmup like there is for iflash. My questions are then a) can I use iflash.exe for the older Batman's Revenge board, and if not, then b) since there is no monitor output in recovery mode, would I need to practice the exact sequence of keystrokes with fmup.exe to go through fmup's GUI to get 1013AF2_.BIO programmed once the system has booted in recovery mode? e.g. once floppy activity ceases, <Press Enter>, then <Press down arrow twice, followed by Enter; this gets you into "Update FLASH Memory Area from a File">, then <Press Enter; Update System BIOS>, wait for diskette access, then type in <A:\1013AF2_.BIO>, followed by <Enter>, wait for diskette access to cease, then <Press Enter; Continue with programming> After this point, I'm not sure what to do because in the test drive, I get the above error (first photo shown with red highlight).

Or perhaps simply editing the autoexec.bat file with FMUP.exe A:\1013AF2_.BIO ?

According to the release notes, BIOS v.13 has the following advantages over .10. With the current BIOS and hardware, I am unable to get the RZ1000 PCI IDE chip to work with a CD-ROM drive in either DOS or Windows. Perhaps this has been addressed since v.12, and I am using the Pentium Overdrive 133, so it would be nice to have proper freq. display.

Code: Select all
Release 13.0
o   Support for P5 Overdrive added to setup. “Pentium Overdrive” is now displayed and
   CPU frequency now displays 133MHz.
o   Fixed setup’s CPU frequency display. 66MHz used to display 50MHz.

Release 12.0

o      Changed setup .  On boards that have the RZ1000 chip, the  CMOS option for
    prefetch buffer is grayed out. The Default of this option is now set to disabled.  Only
         on boards that have the RZ1001 will  the user be able to enable the prefetch buffers.
   
o      Added a check to make sure that the prefetch buffers are never enabled on boards that 
        have the RZ1000 chip. On boards that have the RZ1001 chip the prefetch buffers will
        be enabled or disabled based on the CMOS option selected.

Release 11.0

o   Removed Hot-Key Security option since not supported in this revision of the BI
OS
User avatar
feipoa
l33t++
 
Posts: 6267
Joined: 2011-3-07 @ 13:54
Location: Canada

Re: HELP! - Dead Intel VS440FX motherboard with BIOS EEPROM overheating

Postby SSTV2 » 2019-8-11 @ 00:10

feipoa wrote:It may simply be the case that the *BC chip was added to the list because nobody got around to it. There are so many EEPROM chips past and present. How would it be possible to support the *BV chip if it cannot provide 12 V to RP#? How else would it program the boot block? Same for its support of the Micron counterpart, MT28F002C1. It also uses RP#. The support list is here: http://www.autoelectric.cn/MiniPro/TL866II_List.txt

I think the problem here lies not in its inability to supply 12V (TL866II can bump Vpp voltage for up to 18V), but in its inability to route 12V to the RP# pin. These automatic pinout-adjusting programmers must have some sort of MUX/DEMUX circuitry implemented for that reason and I think TL866II's MUX/DEMUX circuit cannot pass such high voltage, because:

1. output voltage of any IC cannot go higher than the Vcc voltage it receives (which probably is 5V in this case);
2. no additional circuit/software was added or adapter made to support this "one of a kind" 28F002BC's pinout.

The TSOP48 adapter for a 28F002BV probably ignores RP# pin existence altogether and just uses WP# pin for unlocking the boot block.

I don't have a schematic for TL866II or any block diagram to support my opinion, purely speculating here.

feipoa wrote: I don't understand how the TL866IIplus doesn't use an external power supply. It seems to rely on power from the USB cord. My Wellon programmer uses a 12 V wall wart for its 12 V needs. The TL866IIplus would need to have some step-up transformer or other IC regulator to get 12 V from 5 V. The programmer itself is tiny, so makes me wonder a bit.

This programmer definitely uses a boost converter for upping the voltage. It's basically the same thing as a buck converter, except main components are arranged differently.

feipoa wrote: EDIT: Looking again at the datasheets, it seems that E28F002BV can get away with 5 V to erase/program the boot block, while the *BC cannot. But then I see in the TL866II support list, for example, that P28F001BX-T and P28F010 are supported, and don't these EEPROMs require 12 V for Vpp? How does it work around this? [And I think the Micron TSOP-40 I have on order, MT28F002C1 also need 12 V, and it is...EDIT: not supported; the *B1 is supported, which is 5 V or 12 V for Vpp.

P28F001BX-T and P28F010 are supported for 12V programming, because they use standard Vpp pin location (pin 1).

feipoa wrote:If I program the *C1, then I need to connect 12 V to Vpp and RP#?

If that programmer will output such voltage in 28F002BV's programming mode, then it's possible.

feipoa wrote:I would then use the TL866IIplus to program this, solder the EEPROM to the motherboard, then follow the BIOS recovery procedure? Is the ideal order of operation to use the BIOS from kixs, or the boot block recovery approach? Due to soldering and desoldering and small delicates, the method with the highest probability of success would go first.

It would be best to flash whole, known good, BIOS image, but if soldering/desoldering job for you is not too difficult, you should first try the recovery method, who knows, maybe there is some serious fault with the motherboard and it won't even show signs of life or some other miscellanious problem will appear. You should also inspect Vpp and RP# pads on motherboard, maybe they constantly receive 12V.

feipoa wrote:EDIT: Comment from the primary eBay distributor of TL866IIPlus,

Hello,Sir
E28F002BC or E28F002BV ,It's the same programming voltage, VCC/VPP:5V/12V.... , Ignore "check ID"

This response is ambiguous, because both BV and BC can use 12V for programming, but only BV supports 5V programming, so the quesstion is, will TL866II even supply 12V at Vpp for those? Anyway, don't forget that you can rewire BC to the pinout of a 28F020 and flash it in either one of programmers.

feipoa wrote:I was considering again the recovery method for updating the BIOS on my Batman's Revenge board as this board was already on my testbed. It contains BIOS 1.00.10.AF2T, the T being OEM for Gateway. I run FMUP.EXE on the BIOS update disk and get the following error:

Recovery method for a working motherboard? Why? Were you trying to flash the one part/two part images or both? What do you get when using "BIOSUP.EXE"?

feipoa wrote:/p has a different function in fmup and it doesn't look like there is a program flag for fmup like there is for iflash. My questions are then a) can I use iflash.exe for the older Batman's Revenge board, and if not, then b) since there is no monitor output in recovery mode

No idea, you'll have to experiment, but if FMUP failed to recognise the image you've tried to flash the first time, then the same thing most likely is going to happen in blind mode.
User avatar
SSTV2
Member
 
Posts: 342
Joined: 2014-11-19 @ 20:30

Re: HELP! - Dead Intel VS440FX motherboard with BIOS EEPROM overheating

Postby feipoa » 2019-8-11 @ 08:56

Your hypothesis concerning inability to supply 12V to RP# sounds plausible. Could be that the multiplixer ports used for switching connections are all used up for other pins; could also be that there is a software way to get the job done, but they don't want to bother. I'm guessing the the Micron *C1 chip will not work on the TL866II's *B1 setting. Most likely it is supplying Vpp at 5 V, rather than 12 V. Same for RP#. Luckily, the Intel *BV chip is supported and hopefully it works on the VS440FX board. A sad end to the story would be having to rewire the Micron *C1 chip to 28F020 pinout, spend days of effort, and nothing works.

I ordered three *BV chips, so even if I goof the soldering, I have another chance. I got the current BC* chip off alright and I have some thing solder and tips to get the *BV on. We'll see. My hands aren't as steady and my visibility isn't as good as it was just a few years ago.

I pressed the distributor a bit more concerning the ability to set 12V to RP# on the BC* chip and C1* chip can. Again, his response not direct.
Code: Select all
TL866II can support VPP 18V and VCC 1.8V-6.5V, You don't have to worry about 12 volts. You can test it yourself with a blank chip.if it can not work,Please send me the screenshot of the error message,I'll report your problem to the manufacturer.

Recovery method for a working motherboard? Why? Were you trying to flash the one part/two part images or both? What do you get when using "BIOSUP.EXE"?


Well, on the Batman's Revenge official recovery disk, it creates a single BIOS image, but it is only revsion 8. So that's of no use... Unless it forces the BIOS to be an official Intel one (non-Gateway OEM), and after such, I can program the .13 official latest Intel BIOS. That's an idea I just thought of. But what I was hoping to do was update my Gateway branded Batman's Revenge board with .10 BIOS to that of Intel .13. It was shown, at least for the VS440FX board, that you can use the recovery method to change the Gateway OEM BIOS to that of Intel for upgrading purposes: viewtopic.php?f=46&t=65168

So I was hoping to use the recovery mode with the two part Intel v.13 BIOS, but not sure which flash updating software will work with both this motherboard/chipset and the two part Intel BIOS. To go from Gateway to Intel, it must be in recovery mode. I'll try using BIOSUP.EXE from the v.13 Intel BIOS software disk and see what happens.

I think FMUP failed to recognise the image because my current BIOS is from Gateway, and FMUP was trying to use an Intel BIOS. From the exampled in the above linked thread, in going from Gateway to Intel BIOSes, it was done and worked in recovery mode, but not in standard mode.

You should also inspect Vpp and RP# pads on motherboard, maybe they constantly receive 12V.

That's a good idea. I'll check that once I get the socket 4 off the test bed.
Last edited by feipoa on 2019-8-11 @ 10:16, edited 1 time in total.
User avatar
feipoa
l33t++
 
Posts: 6267
Joined: 2011-3-07 @ 13:54
Location: Canada

Re: HELP! - Dead Intel VS440FX motherboard with BIOS EEPROM overheating

Postby feipoa » 2019-8-11 @ 10:03

Here are the startup options for BIOSUP.EXE . It even has /r to automatically reset the system when finished updating.
BIOSup_flag_options.jpg

BIOS1013_recovery_using_BIOSUP.jpg


This is the error I get when using BIOSUP.EXE to update the .10 Gateway version to my the .13 Intel version. I with the error was more deterministic, but to me is sounds like it won't update because it doesn't want to let you run Intel BIOSes onto Gateway ones. But apparently recovery mode allows for this.
BIOSup_with_programming_Intel_on_gateway.jpg


I'm at the point now of: to try it, or not to try it. Am I feeling lucky today? The boards on the testbed now, and it won't be on Monday when the EEPROM's come for the VS440FX board.
User avatar
feipoa
l33t++
 
Posts: 6267
Joined: 2011-3-07 @ 13:54
Location: Canada

Re: HELP! - Dead Intel VS440FX motherboard with BIOS EEPROM overheating

Postby SSTV2 » 2019-8-11 @ 13:51

I see, you should then start recovery process using "10013AF2" BIOS update utility. I can see that the boot block is not included into two part BIOS images there, nor a sepparate 16KB file is present. In worst case scenario, you'll end up with corrupted main block, in best case - fully functional latest BIOS.

I find it strange, that there are no instructions in *.txt form, regarding the BIOS recovery in the update utility... Why is there even a identical file to the "1013AF2_.BIO" called "BIOSAF2.REC"? Perhaps FMUP must be used with it in recovery mode, but other than the file name and extension, *.BIO and *.REC files are identical.
User avatar
SSTV2
Member
 
Posts: 342
Joined: 2014-11-19 @ 20:30

Re: HELP! - Dead Intel VS440FX motherboard with BIOS EEPROM overheating

Postby feipoa » 2019-8-11 @ 23:18

For the Batman board, FMUP is the supplied utility. So you think I need to follow the blind manual procedure I outlined previously? Or perhaps it automatically looks for _.BIO? That would be surprising since the newer VS440FX board needs to be directed to it with the /p flag.

I've attached the two BIOSes from user kixs. the NSSI created BIOS is only 64K, while the UNIFLASH created BIOS is only 128K. Does this make sense to you? Shouldn't it be closer to 256K? According to the UNIFLASH documentation, the boot block is automatically recorded starting with v1.33. I sent kixs version v1.40.
Attachments
Intel_VS440FX_Uniflash_and_NSSI_BIOS_from_kixs.zip
(129.57 KiB) Downloaded 4 times
User avatar
feipoa
l33t++
 
Posts: 6267
Joined: 2011-3-07 @ 13:54
Location: Canada

PreviousNext

Return to General Old Hardware

Who is online

Users browsing this forum: Errius, HanJammer, JudgeMonroe, sofakng and 22 guests

cron