M755LMRE Bios Problem

**davicito** · October 26th, 2005, 02:32 PM

I have an M755LMRE MoBo with GfXcel Pc133.........I have programed a new bios chip with 020514s.rom but the board does not BOOT. I would like to make the bootlock work but it only did it once (I loaded the amiboot.rom but afterwards it died on me) and even while pressing CTL+HOME it won´t work. I´ve tried several bios and different memory. How can I differenciate between BIOS nad MEMORY errors ? Also would like to make the bootlock work ! Someimes I get blinking on the keyboard LED´s but can´t decipher it !

Would appreciate help

**Garak** · October 26th, 2005, 05:56 PM

Hi,

If the BIOS has become corrupt, which it seems to me is the case, then there are a few things you can try. Take a look at the BIOS recovery sticky - I think you maybe able to salvage this via the boot floppy, but can't guarentee that. The alternative would be to use a BIOS recovery company. As for bootlock, I'm not entirely sure what you mean, can you explain a bit more?

Gary.

**davicito** · October 27th, 2005, 12:43 PM

I think the Bios issue is solved..but now it boots sometimes only..!!!
Th processor I have is a Celeron 700/128/66/1.7v From this I can only use PC66 or PC100 memory chips. The weird thing is that it was working with a PC133 memory !!! I´ve tried several PC100 chips and when the system starts it will only work for a few minutes and then it dies.

Would this be a memory or bios problem. At this moment it will not boot at all!

Thank you for your reply

**Garak** · October 27th, 2005, 12:50 PM

Well, to me it sounds like bad caps - and since the model number of the motherboard points to a PC Chips clone - I wouldn't at all be surprised.

If you feel confident enough (and I say this because some people aren't), take the side panel off your PC and look at the motherboard, if you see any obvious damage as illustrated in the image below - its likely the motherboard its self is the cause of the issue (which can be repaired - the big question is, is it worth it to yourself).

I hope this is of some help to you.

Good luck.

**davicito** · October 28th, 2005, 09:25 AM

I followed your lead and checked out the caps but they all seem ok. I measures voltages with the board on and they all showed voltage readings between 1.5 and 3.5 volts. The board looks perfect, not a scratch or burn and everything I was able to check, including transistors and voltages on BIOS ( 5V except for ground & one other pin).

I left the board overnight and this morning I connected keyboard , hard disk and floppy and it booted ok. I entered setup and then booted Windows.....
but it died halfway thru initializing......turned it off and now witout changing anything.....it´s dead again !!! This is driving me crazy but more because curiosity is killing me !!

Is there perhaps a log that windows98 leaves on the HD. By the way I booted the HD on another machine and it works ok so I discard the HD.

I get absolutely no beeps when it won´t boot....does that mean that the bios doesn´t even start ??

Thanks for your help Gary

**Platypus** · October 28th, 2005, 09:56 AM

Originally Posted by davicito

I measures voltages with the board on and they all showed voltage readings between 1.5 and 3.5 volts.

Is there perhaps a log that windows98 leaves on the HD. By the way I booted the HD on another machine and it works ok so I discard the HD.

I get absolutely no beeps when it won´t boot....does that mean that the bios doesn´t even start ?

Depending on where you're measuring, you may have a faulty power supply, certainly another PSU is something to try with those symptoms.

The voltages to the mainboard are shown here:

http://pinouts.ru/data/atxpower_pinout.shtml

for an ATX connector - notably +3.3V and +5V should be within 5% tolerance.

You'll find two log files in the root folder C:\ , bootlog.txt is the current log, and if Windows has had a partial start, it can be read from a DOS bootdisk to see where the startup failed.

Bootlog.prv shows a previous good boot, and can be used for comparison.

No beeps at all does indicate the BIOS code isn't getting to run, ie the CPU can't operate correctly for whatever reason.

Gary, I take it davicito means the bootblock where bootlock was mentioned.

**davicito** · October 28th, 2005, 12:42 PM

Thank you for your reply...

I have used several power supplies..same problem..voltages measure all ok.

The bootlog does show an error :
.......
Terminate = RIT
EndTerminate = RIT
Terminate = Win32
EndTerminate = Win32
EndTerminate = KERNEL
[001490C4] ASD: failing Calling a PnP BIOS, d=P0A000000B800F000, rc=00000001[001490C4]

What does it mean ? The last line was the only difference with the bottlog.prv.

David (davicito = dimunitive of david in spanish)

**NooNoo** · October 28th, 2005, 02:00 PM

some light reading for you about ASD

**davicito** · October 28th, 2005, 05:05 PM

Thank you all for your help .
I am about to give up. The problem does not seem related to Windows since it won´t even boot at all. I have the MoBo with a keyboard and floppy only.
Not a BEEP to be heard ! I have used an AT Power Supply as well as ATX. NADA !

**Platypus** · October 28th, 2005, 08:25 PM

Unfortunately, PCChips motherboards have a reputation for using poor quality EEPROM chips, which can fail when flashed. Normally the board will then just fail to boot, rather than running for a while then failing like you had at first, although I suppose this is still possible.

Originally Posted by davicito

How can I differenciate between BIOS and MEMORY errors ?

Try to boot with no memory fitted. If the BIOS (or at least the bootblock) is active, there should be a "memory not found" error beep. This varies from BIOS to BIOS, but it makes it clear it isn't happy. If you still get no beeps, the memory isn't causing the non-boot condition.

**davicito** · October 29th, 2005, 01:05 PM

Well I have finally isolated the problem...THe voltage of 1.7 is not reaching the CPU ......when that voltage is present it runs but it eventually falls to .3V
and crashes...That voltage should be present on the rectifier MOSPEC S20S40C.....the question is what causes this ? Is it the regulator chip or ???? I do know it´s not the Power Supply, since I´ve used at least 4 different ones..

The big question is how do I fix it ????

At least it´s been a learning experience !!

**Platypus** · October 29th, 2005, 10:35 PM

You're doing a sterling job of fault-finding.

Although I do component-level repairs, Ruslan is our member with the most direct experience in this area. If he doesn't notice this thread of his own accord, perhaps you could attract his attention with a private message.

It is still possible that electrolytic capacitors could be contributing to the problem, if they are developing high series impedance, even if they have not begun to visibly leak or bulge. It may be more likely that a switching mosfet transistor is going bad, PCChips is known for being weak there too.

**davicito** · October 29th, 2005, 11:42 PM

Thank you for your motivation!

Without removing the caps how can I tell if one is goin bad ?

**Platypus** · October 30th, 2005, 05:01 AM

For that kind of fault an ESR (Equivalent Series Resistance) Meter can reveal the deterioration, and may be effective "in-circuit". Where capacitors are in parallel, as they often are on supply lines, it's best to be able to test each individual component. In the absence of test equipment, but if you have suitable soldering facility, for the cost of a few high ripple-current rated capacitors, substitution is the simplest action.

If you have ready access to the reverse of the board it could also be worth checking for evidence of bad solder joints around the CPU core voltage conversion components.

**Ruslan** · October 30th, 2005, 09:41 AM

It could be actually bad capacitors... Take a closer look again - especially at those 7 caps located near power connector as they are the ones responsible for stable CPU voltage. By the way, on a picture given by Garak, were 8 defective caps (out of 12), did you spot them all?

Any capacitor with top other than ideally flat is defective! Also, capacitors could be damaged on pins (bottom) side, not only tops... Check it out too...

I wouldn't trust much an ATX power supply (especially the one what came with a cheap motherboard like this one) as PC builders usually use the cheapest PSU/Case combo and cheap PCChip/ECS MB to build the cheapest PC...
You can open that PSU and check if there is no any burn signs and/or defective caps inside... AT PSUs were much more reliable, IMHO.

Unfortunately, bad power supply could destroy some of motherboard's components - fisrt of all, CPU voltage regulator. And looks like you have problems exactly with that. If all the capacitors are fine, check the power MOSFET (Field-Effect Transistor) located near that MOSPEC chip...
Measure Case-to-pins resistance at lowest range - shouldn't be Zero in any direction unless that MOSFET is fully opened. It is possible to open it by applying about 5-10V to the Gate-Case pins (some of multimeters can do it, some - not)... Apllying reverse polarity voltage to the same pins should close the MOSFET then. It that MOSFET if good (and I'm pretty sure it would be so), then voltage controller chip is the one to be blamed. Most of an older PCChips motherboard are using 34063A voltage controller chips, one (at least) or two. This motherboard has 34063 chip soldered (see picture below)
close to AT power connector... Locate it, download the PDF file for typical schematics... If you have an oscilloscope, you can check the signal coming out of that chip to the gate of MOSFET... May be this chip needs to be replaced - you can find it in local electronic shop (cost around 30-50 cents) or desolder it from another PCChips board...

There were at least two different revision of that board, by the way http://www.pcchipsusa.com/prod-m755mr.asp - so make sure you have a proper BIOS version...