Replacing laptop motherboard Super IO chip ITE8987

[dominicM]

I have a hp motherboard that has a short on the 5V rail due to liquid damage. Oddly enough it actually worked after cleaning but failed to power on after a couple of weeks of just sitting on the desk awaiting a new keyboard. I removed the 5V IC and applied external 5V to the rail to troubleshoot. The ITE8987 chip overheats. It seems to me like the chip is faulty as I see no other issues. The question is can this chip be replaced without programming it first?

[testpoint]

yes, IT8987 include program inside, you may use a second hand chip from a same board.

[dominicM]

So a generic part will simply not work at all? There is no scrap parts I can find so if it needs to be programmed then it's not fixable and the whole board will need to be replaced. I also assume a different board with same chip number will not work either as the software is different?

It's odd since I see plenty of videos of people replacing the ITE chip with no mention of reprogramming and it seems like the chip is new and not savaged. Can you link to some info about this as I cant find much about replacing these chips?

[testpoint]

can you give me your HP PC model? I can help you search the broken board in China.  IT8987E  ? which package ?

pls show me the picture of  IT8987E, the vendor said, particular version is NOT include program.

[dominicM]

Laptop is HP 15-p199na and the motherboard is DAY11MBB6Eo REV : E     MODEL : Y11A

Chip has this on it:

Code: [Select]

ITE
IT8987E
1435-BXA
SC26LA

[difinity]

The IT8987 is in the list of supported devices from programming with the SVOD programmer. One of a couple of devices for programming Super IO chips.

[testpoint]

the vendor said, BXA can be replaced directly, price is 2 USD only. new chip.

[dominicM]

the vendor said, BXA can be replaced directly, price is 2 USD only. new chip.

So any new chip can be used without programming? Could you elaborate on why BXA can be replaced? How can I know when a chip needs to be programmed and when not?

[testpoint]

do not know, the vendor said, maybe they have a document

[dominicM]

do not know, the vendor said, maybe they have a document

What vendor is it? Do they sell this stuff directly or do you just know someone personally?

[khaahk]

The 8987 should be programmed from main bios when mainboard is powered on. At least on this mainboard. It initially might contain some basic code to power on mainboard. Where did you inject the 5V, because the 8987 should be connected to 3.3V ?
Quanta Y11 schematic is freely available online.

[dominicM]

The 8987 should be programmed from main bios when mainboard is powered on. At least on this mainboard. It initially might contain some basic code to power on mainboard. Where did you inject the 5V, because the 8987 should be connected to 3.3V ?
Quanta Y11 schematic is freely available online.

You are correct, it is 3V in the schematic. I assumed it was 5V rail for whatever reason but it could well be 3V rail. Can't be sure as the markings on the 3V and 5V IC's do not match the schematic and all the ones for sale have varying markings too. Voltage never exceeded 1V due to short circuit so it should be fine.

So how can you tell that this ITE chip does not need to be programmed first?

[khaahk]

I believe there is no obvious way to tell - only experience - if not your's then somebody else's who has the will to share.
One indication is if the code is kept inside main bios - there would be no point to keep it there.

[dominicM]

I believe there is no obvious way to tell - only experience - if not your's then somebody else's who has the will to share.
One indication is if the code is kept inside main bios - there would be no point to keep it there.

So there is no extra hardware required for this un-programmed chip to work? If it's software only I don't understand the logic in making some motherboards require the ITE chips to be programmed externally, can you see a reason? Commenter above implied that BXA markings on the ITE chip are indicative of it not requiring programming, any idea what BXA means at all as I can't even find a datasheet for this chip.

[amyk]

The first two are paysites, the last one I was unable to successfully register. But if anyone can get the datasheet, please share

http://bios-fix.com/index.php?threads/it8987e-datasheet.481/
https://vinafix.com/threads/it8987e-datasheet.26578/
https://www.phang-nganotebook.com/forum.php?mod=viewthread&tid=8536

[testpoint]

can not find any document, the vendor just according the model number, they are not sure it was programmed or not, I think it has program, so you just use the used chip is OK.

[dominicM]

can not find any document, the vendor just according the model number, they are not sure it was programmed or not, I think it has program, so you just use the used chip is OK.

So which is it does it need to be programmed or not?? You said it doesn't need to be programmed but now it sounds like you are saying the opposite! There are no used chips I can see anywhere and certainly not from the same motherboard.

[khaahk]

It is a microcontroller to be used by many manufacturers, so it is made to be universal. By pulling some pins low it enters the spi programming mode. That depends on the mainboard on which it is installed. So i suggest you buy the new chip, put it on and see what happens. In worst case you lose a few $. If you want you can buy the programmer later if needed.

[dominicM]

It is a microcontroller to be used by many manufacturers, so it is made to be universal. By pulling some pins low it enters the spi programming mode. That depends on the mainboard on which it is installed. So i suggest you buy the new chip, put it on and see what happens. In worst case you lose a few $. If you want you can buy the programmer later if needed.

Programmer is not worth the cost but I will try a replacement chip, and report back then.

[Rasz]

why not buy preprogrammed chip?

[dominicM]

why not buy preprogrammed chip?

Because they are not available not to mention who knows what programming it needs for a specific board.

[Rasz]

why not buy preprogrammed chip?

Because they are not available not to mention who knows what programming it needs for a specific board.

 
people who repair laptops do, obviously
dumps of this KBC firmware are available
https://dr-bios.com/get/quanta-y11a-day11amb6e0-it8987e-ec-bios-download/29321
https://www.elvikom.pl/schemat-hp-pavilion-15-quanta-y11a-y11c-day11amb6e0-t43358.html
http://vlab.su/viewtopic.php?f=53&t=5916&start=760

its pretty self explanatory, find someone willing to sell you programmed EC on one of the russian/polish forums

[dominicM]

why not buy preprogrammed chip?

Because they are not available not to mention who knows what programming it needs for a specific board.

 
people who repair laptops do, obviously
dumps of this KBC firmware are available
https://dr-bios.com/get/quanta-y11a-day11amb6e0-it8987e-ec-bios-download/29321
https://www.elvikom.pl/schemat-hp-pavilion-15-quanta-y11a-y11c-day11amb6e0-t43358.html
http://vlab.su/viewtopic.php?f=53&t=5916&start=760

its pretty self explanatory, find someone willing to sell you programmed EC on one of the russian/polish forums

This is utter overkill for such a low cost board. It is absolutely not worth the time required to find it if it's not readily available.

[electra2017]

Can You please tell me with the your results about this issue as I am having same problem with mine?
And if anyone can provide me with vendors who sell this chip I will be very Thankfull

[dominicM]

Honestly I still haven't gotten around to it. Will try and do it shortly, will report back if I do...

[BlackCore]

Hi, I'm repairing dead HP Pavilion 17-ab002nm. The motherboard is DAG37AMB8D0, REV: D. For some time motherboard has been intermittent, if I had disconnected the battery and waited for some time, then plugging first the battery in, then AC adapter, the laptop would turn ON, and stay ON on as long as it would have power. Now even this power sequence doesn't work. After some measuring and testing, I came to conclusion, that there is either problem with KBC or BIOS, after consulting some experts, we came to the conclusion that it's KBC that's making problems.

KBC is ITE brand IT8987E, 1604 - BXA version.

Anny suggestions of your experience, what could be the fault ?

Now I'm searching for a programmer for KBC and I found three different ones.

1. Vertyanov ITE ENE IO Programmer - eBay: http://ebay.to/2E4omSc
2. ENIT SIO Programmer - Aliexpress: http://bit.ly/2nOoGxs
3. SVOD Programmer - You can buy them on dr-bios.com and link to how to use: http://bit.ly/2EnfcTW

I would like to know if anyone has used them and what do they prefer or recommend.

Thanks.

Never give up 

[electra2017]

Did you have any chance to fix the problem?

[dominicM]

I did replace the chip however it did not fix the problem. It seems I get about 3.3V on both regulators instead of the expected 3.3V and 5V.The ICE chip doesn't get hot anymore so I do believe it was indeed shorted internally. Not sure what's causing this and I don't think I want to spend anymore time on this. That being said if anyone know why this might be happening let me know and I will test it out, it would be interesting to find out the fault.

[BlackCore]

Where did you buy 8987E? If you got 3.3V on 5V rail, there still can be a problem somewhere else. Did you check for short to ground on 5V VALW/LDO rail?

I'm still solving the problem.. I've got a new chip, when I first tried to program the new chip with Vertyanov and JAWS, there was no problem, but now when I try again chip does not respond. I tried it to flash it with bios and without bios chip attached. So, for now, I'm just struggling to program the chip. It looks like a dead chip, now I'm waiting for new one...

[dominicM]

Where did you buy 8987E? If you got 3.3V on 5V rail, there still can be a problem somewhere else. Did you check for short to ground on 5V VALW/LDO rail?

I'm still solving the problem.. I've got a new chip, when I first tried to program the new chip with Vertyanov and JAWS, there was no problem, but now when I try again chip does not respond. I tried it to flash it with bios and without bios chip attached. So, for now, I'm just struggling to program the chip. It looks like a dead chip, now I'm waiting for new one...

It was just a cheap ebay purchase. I didn't know if it needed to be programmed so just gave it a shot, but from what you said it seems it does need to be programmed so I can't do that anyway without a programmer.

I did not find any shorts. I also tried removing the 5V regulator and feed 5V into the line, power adapter light went on but still no power button led light. Maybe it's the fact that the BIOS chip is not programmed?

[BlackCore]

Yea IT8987E and some others SIO chips as NUVOTON and ENE needs to be programmed for each individual motherboard separately, and so is the BIOS chip. Both are unique to each MB.

[BlackCore]

I have an update. I've reprogrammed IT8987E and also reprogrammed BIOS FLASH. I got the computer working. There is only one problem. It works only with battery in. If I disconnect the battery then the computer shuts down... Will look into it. But it seems that there was a problem with ITE8987E. I don't know if the reprogramming BIOS helped, but I did it anyway Will let you informed if I solve the battery problem. Take care.

[dominicM]

I vaguely recall reading that battery may be required by (bad) design, you should check if it's not the case with this laptop.

[BlackCore]

I think that's not the problem in my case... I have booted into the system, and I see that Battery will not charge when the computer is on. When the computer turns off, battery immediately starts charging (i can see that's charging via a status LED light on input jack). It seems that the charger IC has problems with recognizing when to turn on battery charging FET's. I will look into it...

[Rasz]

sounds like one of the power routing transistors is blown

[BlackCore]

I've checked both ACFET and BATFET in the charging circuit, and they are both fine...

1. So, for now, I've figured out, that SIO and BIOS are OK.
2. Computer boots normally, when both battery and DC plug are connected in specific order. First, I need to plug the DC supply then Battery and computer starts. When turned ON, I see that it draws about 1,4A from the charger, then in the next second, it cuts away ACFET and powers the board from Battery. 
3. Charging circuit is charging battery ONLY when the computer is OFF. When the computer is on and I try to reconnect DC power supply, again it charges for a second, then stops.
4. I've checked Power adapter with my Dummy load and it easily provides 6A, voltage drops to around 18,5V, but this is expected due to cable length. And schematic says that VIN is set to >15.2 (AC GOOD), so this cannot be the problem).

I'm going to measure SRN/SRP vs. Iout (it should be 20x SRP-SRN). As discovered I suspect, that there is a problem somewhere in the charging circuit. Thank you all for your suggestions...

I will try to hold tongue at the right angle and hope it helps. 

[BlackCore]

I have an update to make.

When looking at 3VPCU (that's LDO of 3.3 V Regulator that supplies power to the SMC, BIOS and some other components) voltage jumps from 3.3V to 4V and in some point, it was stuck on 4.2V. I'm still going further in though.. I've removed 3V PCU line (used Exacto knife and cut the trace from +3.3V LDO - SY8208B) and on the system side of +3VPCU I get 1.3V From where

After cutting the line, LDO regulates normally and outputs 3.3V even when loaded. So LDO is working properly.

So investigating further, I measured all pull-ups that are tied from +3VPCU to system and ending removing them all, but still measuring this +3VPCU rail gives me 1.3V from thin air. SMC has been removed and replaced with working one, reprogrammed both BIOS and SMC (SIO IT8987E)... So I'm pretty sure that there is nothing wrong with SMC and BIOS.

Then I've spotted something strange. When lowering the voltage on my power supply from 19V to 0 the voltage on +3VPCU drops to.

At 19V input I get 1.347V, measuring short to ground 7,6mA
at 10V there is 1.089V; 4,25mA
5V input +3VPCU is at 0.921V; 2,13mA
2V ----> 0.656 V 0,79mA
1V ----> 0,484V 0,39mA

I suspect that something is supplying voltage from DC input to SMC through one of the pins..

Anyone has some clues?

[BlackCore]

Where did you buy 8987E? If you got 3.3V on 5V rail, there still can be a problem somewhere else. Did you check for short to ground on 5V VALW/LDO rail?

I'm still solving the problem.. I've got a new chip, when I first tried to program the new chip with Vertyanov and JAWS, there was no problem, but now when I try again chip does not respond. I tried it to flash it with bios and without bios chip attached. So, for now, I'm just struggling to program the chip. It looks like a dead chip, now I'm waiting for new one...

It was just a cheap ebay purchase. I didn't know if it needed to be programmed so just gave it a shot, but from what you said it seems it does need to be programmed so I can't do that anyway without a programmer.

I did not find any shorts. I also tried removing the 5V regulator and feed 5V into the line, power adapter light went on but still no power button led light. Maybe it's the fact that the BIOS chip is not programmed?

If you want you can send it to me and I will reprogram it for you. You need to send both BIOS and SIO chip, they both need to be programed/checked.

[dominicM]

Where did you buy 8987E? If you got 3.3V on 5V rail, there still can be a problem somewhere else. Did you check for short to ground on 5V VALW/LDO rail?

I'm still solving the problem.. I've got a new chip, when I first tried to program the new chip with Vertyanov and JAWS, there was no problem, but now when I try again chip does not respond. I tried it to flash it with bios and without bios chip attached. So, for now, I'm just struggling to program the chip. It looks like a dead chip, now I'm waiting for new one...

It was just a cheap ebay purchase. I didn't know if it needed to be programmed so just gave it a shot, but from what you said it seems it does need to be programmed so I can't do that anyway without a programmer.

I did not find any shorts. I also tried removing the 5V regulator and feed 5V into the line, power adapter light went on but still no power button led light. Maybe it's the fact that the BIOS chip is not programmed?

If you want you can send it to me and I will reprogram it for you. You need to send both BIOS and SIO chip, they both need to be programed.

Thanks very much for the offer. I don't know yet if I will get it fixed, will have to come back to it as I had given up for the moment. I assume that the system should start even with (un)programmed ITE chip right? Right now I get no fan spin at all.

[BlackCore]

Where did you buy 8987E? If you got 3.3V on 5V rail, there still can be a problem somewhere else. Did you check for short to ground on 5V VALW/LDO rail?

I'm still solving the problem.. I've got a new chip, when I first tried to program the new chip with Vertyanov and JAWS, there was no problem, but now when I try again chip does not respond. I tried it to flash it with bios and without bios chip attached. So, for now, I'm just struggling to program the chip. It looks like a dead chip, now I'm waiting for new one...

It was just a cheap ebay purchase. I didn't know if it needed to be programmed so just gave it a shot, but from what you said it seems it does need to be programmed so I can't do that anyway without a programmer.

I did not find any shorts. I also tried removing the 5V regulator and feed 5V into the line, power adapter light went on but still no power button led light. Maybe it's the fact that the BIOS chip is not programmed?

If you want you can send it to me and I will reprogram it for you. You need to send both BIOS and SIO chip, they both need to be programed.

Thanks very much for the offer. I don't know yet if I will get it fixed, will have to come back to it as I had given up for the moment. I assume that the system should start even with (un)programmed ITE chip right? Right now I get no fan spin at all.

SIO is responsible for almost all voltage regulators (They call it Brains of the circuit) so, there is no way the computer starts without that chip... And I think that if the chip is not programmed the outcome will be the same... No power no life.

 You can check if the firmware of SIO and BIOS is OK. Find #RSMRST signal and check it. If you have 3.3V on #RSMRST line than the BIOS and SIO are most probably OK. If #RSMRST is not present then you most probably have a problem either with SIO or SIO BIOS.

[dominicM]

Where did you buy 8987E? If you got 3.3V on 5V rail, there still can be a problem somewhere else. Did you check for short to ground on 5V VALW/LDO rail?

I'm still solving the problem.. I've got a new chip, when I first tried to program the new chip with Vertyanov and JAWS, there was no problem, but now when I try again chip does not respond. I tried it to flash it with bios and without bios chip attached. So, for now, I'm just struggling to program the chip. It looks like a dead chip, now I'm waiting for new one...

It was just a cheap ebay purchase. I didn't know if it needed to be programmed so just gave it a shot, but from what you said it seems it does need to be programmed so I can't do that anyway without a programmer.

I did not find any shorts. I also tried removing the 5V regulator and feed 5V into the line, power adapter light went on but still no power button led light. Maybe it's the fact that the BIOS chip is not programmed?

If you want you can send it to me and I will reprogram it for you. You need to send both BIOS and SIO chip, they both need to be programed.

Thanks very much for the offer. I don't know yet if I will get it fixed, will have to come back to it as I had given up for the moment. I assume that the system should start even with (un)programmed ITE chip right? Right now I get no fan spin at all.

SIO is responsible for almost all voltage regulators (They call it Brains of the circuit) so, there is no way the computer starts without that chip... And I think that if the chip is not programmed the outcome will be the same... No power no life.

 You can check if the firmware of SIO and BIOS is OK. Find #RSMRST signal and check it. If you have 3.3V on #RSMRST line than the BIOS and SIO are most probably OK. If #RSMRST is not present then you most probably have a problem either with SIO or SIO BIOS.

Do you have the datasheet for this board? I never found one, so I am not sure how I would find #RSMRST line. I didn't replace the BIOS chip just the SIO/ITE chip but it's not programmed sso I guess that would be same as no chip at all in this case. Would be interesting to check anyways.

[Rasz]

Anyone has some clues?

diagram is too big to directly upload (2.8MB vs 2MB limit), so link to some shitty pdf hosting service https://www.pdf-archive.com/2018/04/06/hp-pavilion-17t-ab-series-quanta-g37a-g37b-r1a/


is power supply legit? whats on AD_ID?
thats with battery disconnected?
lift/check  Pr1001? might be charger chip after all

[BlackCore]

is power supply legit? whats on AD_ID?
thats with battery disconnected?
lift/check  Pr1001? might be charger chip after all

Yes, supply is OK, middle pin of the power supply connector is outputting the same voltage as the + (inside rim) of the power supply.  I've tried to connect both original PSU and my LAB power supply and I get the same results from both of them.

You guessed it !!! AD_ID is feeding the SIO with voltage from the output... Zener is clamping voltage to 5.6V, but I measured current draw on AD_TYPE line to the SMC (I've cut PCB trace between kathode of Zenner and pin 67 on SMC) and it's about 6mA, so that's limit of the 2k resistor in series with AD_TYPE line from the output.

When I connect the battery, there is still ~6mA's of current draw on AD_TYPE, but the voltage on +3VPCU immediately drops from 4.2 V on 3.3V after pressing the power button when the battery is connected. The laptop still only starts with the battery connected and the current draw is still ~6mA when the battery is connected.

PR1001 is 100k as it should be, and as I said I have removed all the resistors on the +3PCU line and tested them, they are all fine.

I've changed the resistance of R5117 to 1k to increase current (i taught that Zener is failing) but I think that this is not the case. With 860 ohms, Zener stabilized on 6.5V but, I was thinking that this is too much current send to SMC so I'm just rethinking my strategy

How and where is this current sinking through the SMC, that's my next question?

[Rasz]

AD_ID should be ~ 1.4V under load in a working circuit ?
is D5100 a dead short by any chance?

I measured current draw on AD_TYPE line to the SMC (I've cut PCB trace between kathode of Zenner and pin 67 on SMC) and it's about 6mA, so that's limit of the 2k resistor in series with AD_TYPE line from the output.

measure resistance to ground on AD_TYPE
any caps between zener and ec on that line? also check C5111/C5112 just in case, shorted cap could cause excessive current draw

> I've tried to connect both original PSU and my LAB power supply

wait, you put 19V lab supply on ID line? now I have no idea how it works, but I have a strong suspicion ID should be severely current limited, and just blasting 19V from unlimited source will fry something, maybe even EC input

also check ADAPTER_SEL_EC

[dominicM]

Thank you for the datasheet, it's very helpful. I am also trying to check AD circuit as I am not even getting VIN led. Where is d5100 though? There are only a  handful of diodes with same package but none of them have anything close to D5100 printed next to them, how are you finding these components on the board?

[Rasz]

ctrl-F on the pdf
but this is pavilion 17 diagram, yours is https://drive.google.com/file/d/0Bz_l_sRYth4MZU9yS0d3NTBNaW8/view

[BlackCore]

AD_ID should be ~ 1.4V under load in a working circuit ?
is D5100 a dead short by any chance?

As you can see on Page 40 of datasheet, that AD_ID is 8 pin on DC_JACK and it's supposed to have 19,5 V feed in. I've measured the resistance between VDD and AD_ID pin on original PSU and it's 200k Ohms. My DC_JACK converter (for lab power supply) has also 200k resistor inside, so the current is limited on AD_ID line, and the AD_ID line is 19V in working circuit under load. D5100 is not installed on the board, so it's open contacts.

>measure resistance to ground on AD_TYPE
>any caps between zener and ec on that line? also check C5111/C5112 just in case, shorted cap could cause excessive current draw

Resistance on AD_TYPE is 14k
No there is no caps on that line. Both capacitors C5111/C5112 are OK..

>wait, you put 19V lab supply on ID line? now I have no idea how it works, but I have a strong suspicion ID should be severely current limited, and just blasting 19V from unlimited source will fry something, maybe even EC input

I have DC Jack converter, that has 200k Ohms resistor build in between +VDD and AD_ID pin, so the pin is current limited..

> also check ADAPTER_SEL_EC

ADAPTER_SEL_EC line seems also OK.

I'm thinking that something may be related to the charging circuit, cuz battery is charging when the computer is off, but when turned on, It stops charging and draws power from the battery. Problem began when computer was turned on, then it stopped charging while ON, then user had plugged/unplugged adapter 5 times and for some time the problem was solved. But after few months even that didn't help. Now the battery is charging only when the computer is OFF...