Dell G3 3579 BIOS chip ID

[ezenia]

Hello,

I bought a Dell G3 3579 eBay. The fault was not mentioned and it did not come with charger. Before trying to power off battery, I did some basic checks on the coils (in resistance mode) to see if there was any short. None were reading short. Decided to power the board to see if anything happened. The keyboard backlight came on. About 5 seconds later, the fans begin spinning at max speed. The white LED comes on at this time, stays solid for a bit and goes off. Fans continue to spin at max. Display is black (no backlight). If I hold down power button, board goes off. I did give the board some time to see if it did anything (like reboot), but it does not. It just stays in this state until I press the power button.

Tried the usual basics of removing all non-essentials (like HDD, keyboard, touchpad, etc) and also tried another RAM stick (in both slots, this stick worked in another Dell laptop). Still same exact problem.

So my next step is trying to reprogram the BIOS. I have a CH341A programmer that I want to try using (haven't tested it since I bought it). My issue right now is locating the BIOS chip. I can't see any "Winbond" chip. There's one 8-legged IC next to the chipset that I think might be the BIOS, but I am not sure. The markings from what I gather is "AH1921 2501270510 U96224"(could be SLG instead of 510, not too sure). Searching those doesn't mention anything about BIOS. Just shows some results about some 3-legged Hall Effect IC which doesn't isn't the same as this IC.

Here's a close-up of the IC I think might be the BIOS:

Here's an image of more of the board:

I can't see any other chip that looks similar on the board. None have "Winbond" written on them anyway. Maybe the IC looks completely different to what I usually see, so maybe I am missing it.

The board itself is a CAL53 LA-F611P REV 1.0 (A00)

Appreciate any help on this. Before unnecessarily blasting the board with hot air, I'd like to make sure. I don't want to remove the IC if it's not the BIOS.

[ezenia]

Found a schematic for this board that showed this IC was the BIOS. So I've now desoldered the chip and soldered it to the CH341A. It's read the chip fine and I've saved the original on my PC (file size is 16,384KB which seems to be the correct size).

Now I am stuck on getting another BIN file. I tried going to the Dell website but they only offer .EXE and .RCV files. Is there a way to change these BIOS files to a .bin so I can try programming it?

[ezenia]

Decided to try connecting my bench power supply to the laptop since I don't have the charger. The laptop behaves slightly differently with this. I did not connect battery. As soon as I inject power to the laptop at the DC-IN connector, I see the current jump to around 0.500A without me even pressing the power button. Moments later, the board comes on by itself and current jumps up to 0.800a and stays there. Keyboard backlight comes on and stays on. A white LED still comes on briefly (for maybe 2-3s) and goes off. Laptop seems to stay in this state until I cut my power supply or press the power button. When I tested just with the battery, I had to press the power button to get it to turn on (same fault occurs after). So it turning on by itself with the bench supply is different. Not sure why it's doing that.

I did try programming another BIOS file I found in the BIOS request thread (was for the exact board model) just to see if anything changed. Nothing did. Same fault as before. Not sure if this rules out a BIOS problem yet since it's still possible the one I downloaded isn't good for this machine.

Does the current consumption help tell anyone anything about what might be going on here? I did the same test without any RAM inserted. It's the exact same. Makes no difference whether the RAM is inserted or not.

[mon2]

All bios questions should be posted inside the dedicated bios forum. See above. The bios gurus reside there.

Keep in mind this is a Dell. Dell demands the use of the 1-wire ADAP_ID / PS_ID (or similar) communication line to chat with the power adapter IC. This is critical. In the absence of this communication, the CPU will be throttled and the battery charge feature will shut off. Respectively, a DC lab power supply is not suitable. Suggest to test with the OEM power supply for this laptop.

[ezenia]

I will be getting a charger soon, just wanted to probe around and see if this board is actually fixable.

I found some schematics posted and a boardview on this forum for this board. I've been looking at them and doing some more voltage measurements. Everything seems to look OK, but there are some measurements on PQH01 I am not too sure about. Schematics show it's for the VIN +1.05V rail.

I've marked some of the voltages I've been measuring. Does that look normal or could it explain why the board is not passing POST?

This is with the battery connected. I get similar readings with the DC power supply (just the 14V jumps to like 18V).

Link to schematic - https://www.badcaps.net/forum/docume...atic-boardview

[khaahk]

And why dont you measure the voltage on the coil PLH01, if it is 1,05V then you can move on. On the gates of the dual transistor there are fast moving impulses and measuring them with voltmeter does not give much information. The 14/18 V should be ok.

[ezenia]

Yes that coil reads 1.05V. Got an OEM Dell charger, plugged it in and same exact symptoms. When plugging in the charger, I do see the white led dimly flicker. After like 4-5 seconds, white LED comes on along with max fan speed and no display. Stays like this indefinitely and doesn't restart. If I press power button (just press not hold), it turns off. Some seconds later, it turns itself on again with the same symptoms. It only seems to do this auto-on with the charger. If on battery only, I have to press the power button to get it to do anything.

Recorded coil voltage measurements and checked both ends against GND. The 0.002V was one side of the coil that I saw sticking out enough to get my probe in. Not sure on other side as heatsink is blocking it (will take it off if needed to check voltages there). Almost all other coils have the same voltage values on both sides (that's why I only added one box with the value) except the one 0.954V and 0.996V - that seemed like a bigger difference, so I added labels for both sides.



On battery, all voltages are the same except for the 0.972V coil. It's at 0.

With battery and charger connected, this coil goes up to about 15.7V. The solid white LED also comes back after it goes off. So it turns on for a few seconds, goes off and then comes back on (stays on after this). Guessing this is a charge LED. I checked the voltage at the battery connector and the voltage is slowly increasing.

I held down the D key during a power up and it ran through the LCD test. That test seemed to work OK.

I was hoping this would be a BIOS issue, but I am starting to think it might be a bad CPU/GPU (something critical like this that can't be easily fixed). What is strange is why it seems to auto power on if trying to turn it off when on charger. It's like it doesn't want to stay off. Soon as I remove charger, force power off and leave battery connected, it doesn't try starting by itself. I have to press the power button to get it to come on.

[ezenia]

Took heatsink off and checked coils around CPU/GPU.

CPU coils:



GPU coils:



Seems like the issue might be a shorted CPU? I've read that CPUs can read very low ohms. However, 0.02 seems way too low and is the same as when my probes touch so it doesn't look right. GPU is also low, but it looks a lot better than 0.02. It's closer to 1 ohm and these coils do get voltage when power is applied. So the GPU looks OK. The CPU doesn't. The PCH looks OK because those rails are all up and there's no shorts around that area. Really don't want it to be a CPU issue, but looking more and more like a failed CPU.

[Sephir0th]

You have a Multimeter which shows TWO decimal places? I would like to see the display of your MM when measuring, because especially in the GPU area it's a big difference between 0.08 and 00.8 Ohms.

However, you are most likely right and the CPU is shorted. Only other possibility would be that one of the DrMOS has shorted the whole Core Rail to GND

[ezenia]

Oh right, it shows "000.2" on my meter for the CPU. GPU one shows "000.8".

I can't see what else could be wrong. I tried programming another BIOS file that was posted in the requests thread for this model just to see if anything different happened. Same exact symptoms. I don't think it's a BIOS issue which is a shame. Some others with this exact issue (black screen, max fan) fixed it with a BIOS flash so was hoping the same would happen here, but the low ohms on the CPU coils says otherwise.

Would it be worth trying to desolder the mosfets one by one to see if that makes a difference? Or is it pointless going further (CPU is dead regardless of what is shorting the rail to GND)?

[ezenia]

Tried lifting all the coils on one side to see if that led to any changes in resistance. The side towards the Mosfets now shows very high resistance (it's in the kilo ohms). The side towards the CPU still shows that very low 000.2 ohms value. Surely that means the CPU is gone?

[Sephir0th]

Yes, surely.

PS: No BIOS firmware in the world can resolve a hardware short circuit. This part of the story was absolute nonsense and belongs to the myths and fairytales.