Introduction: Dell Precision M4300 Repair Attempt
I bought this laptop from a Swedish version of ebay.com called Tradera, to troubleshoot and have some fun. I thought It was not opened and the seller was the original owner of the computer. But, I discovered later that he didn't know so much about the computer, he was not the original owner.
The fault description is " Defect computer, The computer starts, power LED turns ON - flashes several times and then the laptop turns off".
The computer came without hard drive or memory.
Look at the video to see the problem.
Step 1: Tear Down the Laptop
Finding service manual was not so difficult. Generally opening the laptop was not so difficult. It is important to know how and what you take a part first so you can reverse the process.
The first observation I made when I opened the laptop was the conclusion "the laptop was opened before". I found the screw which hold the CPU mounted to the socket/motherboard abused. I found also some cooling paste at several places on the motherboard and on the plastic parts of the computer.. Bad start.
Step 2: Find the Problem and Fixing It
When I removed the heat sink ..I discovered that the processor was missing. It was not good as the description of the object when I bought it, did mention that the computer should comes with the CPU. Only hard drive and memory should be missing.
I managed to fix a CPU for the computer. But even with the CPU mounted to the motherboard, the computer had the same behavior. This indicated to me, the processor has no effect on the motherboard and it is not working. To prove that, I touched the CPU and it was cold. CPU was doing nothing.
I started to look at the motherboard for any physical damages. I found in a very short time the cause of the problem. There is a chip which is near to the CPU-socket which has a hole on the top side of it. The chip is called
EMC4001. It is responsible for the fan and thermal control. Replacing the chip resulted in fixing the startup issue.
But unfortunately, I found later, the computer has another problem which could be the real cause of the defective EMC4001. The fan wasn't working and it was damaged. I ordered a new fan.
Step 3: Laptop Is Working
Step 4: Ordering the Fan .. Fail
I got the fan today. Unfortunately, the fan is not compatible.
Be careful when you order spare-parts so you don't do the same mistake. They have almost the same model-number .. But they are not the same item!!
Please let me know if you have any idea why that number make the difference? The model number should be "GB0507PGV1-A", I guess!
That is what SUNON make it as their item number or model number .. How should I know what text is the correct model number?
bad luck !
Step 5: Problem With the FAN
When I replaced the EMC4001, I could start up the laptop and it was working, I thought that was the end of the story. But, I discovered very soon later that the fan was not spinning. Since I know, the EMC4001 is the responsible chip for this job, I didn't feel so great when I discovered the problem.
Looking for datasheet for the EMC4001 on the net gave me no results. It is a custom chip which is no longer (I guess) produced and the producer (Microchip) has no info about the chip on their website. But, there are other fan controllers, you can find on their home site. Their datasheet could be useful for understanding the pins you have on the EMC4001.
Later, I could find info about EMC4002. This chip is very similar to the EMC4001, but not totally. In fact the chip has the most important thing (that I will explain later) inside it which is missing in the EMC4001 and caused further damages on the motherboard.
Unfortunately, looking to some Youtube videos, they admit that laptop fans could be opened and cleaned (inside the case). Depending on this, I tried to open the fan (original), but I braked it into parts, It was not possible to fix it again (You see the fan). It has some electronics responsible for the TACH signal generation (tachometer signal). The signal gives a feedback to the computer (a chip called MEC5025) about the FAN's RPM. I will come back to this chip later.
First, I tried to re-flow the EMC4001 chip .. and checked all pins .. They were just fin. Later, I tried to measure all resistors (thermo-resistors) you find them near to each sensor.. and they all were just fin.
mmmm... What could be the problem.. ?
There is a way to reset the Bios for the thermal-control & FAN-functionality.. Using fn+z should reset the bios. I upgraded the Bios also for the latest firmware upgrade from DELL. Nothing helped in solving the problem.
I downloaded a program which gives you the ability to control the fan. From previous usage, I often used speedfan for checking S.M.A.R.T of hard drives .. But never used speedfan for controlling the fan (LOL the name is speedfan). So, it was the time to used it for that purpose. I could find another program which is (or might be) better for controlling the fan. It is called i8kfangui31.exe. You can search for that on the net .. easily you can find it.
Just to be mentioned here, the 5 volts (Vcc to the PWM system inside the EMC4001) which should drive the fan is there at pin 5&6. The only problem the computer has, EMC4001 is not giving the power to the fan on pin 7&8 .
As you see in the pictures, the speedfan shows the speed of the fan to be 1928 rpm. This number is not changeable. Even when you change the speed or disconnect the fan, you get the same number. Now, we know from the datasheet, the EMC4001 has no GPIO pins for the TACH signal since TACH siganl go to the page where MEC5025 is drawn. Ohhhhh ... which means, the MEC5025 has problem also.
The i/o pin might be damaged .. or actually it is damaged. I don't know the mechanism of how the MEC5025 control the EMC4001, But I am sure, the MEC5025 is also partially damaged, which we could confirm by the strange constant RPM number we get as speed of the FAN.
What a bad design to not include the Feedback signal inside the EMC4001 itself? If you get a problem with the fan, you will not get damages with one chip, you will get 2 defective chips, as in our case here.
That is very bad. MEC5025 chip is costing about €18. It is expensive, with regards to all money I spent on the computer.
I really don't know if there is a way to control the fan via the EMC4001 without having the real fan speed (skipping the GIOP and TACH siganl). I don't know if you can replace the mechanism available internally by the Bios and the MEC5025 so you do your own calculations for the fan speed and control it in that way.. You have all data (i.e. the sensors reading) and you can in theory calculate the speed you need to force the fan run at desired rpm. But how do you control the EMC4001? Is there any way to directly control it (writing a small program maybe?).
I will continue thinking about the problem. I might find a way to solve the problem .. I know, I can give the 5 volts directly from pin 5&6 to the fan, this will make the fan to run at high speed permanently. But, that is not the best way to solve the problem. I might use that option if I cannot find any other way to solve the problem.
This is enough for this step. I wanted to update you on the progress I have in troubleshooting this laptop. Please share with me your experience .. I will appreciate your comments.
Step 6: Last Step - Acceptable Solution?
I gave up searching further or putting more time on this repair attempt. Connecting two diodes with 0.7V drop voltages across each one gives you an output voltages about 3.5V. This make the fan run always but not in max speed. It works and I think it will be suitable if you don't run program which heat up the CPU/GPU. The 5 volts is from the pin which the EMC4001 gets for supplying the fan .. So it is the write source of current ..but with constant current -- not PWM.
I sold also the computer with no profit .. in fact with loosing some money :) . But my goal was not profit.. It was from the beginning "JUST FOR FUN". I hope you enjoyed this repair.
We have a be nice policy.
Please be positive and constructive.