AMD RS880M/HD4250 (216-0752001) - bad chip!

[Th3_uN1Qu3]

Re: AMD RS880M/HD4250 (216-0752001) - bad chip!

And another HP 625 done.. This board has good quality solder mask. After reballing it flashed a different code but still no display, so i reflashed the BIOS as well, and... it works.

I'll follow this one, for now my recommendation is to reball this chip if it looks green and healthy, replace if browned from overheating, or if you find the thermal rubber broken.

[sorinac]

Re: AMD RS880M/HD4250 (216-0752001) - bad chip!

also for this hp 625 u should remove white wire from fan or cut it so the cooler goes to maximum

[Th3_uN1Qu3]

Re: AMD RS880M/HD4250 (216-0752001) - bad chip!

I was never a fan of this modification... When the fan is maxed, in a short period of time it gathers a lot of dust and clogs the heatsink, so the same thing happens eventually. If i'm worried about it, i'll put a shim between the chip and heatsink, but i think just a replacement of the thermal rubber (which was a bit damaged btw) will work fine in this case.

[sorinac]

Re: AMD RS880M/HD4250 (216-0752001) - bad chip!

I don't like to modify either but some notebook have poor cooling system and require special treatment

[Th3_uN1Qu3]

Re: AMD RS880M/HD4250 (216-0752001) - bad chip!

The HP 625 actually ran at good NB temps, because the big CPU (2.4GHz Turion X2) made the fan run almost all the time.

I've also heard from the guy with the HP G6, and it's doing good after the NB reballing. So from now on i'll be telling my boss to leave that hot air station alone when it comes to laptops, and let me do the business.

[dellxps15]

Re: AMD RS880M/HD4250 (216-0752001) - bad chip!

have u ever encountered pop corning problem with ati chip? (umidity problem that blow chip when reballing)
so rs780 is also a big defective product ?

i have an hp with it and going to replace it and give it a try

[sorinac]

Re: AMD RS880M/HD4250 (216-0752001) - bad chip!

Info


I have a HP dv5 notebook and decided to test some RS780 chips to see if they are compatible.
Mainboard model : DA0QT8MB6G0 with discrete graphics
Original chipset : 215-0674034
Tested successfully : 216-0674022 , 216-0674026

The 216-0674024 not tested but I am sure it will work.

Also same chips i have tested in Asus K70AB with dedicated graphics and it worked.



Notice: In this types of notebooks the incorporated video from RS780 chip is not used

[jeremykum]

Re: AMD RS880M/HD4250 (216-0752001) - bad chip!

In my experience 216-0674026, 216-0752001 both are problematic chips. Very good demand in China, until it went out of stock from various suppliers of mine from Shenzhen, China last year about August.

Last year alone this model chip total up more than 200 pcs in my place. Mostly from Compaq CQ40 AU series, and CQ41 AU series. These 2 laptops are top selling models when it is launched here in Malaysia.

Datecode 13+ so far more reliable, but beware many are selling it cheap for reballed versions. There is actually 3 types available. (applies to almost all chips available)

1. Original New
2. Bulk New
3. Reballed (pulled chips)

1 is the best, factory packed, all similar datecode, dull very white solder balls and fresh green in colour. If you buy a strip of 10s, all has the same datecode, same colour.

2 is cheaper, (I suspect these are factory reject stocks), repackage by the sellers. You will find various datecode if you happen to buy a strip of 10s. Solder ball when compared isn't as white, and slightly less fresher green in colour. If you do not have both chips, it is very difficult to differenciate them. You buy 10, likely you get 6pcs that worked.

3 is rubbish. You buy 10, maybe less than 3 pcs usable. Dark green in colour, due to heat stress a few times in the process if pulling the chip from board, cleaning the balls underneath the chip, and reballing it. Sometimes you may even see the side/corner of the chip is crack.

If you are wondering how to define the datecode, it is printed on the chip. A 4 digit code. Example my 216-0674026 datecode is 1326, meaning Year 13 Week 26. If your chip delivered to you date 08xx/09xx/10xx is definitely not ORIGINAL NEW. Just beware. Some models Original New still date back to 10xx but it won't be these 2 hot sellers

Hopes everyone get smarter and don't fall prey to irresponsible sellers.

[spleenharvester]

Re: AMD RS880M/HD4250 (216-0752001) - bad chip!

Might also be worth noting there is a widespread issue with this chip in the Dell M5030s.

[Th3_uN1Qu3]

Re: AMD RS880M/HD4250 (216-0752001) - bad chip!

Not only the Dell M5030 - it also fails often in the HP 625.

[jeremykum]

Re: AMD RS880M/HD4250 (216-0752001) - bad chip!

But if we look at the failure rate, ATI and Nvidia has shorter lifespan compared to intel. I had read some article previously indicating the shorter lifespan was due to technology that intel owns in manufacturing the chip. ATI and Nvidia doesn't have access to those technology thus it is not as reliable as intel.

I begin to notice that these ATI or Nvidia begins to fail when after 2 years of usage. However intel SLGZS chip seems to have flaws too, very common it failed in a variety of laptops' brand. So this eliminates the options due to board design.

[Th3_uN1Qu3]

Re: AMD RS880M/HD4250 (216-0752001) - bad chip!

Yes i've heard about intel's manufacturing technology being superior... Also as far as i can remember, intel chipsets allow higher reflow temperatures so they create stronger joints. But so far, i've noticed that many of the failed laptops i've had to fix had one thing in common - really tiny pads along the outside of the chip.

Latest one is a Lenovo Y560, with an ATi 5730 (216-0772003). This one died not of overheating but of mechanical reasons - the bottom case was cracked inside and the hinges were loose. The left hinge foot is under the edge of the mainboard, and every time the display was opened, it pushed the board and flexed it (btw, this laptop has a VERY weird-shaped board). So it flexed until the joints broke on the graphics chip. This too had really tiny pads on the outside, where all the signals are.

This laptop was very lucky - all this board flexing by the metal hinge base chipped a corner out of one of the VRAM chips, but it only damaged the plastic. After i reballed the GPU, i glued the hinges to the bottom case with 2-part epoxy to eliminate the cause.

[spleenharvester]

Re: AMD RS880M/HD4250 (216-0752001) - bad chip!

Fails on Toshiba L675D and similar models too. I think you're definitely onto something here.

[Dimitrios]

Re: AMD RS880M/HD4250 (216-0752001) - bad chip!

I have a shorted ATI 216-0752001 northbridge/GPU power supply on a Toshiba but this was due to milk spill and I think it burned the chip. I removed it now completely and is still shorted, so I guess there is a shorted capacitor somewhere. Anyway, it is a very interesting thread here, well done Unique! Thanks you all for sharing these useful information. Its best to help each other in this rare field with misleading info!

[jeremykum]

Re: AMD RS880M/HD4250 (216-0752001) - bad chip!

By the way, out of curiosity, for those who owns and using IR Rework Station. What do you actually think of reflow and reballing makes a different?

Personally, I only reball back the chip whenever the new chip fitted but board is dead. I never reball a chip that the laptop comes in dead due to the BGA chip.

This is my understanding, but do correct me if I am wrong though. I had written some of my understanding here in my fb page.

https://www.facebook.com/ChipwerksRe...&stream_ref=10

Usually I will set the reflow temp that actually melts the balls till it's shining. Adjusting between 220 to 225 and the durations only. I mean until the balls changes it's state towards liquid (forgot the terms that is used though). This method saves a lot of time needed to reball. And doing reflow I don't use paste flux. I use liquid flux. Paste flux only when I fit new chip on to the board.

[Th3_uN1Qu3]

Re: AMD RS880M/HD4250 (216-0752001) - bad chip!

What do you actually think of reflow and reballing makes a different?

In the ideal world, if the problem is indeed from the solder joints cracking and the chipset is not damaged, a proper reflow should be enough.

However, in the real world, the following happens:

• Computer starts acting crazy. Locking up, BSODing and so on. Sometimes it works, sometimes it does not.
• User tries various things. Apparently successful, computer works for a few more days.
• Computer dies entirely. After many more unsuccessful attempts to get the thing running again, user gives up and brings it in to a service center.

Running curent thru the cracked BGA balls makes the cracks oxidize. The more the user tries to power on the machine again, the worse it gets. So even if you completely melt the solder when you reflow it, there will be oxide trapped inside, weakening the solder, and the problem will return. This is why reballing is required.

[Dimitrios]

Re: AMD RS880M/HD4250 (216-0752001) - bad chip!

From my experience all reflows are due to fail after 1-3 months. At least with my equipment: hot air and preheater. I can do a proper reflow with insat flux but they will definitely fail after a while. Only reballing the chip is a permanent fix but many things could go wrong in the proccess. It has many steps and you must know what you are doing.

[jeremykum]

Re: AMD RS880M/HD4250 (216-0752001) - bad chip!

Running curent thru the cracked BGA balls makes the cracks oxidize. The more the user tries to power on the machine again, the worse it gets. So even if you completely melt the solder when you reflow it, there will be oxide trapped inside, weakening the solder, and the problem will return. This is why reballing is required.

Noted about the oxide. So there isn't a way to get rid of the oxide without the tedious reballing work...

From my personal experience though, I seldom practice reballing chips pulled from boards that the chip initially failed, unless the owner requested. However, I do notice that using higher temp for reflow has a very high success rate of the chip lasting above 5 months, for those chips that still live after the reflow process. Some will have the core bursting out solders. Means their life is up, and goes into my collection bin.

Thanks for sharing your experiences and knowledge, hope this will make us all wiser and face less problems in trying to relive a dead laptop.

[jeremykum]

Re: AMD RS880M/HD4250 (216-0752001) - bad chip!

From my experience all reflows are due to fail after 1-3 months. At least with my equipment: hot air and preheater. I can do a proper reflow with insat flux but they will definitely fail after a while. Only reballing the chip is a permanent fix but many things could go wrong in the proccess. It has many steps and you must know what you are doing.

Yes, I do agree. I experienced it too, some even dead in a week time (same results even with IR). Most of it died within the first 4 weeks. Here I am just referring to CQ40-AU and CQ41-AU, since I had too many of these models here. Its AMD 216-0674026 and 216-0752001.

Certain times, due to customer budget, I will do a high temp reflow. Purposely to stress the core. If it live through the process, the chip last for months to come. Or dead on the spot with solders spitting out from the core joints. I accidentally found this method while experimenting with temp profile. So it makes me think that it also could be the issues with the bonding of the core to the chip, not only cold joint between the chip and the board. Well, I am just trying to give my self a logical explanation only, as I couldn't find much info about this topic in google search.

[Dimitrios]

Re: AMD RS880M/HD4250 (216-0752001) - bad chip!

This area of reflowing and reballing chips is a nightmare. You must gain you own experience with your equipment... You must reballing chips all day to see temps and bending boards!! Lol! Everytime I do a reballing I am thinking that next time will be better as long as won't use too much flux or support the motherboards cause it gets bended from heat and so on...