Asus P4C800e

  • Thread starter Thread starter La Pedrera
  • Start date Start date
L

La Pedrera

Anyone here famaliar with the solder problem on the bottom of the board with
a tight heat synch?

I'm wondering if this has been fixed in any board revision??
 
La Pedrera said:
http://www.hardwareanalysis.com/content/topic/7272/

Link with info for the uninformed. :)
Click to expand...

This is not a "revision" issue, but a "manufacturing lot" issue.
Identifying the troublesome boards would only be possible by their
serial numbers. All it takes is a change in the order of assembly
of the product to fix the soldering problem (i.e. solder, then
install the bracket). The revision number printed on the board
will not change.

The problem is not a "tight heat sink". There is a blob of
solder that comes close to bridging two conductors. A little
extra pressure in the area is enough to close the gap and
cause a short. If the excess solder is removed, you'll be able
to apply as much pressure as you want. Since the solder joints
in question will take a lot of heat to melt, you would need a
soldering iron larger than the 25-35 watt irons usually used
for this work. Touching the board with a soldering iron will
void the warranty, and the factory will have a soldering iron
large enough to do the job properly.

So, there is virtually no way for you to distinguish the good
from the bad boards. You can only hope that Asus had the good
sense to inform their distributers of the bad batch of boards
and have them returned to the plant for rework. (But the chances
of that happening are zero...)

You can do what I did when I purchased a certain Asus board - I
went to a local store, and insisted they open the box for inspection
of the contents. Have the clerk handle the board while you look
for whatever design issue you are interested in. Since the place
I purchased from also build systems, this board happened to be one
they were using for system builds, so taking it out of the box
didn't bother them too much.

HTH,
Paul
 
This is not a "revision" issue, but a "manufacturing lot" issue.
Identifying the troublesome boards would only be possible by their
serial numbers. All it takes is a change in the order of assembly
of the product to fix the soldering problem (i.e. solder, then
install the bracket). The revision number printed on the board
will not change.
Click to expand...

Well, after speaking with Asus today, they have not accepted this as a known
issue for this board yet. I'm sending the board back to them, and hopefully
the replace it.
The problem is not a "tight heat sink". There is a blob of
solder that comes close to bridging two conductors. A little
extra pressure in the area is enough to close the gap and
cause a short. If the excess solder is removed, you'll be able
to apply as much pressure as you want. Since the solder joints
in question will take a lot of heat to melt, you would need a
soldering iron larger than the 25-35 watt irons usually used
for this work. Touching the board with a soldering iron will
void the warranty, and the factory will have a soldering iron
large enough to do the job properly.
Click to expand...

On my board and some other ones I have seen, it * is * a "tight heat synch"
that causes the shorts to occur. EG - I had one type of heat synch, then
upgraded to a Thermaltake Xaser, which caused the board not to post. Go back
to the other one, posts fine. :)
So, there is virtually no way for you to distinguish the good
from the bad boards. You can only hope that Asus had the good
sense to inform their distributers of the bad batch of boards
and have them returned to the plant for rework. (But the chances
of that happening are zero...)

You can do what I did when I purchased a certain Asus board - I
went to a local store, and insisted they open the box for inspection
of the contents. Have the clerk handle the board while you look
for whatever design issue you are interested in. Since the place
I purchased from also build systems, this board happened to be one
they were using for system builds, so taking it out of the box
didn't bother them too much.
Click to expand...

I'm just hoping to pass this on to others incase they experience the
problems. The more people that are aware, hopefully Asus will recognize the
problem and acknowledge it.
 
La Pedrera said:
Anyone here famaliar with the solder problem on the bottom of the board with
a tight heat synch?

I'm wondering if this has been fixed in any board revision??
Click to expand...

I bought one of these things with the bad solder, wound up with a DOA board.
I was able to get it to light up the monitor briefly if I started up w/o the
heatsink on, but it would crap out after a couple of seconds. I wound up
RMA'ing it for a replacement, and while I was waiting for the new one, I
went out and bought a different kind of heatsink that mounts directly on the
MB. Since I had to remove the stock mounting bracket to use the new
heatsink, I was able to have a look at the place where the bad solder was.
The new board looked clean, so I don't know if it was a bad run or just
piss-poor quality control and finishing in the manufacturing process. At
any rate, the new board runs great, and is nice and tweakable.

I couldn't get ASUS to admit to the solder problem either.

Good luck with yours.
 
I couldn't get ASUS to admit to the solder problem either.
Click to expand...

It may or may not be relevant, but Intel warns against using the metal
braces on the underside of the board. They are supposed to eliminate
flexing, but the flexing is normal and expected.
 
jaeger said:
It may or may not be relevant, but Intel warns against using the metal
braces on the underside of the board. They are supposed to eliminate
flexing, but the flexing is normal and expected.
Click to expand...

I think the problem is there are two pieces of solder that are just
barely touching one another. If they actually joined to one another,
testing at the factory would reject the board. The boards that slide
through the testing don't have a solid short circuit. All it takes
is mechanical distortion of the PCB, such as happens when you install
a HSF, that causes completion of the connection. The bracket is plastic
and when the board is wave soldered, the bracket causes solder to
"drag" to one side. That is where the blob comes from. Wave soldering
with the bracket missing, would result in a good solder joint and
no problems. Then, the Asus assemblers could put on the bracket, after
soldering is complete.

Paul
 
I think the problem is there are two pieces of solder that are just
barely touching one another. If they actually joined to one another,
testing at the factory would reject the board. The boards that slide
through the testing don't have a solid short circuit. All it takes
is mechanical distortion of the PCB, such as happens when you install
a HSF, that causes completion of the connection. The bracket is plastic
and when the board is wave soldered, the bracket causes solder to
"drag" to one side. That is where the blob comes from. Wave soldering
with the bracket missing, would result in a good solder joint and
no problems. Then, the Asus assemblers could put on the bracket, after
soldering is complete.
Click to expand...

I used to work at a production plant with a wave solder and the boards went
through with no brackets (such as the socket bracket) installed. Those were
mounted after.

After taking the bracket of my board, it was as though it was compressed by
the socket. (clean edge where the bracket caused it).

I'm taking it for RMA tomorrow (from point of purchase).

I plan to take some pics, and can post em here when I get em.
 
La Pedrera said:
Anyone here famaliar with the solder problem on the bottom of the board with
a tight heat synch?

I'm wondering if this has been fixed in any board revision??
Click to expand...
Heat SYNCH? :-)
 
When I had three GB ram, BIOS and Windows reported 3GB. After installing the
4th stick, Windows now reports that I have 3.49GB. According to the manual,
I should have 'somewhat' lesss than 4GB, but I am getting only 50% out of
the 4th stick. Is anyone else running this mb/mem combo? How much memory
does Windows report? (I am running 4x1G PC2100 memory with a P4/1.6A
processor, all of which was left over from my prior system when that mb went
bad. I have thus far upgraded only my mb.)
 
Yup. That is right. Unfortunately, yo udo loose memory. It's reserved for
other resources. Can't do a thing about it. (I'm in the same boat with that
motherboard)

My previous (Gigabyte 81NXP) only lost 256MB, reporting 3.75.
 
"Z Man" said:
When I had three GB ram, BIOS and Windows reported 3GB. After installing the
4th stick, Windows now reports that I have 3.49GB. According to the manual,
I should have 'somewhat' lesss than 4GB, but I am getting only 50% out of
the 4th stick. Is anyone else running this mb/mem combo? How much memory
does Windows report? (I am running 4x1G PC2100 memory with a P4/1.6A
processor, all of which was left over from my prior system when that mb went
bad. I have thus far upgraded only my mb.)
Click to expand...

The 875P Northbridge chip spec has all the details, but basically the
system memory map needs some room for PCI/AGP device mappings. I have
no idea how Windows or the BIOS handle things. My guess would be the
BIOS computes how much PCI/AGP bridge address space is needed and
programs the MBASE and similar registers to cut off the top of memory.

Paul
 
Back
Top