High Death Rate for Intel Boards?

[bgavin]

I seem to be getting a lot of dead Intel boards. None have bulged caps. I recapped two of them, and they remain dead.

Is this a trend with Intel boards, or just the luck of the draw?

[willawake]

Re: High Death Rate for Intel Boards?

model nos. pls

[PCBONEZ]

Re: High Death Rate for Intel Boards?

Yup.
What board(s)

These may have been some of those Intel boards with the HN HM problems that were recapped but after some damage was done to other components.

Also do you see any KZG on them.

.

[PCBONEZ]

Re: High Death Rate for Intel Boards?

To answer you question.

Except for the HN HM issue bad Intel boards are (as compared to other brands) rare.

.

[willawake]

Re: High Death Rate for Intel Boards?

i wasnt that pleased with the newer models. i mean its ok with the bad axe but with the other boards they could sink the fets. the ich8 is also running too hot but i guess that is a feature of all the boards with that chipset. still they did not sink the ich8 with large enough sink in my opinion on the board i tested.

but in the end i guess it depends on the price. some of the intel boards are very cheap.

then again if your clients have many intel boards then more likely to see fails.

[PCBONEZ]

Re: High Death Rate for Intel Boards?

I haven't seen the newer models.
I prefer to wait until new motherboards have been out for 3 or 4 years before I'll buy one.
Let someone else discover all the problems.
Read all the reviews.
And then usually end up with something with bad (*&$^@*^ caps anyway.. LOL

.

.

[kc8adu]

Re: High Death Rate for Intel Boards?

i have seen a ton of the intel seabreeze boards bad.
half were hn/hm and the rest bestec psu vicims in evilmachines.

[bgavin]

Re: High Death Rate for Intel Boards?

The Intel boards are from Dell, Gateway, Compaq, and eMachines.

D845HV, D915GAG, are the two that I can reach fast and easy. The rest are buried in the bin awaiting the Second Coming, a Winning Lottery Ticket, and a date with Anna Nicole Smith.

The two I can see show no visible signs of damage. All have crap PSUs installed. The D845HV has a Newton 160w supply, no visible bulging inside. The board did power up and run Memtest for a long time, but after shutdown, it won't come up again. Installed a known-good quality PSU, still DOA.

The D845HV has LXZ, and two other types of undetermined heritage. I hate to recap this board, when I can buy a working one with processor for $35. I can recap the working one and not worry about wasting caps on a dead board.

[zandrax]

Re: High Death Rate for Intel Boards?

The Intel boards are from Dell, Gateway, Compaq, and eMachines.

D845HV, D915GAG, are the two that I can reach fast and easy. The rest are buried in the bin awaiting the Second Coming, a Winning Lottery Ticket, and a date with Anna Nicole Smith.

Rotfl!

The two I can see show no visible signs of damage. All have crap PSUs installed. The D845HV has a Newton 160w supply, no visible bulging inside. The board did power up and run Memtest for a long time, but after shutdown, it won't come up again. Installed a known-good quality PSU, still DOA.

The Newton shouldn't be a crappy psu, just an underpowered one: 160W are barely enough for a Pentium 4 with, perhaps, a low-lewel video card, an hard disk and a cd player.

Zandrax

[PCBONEZ]

Re: High Death Rate for Intel Boards?

Oh, - That level of "newer" boards.
[Is okay, that's what 'newer' is to me too.... LOL..]

Some i845-i865-i875 Intel boards suffered from the Nichicon HN and HM series cap fiasco. - As I recall there was a second advisory that added a couple of i915 boards but I can't find it right now so I might be mistaken.

http://www.intel.com/cd/channel/rese...tes/193414.htm

It's not an "Intel Issue".
The HN and HM issue affected Asus, FIC, Gigabyte, Abit, Tyan, MSI. Even Apple/MAC took it in the shorts on this one. It is also behind some of the Dell, HP, and MAC G5 recalls.
- I think every MoBo manufacturer got caught on a board or two (or 10) because Nichicon makes good quality caps.
- This 'plague' was due to a screw up at the factory and not a design defect or a low quality materials issue.

The affected faulty caps were manufactured from 2002 to 2004 and only from one of the fab facilities. Some 2002-2004 HN HM are okay but there in no way to tell by looking at them.
-
HN and HM caps date coded 2002-2004 are ????.
Those date coded before 2002 or after 2004 are okay.

Date code is [single letter][two digit year][two digit week of year]

I think there is something similar going on with Chemicon KZG series caps but Chemicon either doesn't know yet (it does take like 2+ years 'in the field' for cap issues to be figured out). Or possibly Chemicon knows but chooses to keep it a secret. .

.

[PCBONEZ]

Re: High Death Rate for Intel Boards?

Now.
The caps were made from 2002-2004 but some idiot companies that weren't paying attention were still using boards with them as late as 2006.
-
http://www.asti-usa.com/support/appnotes/65.html#b

.

[bgavin]

Re: High Death Rate for Intel Boards?

I have three D845HV boards now.

The first one came in with a dead Gateway machine. I found two more of the same board on eBay for $39 each (and with a free AGP card). These arrived safely packed in anti-static bags and worked correctly on the bench. One had KZG, the other HM on board.

Recapped the working board, and put it into the customer machine along with a new PSU. I am recapping the other working board today and will put it into stock or for my own use.

I recapped the dead board and brought it back to life. I normally don't like to spend 30+ caps on a non-working board, but this only had four 3300/6.3 on the VRM so I took a chance and it worked. The field caps are various tiny ones that I left untouched. The ATX12V filter caps are Rubycon ZL.

[willawake]

Re: High Death Rate for Intel Boards?

Now.
The caps were made from 2002-2004 but some idiot companies that weren't paying attention were still using boards with them as late as 2006.
-
http://www.asti-usa.com/support/appnotes/65.html#b

.

# To initiate the return and replacement procedure, a PO must be issued to ASTi in the amount of $1000 for each effected Telestra. However, the net price with return of original motherboard is $0. Do not list the value on the shipping documentation.

whats that?

[PCBONEZ]

Re: High Death Rate for Intel Boards?

What's what?

ASTi?
Genaric/Contract PC builder that was still using bad HN/HM boards in 2006.
Took the 2 years longer than everyone else 'to hear about it'.
.

[bgavin]

Re: High Death Rate for Intel Boards?

I went out eBay hopping for another board for an eMachine T2885 (Intel Seabreeze, D845GVSR).

It appears all these boards die a similar death, i.e. will not power up at all. Evidently the very under-powered PSU used by eMachines contributes to board death. I found a large number of occurrences.

Curious: is this a crap board, or a common symptom of DBE (death by emachine)

[willawake]

Re: High Death Rate for Intel Boards?

What's what?

ASTi?
Genaric/Contract PC builder that was still using bad HN/HM boards in 2006.
Took the 2 years longer than everyone else 'to hear about it'.
.

whats a PO for $1000?

Curious: is this a crap board, or a common symptom of DBE (death by emachine)

i have two 845 intel in the office and rock solid for years. just dont have sata so not that interesting anymore.

[willawake]

Re: High Death Rate for Intel Boards?

Intel BOXDG33TLM LGA 775 Intel G33 Micro ATX Intel Motherboard
gets some mixed reviews on newegg. Quite a few RMAs. This board is interesting as it was sold as a very cheap package for intel retailedge program enabling retail employees working for participating stores to get a quad, this board and vista for extremely cheap as long as they did many quizzes on the site.

its a pity that. also concerning about the current intel boards is remarks that the southbridge is too hot. 70-80oC. One reviewer mentioned a drop to 50oC after putting some cheap chipset heatsinks that were more effective. although i do have one recent intel board in the office with similar issue that is completely stable. i think that is ICH8. i would have hoped they would not persisted with this on more recent boards.

this is quite annoying because i would like a new work computer and i fancy an intel board.

[zandrax]

Re: High Death Rate for Intel Boards?

its a pity that. also concerning about the current intel boards is remarks that the southbridge is too hot. 70-80oC. One reviewer mentioned a drop to 50oC after putting some cheap chipset heatsinks that were more effective.

Well, I see an heatsink: I'd put a 40 mm fan and check if the termal compund is good enough because the stock heatsink seems better than cheap ecs-style ones (these were fine up to P3 era chipset, but after ...).
I agree 70 C is a too high temperature: it should never be more than 50-55 C.

Zandrax

[Bookworm]

Re: High Death Rate for Intel Boards?

I found that the board death in E-machines was generally not the power supply directly, but the fact that the power supply was vulnerable to power hits. The E-machines I put on even CHEAP UPSs never lost a power supply. The ones in power strips often lost two or three.

[bgavin]

Re: High Death Rate for Intel Boards?

I spent a few days putting together a big spread sheet of Intel processors and boards. After that, I dug around to see which were dogs. I came up with the Northwood core as being the most desireable in P4. The matching board is the D875PBZ which has a long reputation for stability.

Both are available on eBay, so I scored a 30-cap version of the 3.4 GHz P4 in Northwood/800FSB. There are many D875PBZ boards available, including a guy selling 100+ new ones. Price is stiff at $169 USD, but they are available. There are many used ones also, so I will pick on up in a few days. I figure two sticks of DDR800 will complete the package. Then I have to find an AGP 8x (non-gaming) card, if one exists. This should give me a nice business machine with plenty of poop.