PSU's on the top of the list

Archived from groups: alt.comp.hardware.pc-homebuilt (More info?)

Well interesting developments, in the last 6 clients I have been
working with, PSU's have failed or were failing. None are from my
builds and most were 1- 2 years old. This has been the least frequent
problem for me and customers for the last 2 years or more, but now
this is a serious development.

None of the PSU's are alike. Some were cheapies and some were
sparkle and antec. 2 AMD's and 4 Intel. all different cases. This is a
concern for me now, as prior I had no reason to be that concerned.
What is interesting is that the PSU's in question were all manufatured
within 3 months of each other. I now (sigh) have to look at each one
of these and find if there is a common cause. I really hate doing this
tedious work, but this is an epidemic, at this point and time.


I have been always on the side of calm in the posts that were
'excited' about PSU's. Now I have to say although I am not panicy
about it, I have changed my way of thinking.

Archived from groups: alt.comp.hardware.pc-homebuilt (More info?)

On Fri, 26 Nov 2004 11:26:16 -0800, JAD wrote:

> Well interesting developments, in the last 6 clients I have been
> working with, PSU's have failed or were failing. None are from my
> builds and most were 1- 2 years old. This has been the least frequent
> problem for me and customers for the last 2 years or more, but now
> this is a serious development.
>
> None of the PSU's are alike. Some were cheapies and some were
> sparkle and antec. 2 AMD's and 4 Intel. all different cases. This is a
> concern for me now, as prior I had no reason to be that concerned.
> What is interesting is that the PSU's in question were all manufatured
> within 3 months of each other. I now (sigh) have to look at each one
> of these and find if there is a common cause. I really hate doing this
> tedious work, but this is an epidemic, at this point and time.

Bad power stressing those supplies?

> I have been always on the side of calm in the posts that were
> 'excited' about PSU's. Now I have to say although I am not panicy
> about it, I have changed my way of thinking.

Consider also that the business goal of "value engineering" is to reduce
costs (and thereby increase profits, always in short supply on low margin
items like PCs). Ideally, the units should fail just after you have
decided to throw them out. That then becomes a statistical estimate, with
some variance (including "infant mortality" ). It is possible that there is
a general trend for power supply lifetimes to decrease, maybe to match the
tech churn/treadmill of some upgrading their PCs every couple of years?

Maybe also common manufacture changing venue? from Japan and/or Korea to
new shops in India and/or China? Might be some kind of process calibration?

One could ask "what do you expect from a $29 retail PSU"? Then (your
question would be) is a $89 retail PSU actually 3x better? in what way(s)?

Might be an opportunity for you, in terms of maintenance business, as long
as you are not affected by any increased business/liability risks?

--
Juhan Leemet
Logicognosis, Inc.

Archived from groups: alt.comp.hardware.pc-homebuilt (More info?)

"JAD" <Hrhackthespam@witchiepoo.broom> wrote in message
newsxLpd.3717$XF2.1811@fe06.lga...
> Well interesting developments, in the last 6 clients I have been
> working with, PSU's have failed or were failing. None are from my
> builds and most were 1- 2 years old. This has been the least frequent
> problem for me and customers for the last 2 years or more, but now
> this is a serious development.
>

I've seen 3 in a couple of months, and 2 internal psu faults, both on 24
port switches.
2 of the psu failures showed clear signs of capacitor failure, no
visible signs on the third psu.
Both switches were also cap failure, and replacement caps solved the
problem in both cases - both emergency repairs, explained the risks and
got disclaimer signed by customer.
All system psu replaced with Antecs, I explained that I don't have the
required test gear to repair a system psu with confidence.

> None of the PSU's are alike. Some were cheapies and some were
> sparkle and antec. 2 AMD's and 4 Intel. all different cases. This is a
> concern for me now, as prior I had no reason to be that concerned.
> What is interesting is that the PSU's in question were all manufatured
> within 3 months of each other. I now (sigh) have to look at each one
> of these and find if there is a common cause. I really hate doing this
> tedious work, but this is an epidemic, at this point and time.
>
>
> I have been always on the side of calm in the posts that were
> 'excited' about PSU's. Now I have to say although I am not panicy
> about it, I have changed my way of thinking.
>

The cap problem seems to come and go, a while back it was all
motherboards but I haven't seen one for a while, now other devices are
showing the signs. I wonder how long we've got before we start seeing
large numbers of domestic appliances failing?, virtually everything
electrical / electronic has some electrolytic caps in it.

--
Ian

Archived from groups: alt.comp.hardware.pc-homebuilt (More info?)

"AMD'r" <*******@***mail.com> wrote in message news:<2n1qd.82109$Le1.1686227@news20.bellglobal.com>...

> Here's an interesting article on the capacitor problem plaguing psu's, mb's
> and other electronics... http://www.niccomp.com/taiwanlowesr.htm. Seems
> there was a bad batch of electrolyte used to produce millions of caps and
> now they're failing and problems are showing up in many different components

I have a Powmax/Raidmax PSU made in 1999 with cheapo JEE or JGE caps
that are still fine (appearance and test results), while an Antec made
2-3 years ago developed a couple of bulging caps (caught in time).
The latter worries me because I have some Fortron/Sparkle PSUs
(considered an excellent brand) with the same Fuhjyyu brand caps,
although they're much larger in volume than those of the same voltage
and capacitance in the Antec.