[OT] PC Hardware Problem

It's been ages since I kept current with PC technology, so I wanted to run this by some of you, to see if it lights any bulbs.
One of my boxen runs for a while, then (in Linux at least) kernel panics and resets (in Windows it resets, but I haven't stood over it to know if Windows notices the problem). My kid and I were working on it today to reinstall Ubuntu on the theory that the software was just royally screwed, which is when I noticed the kernel panicking.
It acts like a thermal problem -- leave it off for a long time and it takes a long time to have a problem, use it a lot and it happens much more often. All the fans work, and at one point I was able to monitor the various system temperatures which showed OK, so it's not something simple like the processor overheating.
At this point I'm about ready to start swapping parts, but part-swapping costs $$, so I thought I'd ask the group if these symptoms sound familiar, and if you found out anything specific to go with them.
--
www.wescottdesign.com

Add pictures here
<% if( /^image/.test(type) ){ %>
<% } %>
<%-name%>
Add image file
Upload
Tim Wescott wrote:

I had the same problem. Starting a few weeks ago my computer started freezing up mainly after I would leave it(say in over night). It turned out that the heatsink compound was dried up... fixed that and it's been running fine ever since(about 2 weeks).
Of course it could potentially have been something else but that seems to have been the issue. What happened was the thermal compound was relatively dry and I guess wasn't making good enough contact and would eventually cause the thermal sensor to trip(most modern CPU's have a shutdown mode to prevent damage).
I was monitoring the temp too but since it always happened when I was off(except the last few times) I never knew what was going on and imagined it couldn't be overheating when I wasn't on it(since it was basically in idle) but after replacing the compound no issues at all.
Anyways, it's worth a look...
It could be the memory or PS... usually one of those is the issue(Which is why I figured it was my memory since I have a monster PS).
Add pictures here
<% if( /^image/.test(type) ){ %>
<% } %>
<%-name%>
Add image file
Upload
On Sun, 21 Feb 2010 01:37:10 -0600, Jon Slaughter wrote:

What did you use for heat sink compound? Just the usual white silicone goo like I may find in my 30-year-old tube? This has the magic melting elastomeric stuff that came with the CPU.
We've replaced the power supply, and drives, and played 'swap the memory' games -- still does it. The caps on the mobo look good, so either it's a bad cap that's not visibly bad, or it's a CPU fit issue.
If I have good heat sink compound I think I'll give that a go.
--
www.wescottdesign.com

Add pictures here
<% if( /^image/.test(type) ){ %>
<% } %>
<%-name%>
Add image file
Upload
Tim Wescott wrote:

Ok, Tim, here is how to locate the problem...
Make a cone of paper that will fit over a component to be tested. Big end up - little end fits the device to be tested. Printer paper and tape work fine. You'll probably wind up with several odd shaped cones for computer parts.
Use a hair dryer to blow WARM air onto the part for a few seconds to try to fail a part.
Use a freon can the same way to try to recover a part.
Try to control your spray area carefully so as to affect only one part at a time.
With computers this is more difficult because once a computer goes crazy it must usually be cooled and restarted before it will run right. So you probably want to plan an attack that keeps parts cool to prevent the crazies rather than cause them.
Old light aircraft autopilots and other avionics are about the only thing expensive enough these days to warrant fixing rather than just replacing. That's where I learned this trick.
--

Richard Lamb
http://www.home.earthlink.net/~cavelamb /
  Click to see the full signature.
Add pictures here
<% if( /^image/.test(type) ){ %>
<% } %>
<%-name%>
Add image file
Upload
On 2/21/2010 11:19 PM, cavelamb wrote:
(...)

(...)
Yup. We used to call Freon 11 "Tech-in-a-can".
--Winston
Add pictures here
<% if( /^image/.test(type) ){ %>
<% } %>
<%-name%>
Add image file
Upload
On 2/22/2010 10:43 AM, Winston wrote:

Only problem with freeze spray is that if there is high humidity, you end up frosting good circuitry that upon melting starts flaking out and maybe even developing corrosion and conductive tracks then you waste time figuring out what just happened. I am not big on that method here in humid Florida.
--
Joe Leikhim K4SAT
"The RFI-EMI-GUY"
  Click to see the full signature.
Add pictures here
<% if( /^image/.test(type) ){ %>
<% } %>
<%-name%>
Add image file
Upload
On 2/23/2010 8:43 PM, RFI-EMI-GUY wrote:

We always kept the lab comfortable WRT both temperature and humidity. Never saw an issue caused by condensation, probably because it dried quickly and there wasn't anything it could pick up that would've made it conductive.
I feel lucky that I was out of that business by the time that lead-free solder became prevalent. I don't like that stuff. :)
--Winston
Add pictures here
<% if( /^image/.test(type) ){ %>
<% } %>
<%-name%>
Add image file
Upload
Winston wrote:

completely concur!
Fortunately I still have several pounds of 70/30.
--

Richard Lamb
http://www.home.earthlink.net/~cavelamb /
  Click to see the full signature.
Add pictures here
<% if( /^image/.test(type) ){ %>
<% } %>
<%-name%>
Add image file
Upload
On 2/24/2010 12:22 AM, cavelamb wrote:

(...)
Just checked the price of my fav 63/37 at ~25.00 for a 1 lb roll. Wow! Should have bought 'solder futures'.
--Winston
Add pictures here
<% if( /^image/.test(type) ){ %>
<% } %>
<%-name%>
Add image file
Upload

Who knew that the valuable heavy metals would be lead, tin and copper?
jsw
Add pictures here
<% if( /^image/.test(type) ){ %>
<% } %>
<%-name%>
Add image file
Upload
On 2/24/2010 7:16 AM, Jim Wilkins wrote:

The stuff is inflation proof, apparently. Huh.
--Winston
Add pictures here
<% if( /^image/.test(type) ){ %>
<% } %>
<%-name%>
Add image file
Upload
wrote:

In US dollars/metric ton
Lead is $2,290/ton Tin is $16,975.00/ton
Assuming the 63/37 is by weight, the weighted average is 11,500 per tonne, or $5.23/pound. So the metal is only 20% of what you are buying.
Add pictures here
<% if( /^image/.test(type) ){ %>
<% } %>
<%-name%>
Add image file
Upload
On 2/24/2010 12:57 PM, Spehro Pefhany wrote:
(...)

Logically then, the solder I've been storing for years cost only 1.60/pound in raw metal. Of course, 1980 dollars are not the same as 2009 dollars.
I guess there's value added in the alloying, extrusion, drawing processes, then profit down the retail chain.
The machine that makes the stuff must be interesting. I've never seen one but I envision a couple coaxial stainless funnels that continuously extrude the solder (with its flux core). Perhaps rows of rollers follow to draw it to final size. I imagine much of the operation would be conducted under a inert gas blanket to exclude oxygen.
--Winston
Add pictures here
<% if( /^image/.test(type) ){ %>
<% } %>
<%-name%>
Add image file
Upload

    Perhaps part of the cost is some kind of ROHS penalty fee.
    [ ... ]

    I would find the machine which makes the Ersin Multicore solder even more intersting. IIRC, it is five rosin cores whose centers form a circle about half the diameter of the whole solder.
    At a guess -- slitting from the sides, squeezing the rosin into the slits, then swaging the slits closed again over the solder.
    Is Ersin Multicore still made?
    Enjoy,         DoN.
--
Email: < snipped-for-privacy@d-and-d.com> | Voice (all times): (703) 938-4564
(too) near Washington D.C. | http://www.d-and-d.com/dnichols/DoN.html
  Click to see the full signature.
Add pictures here
<% if( /^image/.test(type) ){ %>
<% } %>
<%-name%>
Add image file
Upload
On 2/24/2010 7:56 PM, DoN. Nichols wrote:
(...)

I guess the wire *could* be extruded 'open' and then swaged over the flux, too. Envision a gear with 5 teeth.
I guess that means single-core solder *could* start out as an extruded flat ribbon and the first die actually wraps it around the flux core sort of like a hot dog bun.

Yup. Though I think they dropped the 'Ersin' name. <http://www.newark.com/multicore-solder/mm00975/502-cored-wire-solder/dp/32M7025
I bet flux smells exactly the same though. :)
--Winston
Add pictures here
<% if( /^image/.test(type) ){ %>
<% } %>
<%-name%>
Add image file
Upload

    Yes. That would make sense.

    :-)
    Which hot-dog bun? The New England style, which is slit down the top (and baked joined with several others on the sides), or the more widespread US one which is baked as individuals, and slit on the side?
    What do you use for mustard on the solder? :-)

    O.K. The Multicore logo looks the same as it used to at least.

    That brings back memories. :-)
    Enjoy,         DoN.
--
Email: < snipped-for-privacy@d-and-d.com> | Voice (all times): (703) 938-4564
(too) near Washington D.C. | http://www.d-and-d.com/dnichols/DoN.html
  Click to see the full signature.
Add pictures here
<% if( /^image/.test(type) ){ %>
<% } %>
<%-name%>
Add image file
Upload
On 2/25/2010 2:58 PM, DoN. Nichols wrote:
(...)

Either will do for the purposes of this laboratory.
For extra credit use a 'hamburger' metaphor to explain capacitance. :)

French's errr.. Victory Flux!
(...)

Heh! That will melt a few decades away, Right Now. :)
--Winston
Add pictures here
<% if( /^image/.test(type) ){ %>
<% } %>
<%-name%>
Add image file
Upload
I have some - I want to say it was British / English made. Likely bought out or patent ran out. Maybe the get the lead out did it.
IIRC, I have some with copper in the alloy. It is tougher.
Martin
Winston wrote:

<http://www.newark.com/multicore-solder/mm00975/502-cored-wire-solder/dp/32M7025

Add pictures here
<% if( /^image/.test(type) ){ %>
<% } %>
<%-name%>
Add image file
Upload
On Thu, 25 Feb 2010 21:06:10 -0600, "Martin H. Eastburn"

From the website of the manufacturer of Ersin solders, Multicore Solders Ltd
Multicore Solders Ltd Kelsey Hse, Wood Lane Hemel, Hempstead U.K. HP2 4RQ
Flux-cored wire is universally made by an extrusion process followed by drawing through dies progressively reducing the wire diameter from say 16mm to the required diameter which is normally in the range 3mm to 0.2mm. As solder is opaque and usually contains lead it is not possible to see and test inside the wire to check for flux presence. The extrusion process is subject to variations of pressure and temperature which can lead to occasional blockages or contractions of the flux core. The shortage of such a flux void at the extrusion stage results in a much longer length without flux in the drawn down wire. Any gap in the flux continuity will result in a 'dry' faulty joint. Solder cannot join metals without flux. Multicore Solders Ltd. has employed its own unique extrusion process for over 40 years to guarantee flux continuity. It has earned an enviable reputation for this reliability that has never been challenged.
Firstly we use machines of our own design to minimise variations of temperature variations etc. during extrusion.
Secondly we are the only company in the world, to the best of our knowledge, that extrudes 5 truly separate cores of flux inside the solder wire.
Until 1967 we had patents covering our principle of extruding more than one core of flux. Most competitors still make single core. Some have tried to imitate our technology but we have proved that the shape of their flux cores (which sometimes collapse together at the centre as in the photo above) is due to flux being injected through a single nozzle having 5 holes in it. Such a crude imitation is no more technically reliable than a single cored solder.
The probability of all 5 cores being absent from Ersin Multicore 5-Core Solder is 5x4x3x2x1 = 120 times less than competitors' cored solder, even if the single core solder process was as closely controlled as Multicore's.

Add pictures here
<% if( /^image/.test(type) ){ %>
<% } %>
<%-name%>
Add image file
Upload
On 2/25/2010 10:03 PM, snipped-for-privacy@snyder.on.ca wrote:
(...)

Ah! Reading between the lines, they squirt a 16mm diameter solder slug out of the extruder (with flux holes in place) then pressurize the flux at one end of the slug whilst (g) drawing the other end through a series of reducing rollers.
Pretty darned clever.
--Winston
Add pictures here
<% if( /^image/.test(type) ){ %>
<% } %>
<%-name%>
Add image file
Upload

Polytechforum.com is a website by engineers for engineers. It is not affiliated with any of manufacturers or vendors discussed here. All logos and trade names are the property of their respective owners.