Rebuild?

[c_hegge]

Those D945GTP motherboards do seem to get hot VRMs a lot. I've seen them get pretty severe PCB discolouration around some of the FETs and coils before.

btw, I still have the D945GTP which I used to test the CM N200 case. Mine doesn't have any discolouration yet, After seeing just how hot it got with that 130W Pentium D CPU, I decided to mod the VRM somewhat...

[LongRunner]

The curious thing here is that it was only the +12V pins that overheated. Either too much current was being drawn, the contacts have loosened up, or some combination of the two.

Doesn't look like very good design on Intel's part — and they seriously thought a single 1200µF Nichicon HD on the input (though most of mine have three Chemi‑Con KZEs) was good enough?

[c_hegge]

The curious thing here is that it was only the +12V pins that overheated. Either too much current was being drawn, the contacts have loosened up, or some combination of the two.

Interesting. I didn't think a Celeron CPU would be capable of overloading an ATX12V connector. After all, the maximum recommended is 6A/pin (144W for a 4 pin connector). I'd say bad contact at least contributed to the problem.

Doesn't look like very good design on Intel's part — and they seriously thought a single 1200µF Nichicon HD on the input (though most of mine have three Chemi-con KZEs) was good enough?

It seems like it. I fixed that up too while I was at it, though. I haven't got a pic of it yet, but it now has three 470uF OSCON SEPCs. Silightly higher capacitance (now 1410uF), and much, much lower ESR

[Wester547]

That sounds like something ASUS would do. I wonder what other boards Intel underspec'd.....

[LongRunner]

Comparing the drive bay covers from the PS08 and the old AOpen case shows that Silverstone used taller reinforcing ridges on their sides, so the rigidity is quite deliberate:

Drive bay cover comparison.jpg (90.04 KiB) Viewed 24677 times

I did find their mechanism for "tool-less" removal rather clumsy, however (and what's the big advantage when the case still requires screws to mount the drives?).

The front fan is a 120mm Honghua sleeve bearing unit labelled at 1200RPM (which was confirmed by using SpeedFan to monitor the tacho output) and with a 9-blade design a bit like NMB's 4710KL (which, compared to competing 7-blade designs, provides higher airflow but at lower pressure). Examining the bearing showed a bit of grease — which is not good, although you may get away with it if you use temperature control. The intake air filter is one of those plastic-mesh designs that aren't all that effective.

(For now, I'd actually rather have 120mm than 140mm fans; 120mm fans are very easy to obtain, but none of the top brands — ebmpapst, Nidec, NMB, or Sanyo Denki — offer a 140mm model. OK, Delta does make a few FFB and TFA models in that size, but those are jet-engines absolutely not suitable for a home or office PC and even if you could stand the noise, they only come in 24V and 48V versions anyway. The 172mm round fans aren't even close to quiet either.)

The case does provide for an additional 80mm front fan below the installed 120mm unit but that would go behind the front part of the chassis — which would limit you to using the rearmost set of mounting holes for the lower 2 HDDs (it would be 3 if they didn't render the third from the bottom unusable with that last-minute "reinforcement"), making an already cramped case even more cramped. Without the "extra" fan installed, the opening for it does nothing more than let some of the air taken in by the 120mm front fan go right back out.

My assessment of it:

Pros

• Low price (not that I care much)
• Front audio jacks actually work

Cons

• Dubious front fan (−1)
• Not that solid (−1.5)
• Lazy adaption of an old design (−2)
• Poor air filter (−0.5)
• LEDs are too bright (and placed behind the air filter) (−0.5)
• Cramped interior — HDDs block off parts of the board (not unique to this case, but still…) (−1)

Score: 3.5/10

YMMV on the painted interior, but IMHO it does little for the system's looks (considering that no ordinary user is actively looking at the back of their PC), adds to the cost for no real benefit, and as the paint is non-conductive, can potentially interfere with correct earthing of each part of the chassis (which is important for shielding — otherwise RFI can increase). So not quite a Fail Award, but I still would not recommend it to anyone.

Frankly, I'm awfully tempted to think that I could design a better PC case than many of the current budget offerings.

[LongRunner]

I added a rear fan (92mm, sleeve bearing, made by Power Logic). It's pretty quiet, with a top speed of about 1500RPM; a bit odd as the slowest documented variant runs at 1900RPM, so it must have been custom-configured for the case it came from (which also had two 80mm rear fans, from the same manufacturer and again with sleeve bearings, running at 1100RPM, the slowest documented variant being 2000RPM). I'm not sure how much this will help, but it has to have some effect.

[c_hegge]

Ouch. I didn't think that the PS08 would be that bad. It seems like the TJ08 may be well worth the extra money. I have built a few computers in that case before and it's a very solid case.