Wednesday, March 23, 2005

More on the great capacitor scandal -- Nichicon

MacInTouch Home Page

Macintouch has been tracking capacitor and motherboard failures on Apple's G5 machines. Now they expanded to reporting on Dell issues. Macintuch really does journalism. They use their readers as reporters and deploy their own editors and investigators. An interesting and underappreciated model!
Ron Royer reports that the current rash of iMac G5 failures may be caused by a bad batch of Nichicon capacitors that is also causing Dell computers to fail:

I don't know if you have seen this, but other manufacturers seem to have this issue too... Postings from Badcaps.net:

"I work as a network engineer/administrator at a company in Washington DC, with about 600 installed workstations. As many as 200-300 of those are the GX270 models, which are 2.4-2.8GHz P4 machines.

We've had a rash of motherboard failures on these machines. I only recently had a chance to inspect a dead board before it was returned to Dell, and it turned out to have swollen Nichicon caps.

I finally had a Dell tech admit that they were aware of the problem and were replacing boards under warranty (the corporate machines usually have 3-year extended service plans). As of today we have 7 new boards on the way to us and at least 15-20 more have already failed. (I'm pretty sure that the GX270 boards are OEM-built by MSI.)

I'm just posting this for information, and to see if anyone else has seen this in a large-scale IT operation, especially with Dells."

"Well, we did 12 boards today, and we have 6 more to call in. I spot-checked some of the bad caps, and they all seem to be Nichicons with the same lot numbers and about the same date range (early 2003). All the replacement boards that were installed had Rubycon caps instead.

I did notice a few boards where, while the caps looked identical, not all of them were bulging. There is a picture of this here: [Dell Motherboard Capacitor Failures]

I'm starting to agree with the idea that this is Nichicon's own quality control problem, rather than the same batch of bum electrolyte as last time. I also think that the lousy ventilation inside the Dell small-form-factor case speeds up the failures, since they don't like to waste money on things like fans."
The G5 interiors run hot -- as do small-form-factor Dells. I think the Mac Mini is quite a bit cooler. This problem may be a combination of high temperatures and manufacturing problems. Are computer manufacturers running at temperatures that exceed manufacturing specs for their components?

I also wonder if setting the iMac to 'full performance' mode might accelerate this problem.

No comments: