Thread

Two dead G4s

Two dead G4s Hardware 42 posts Aug 14, 2010 — Sep 12, 2010

#31 Wed, 25 Aug 2010 - 21:11

I bet he did the rest but forgot about the -5v. But yes, he should have examined everything more closely.

At least nothing got toasted though.

#32 Wed, 25 Aug 2010 - 21:19

You powered it up without even looking at the pinout? :O

Hey I'm new to all this! Plus it was late and most of my thinking faculties had gone to bed for the night. I did look at the pinout but I was mainly concentrating on the +5vSB pin. Anyway it all worked out good in the end, thankfully.

I've bunched up three of the sets of cables coming from the PSU, tied them and wedged them in behind the fan bracket to keep them out of the way. Whoever took out the hard drives also took the hard drive caddy for the IDE drive. So I used the middle hard drive carrier and placed some little dampeners between the carrier and the drive to help with any vibration. I dug out the horrible, horrible pair of "pc speakers" I got with my first PC. Should do the trick for hearing sound out of the line out for now.

So now I finally have my first G4. A whole new world of Altivec awesomeness unfolds.

OK, with the speakers plugged in I hear a faint chime, so there must be something up with the internal speaker. Something to look into later...

#33 Thu, 26 Aug 2010 - 02:42

maybe some bad caps in the amplifier circuit? dont know....

#34 Fri, 27 Aug 2010 - 10:35

So now I finally have my first G4. A whole new world of Altivec awesomeness unfolds.

Congratulations on a successful transplant, Dr.

#35 Fri, 27 Aug 2010 - 13:18

Thanks

I didn't get to spend much time on it yesterday but a couple of notable things happened. The chime came back (intermittently) so the speaker is working fine.

On the not-so-good side, at one stage the hard disk started powering down and back up again (like when its not getting enough power). I hit the reset switch and it stayed powered on. It may have been a glitch of some kind but if it happens again I wonder if it might point to the 300w PSU not being up to the task. I also ran MacTest Pro on it and it failed some of the Video tests in the logic board section. It may have been running the wrong tests I'll have to double-check as it was late when I was doing this. Everything else passed OK and video is working fine (passed the AHT tests).

I ran the test again and I think the logic board test may be a little flaky. If I deselect all the logic board tests and run each video test individually they all pass. But if I try to run all of the video tests it fails on the second one (Video DAC Read/Write).

#36 Sat, 11 Sep 2010 - 18:26

Seeing as the ATX transplant worked so well with the GE I was wondering if anyone has ever tried putting the guts of a 1U server PSU into the MDD's PSU casing for a perfect fit replacement? Or maybe there's a 1U unit that would fit nicely in it's place? Shoehorning in a regular ATX PSU doesn't look that great.

It was asked here but nobody gave an answer. However there is a quote that seems to shed light on why these PSUs were so prone to failure.

#37 Sat, 11 Sep 2010 - 20:12

That's very interesting. I can't imagine why it would continue to pull from the +28V when it's off. I resurrected my MDD with an ATX mod and nothing going to the +28V pin at all and it works just fine (minus ADC and powered firewire) Was it Just a design flaw?

#38 Sat, 11 Sep 2010 - 20:34

That's very interesting. I can't imagine why it would continue to pull from the +28V when it's off.

because on those models the 28v is also the standby.

I resurrected my MDD with an ATX mod and nothing going to the +28V pin at all and it works just fine (minus ADC and powered firewire) Was it Just a design flaw?

MDD has a normal 5v standby and separate 28v. Apple might have noticed the overheating while off issue and decided to fix it the proper way.

#39 Sat, 11 Sep 2010 - 20:40

Congrats on the successful repairs.

My startup-chime disappeared when I installed linux on my Sawtooth. Apparently setting a non Mac OS drive as the default boot disk does something. I got it working again at one point when I reset the PRAM. (But then I installed a new linux OS.) Maybe resetting the PRAM will get your startup-chime working.

#40 Sat, 11 Sep 2010 - 21:16

The chime is working fine now. I think the reason it was intermittent at the start was because I was running tests with MacTest Pro and it didn't seem to chime when I rebooted out of the MTP disc.

It's dual-booting 9.2.2 and 10.4.11 and got a RAM bump to 1GB

#41 Sun, 12 Sep 2010 - 01:11

Any reason you installed tiger instead of leopard (you are talking about the MDD right?). I just have to throw in my hat and say that any of the MDD machines make wonderful snappy leopard machines

#42 Sun, 12 Sep 2010 - 01:28

I'm talking about the GE. The MDD is still sans-PSU which is why I was asking about the 1U stuff.

I'll put Leopard and 9.2.2 on that one whenever I get it running.