If a drive in a Sawtooth gets erased (zeroed) and a fresh install of Leopard will that same drive (system) function without difficulty in a MDD?
Just curious (troubleshooting).
Yes, assuming it's jumpered correctly for where it is on the cable, MDDs generally use Cable Select, Earlier G4s used Master, Slave, or Single depending on make of drive.
Found a molex with a branch line the 120mm would plug to and tried powering it that way (assuming it's the 12v side) and the fan didn't work.
With the case opened the temp continues to drop, now at 45.9º.
So. Further troubleshooting showed there was power on the MB for the 120mm fan. Used a replaced PSU fan to check power. The SilenX is a 3 wire fan without color coding - I had the wrong 2 wires in use, a little trial and error got the right 2. This is the second SilenX fan for this unit and only a month or so old. I thought I had it wired same as the other.
At the moment the CPU is at 37.9º. It certainly appears I have the cooling issue in hand.
I still have a 'short' somewhere though. I have been able to open the case once while powered and I have been able to start up once while the case is opened. Subsequent efforts still produce a shut down. I'm 95% certain it is the main plug at the MB. Manipulation of the harness at the plug is what has made the difference. Any ideas how to troubleshoot the 24 pin connector for a loose connection, after all the poking and prodding to check and repair the PSU, thats what I believe it is?
At the moment the CPU is at 37.9º. It certainly appears I have the cooling issue in hand.
That's more like it!
Glad that you got the fan going.
Any ideas how to troubleshoot the 24 pin connector for a loose connection,
Like a Molex, the pins can move up and down in the plastic housing if they have ever been removed, or, if they have ever been forced.
Close, visual inspection is the only way that I know of to find the culprit.
Like a Molex, the pins can move up and down in the plastic housing if they have ever been removed, or, if they have ever been forced.
Close, visual inspection is the only way that I know of to find the culprit.
Yeah ok, I'll see what I can find.
I don't think this machine has ever run this cool and quiet. Do you know anything about CPUTest? and how to use it? I've been running progressively larger tests and the MDD never gets above about 60º, the fans only barely turn up. May be no point in it, but thinking I can get a better idea where I'm at with it.
Here's my best guess. After careful examination of the female pins in the PS plug, they all appeared to be seated with very little vertical movement. Using a plastic probe I could push them slightly till they hit a stop. The motherboard side was rock solid, both each individual pin and the receptacle.
With the machine running I probed for lateral movement of the wires and found 5 (maybe 6) that with the slightest pressure would shut down the machine. This tells me why the machine would shut down when opening and closing the case. As that wire bundle torques on the bottom of the case it correspondingly moves those wires and *blink* out goes the power.
So!! what to do. The connection at this point is so sensitive I cannot get the computer to start with the case closed. I can only get the computer to start now by careful manipulation of the wire bundle. Wish I could think of a way to clamp/close/compress the female ends to fit tighter to the mobo pins. Seems that would do it.
Progress.
Those connectors on the PSU end are like little square tubes that have been split in half. On closer inspection a few of them were visually spread apart. Found a small jewelers screwdriver that fit nicely alongside the connectors and was able to pry them back closed. I'm sure they aren't as tight as 'new' but visually they are much better.
Enough better that at least now I can close the case and start the computer. I tried opening it while running and it still shorts. I don't know what I can do to make the connectors tighter. I can only pry till the 2 sides meet. If I could 'over' pry - passed (or past) zero a bit, it would give the spring necessary to grip again.
I've had this machine close to 10 years and it has never run this quiet and cool. I like my MDD again.
Thanks japamac and BDAqua.
Great work on your part... I was going to recommend just such prying, but I have to watch my toes around here anymore, but thanks for the report! 🙂
On closer inspection a few of them were visually spread apart. Found a small jewelers screwdriver that fit nicely alongside the connectors and was able to pry them back closed.
Without a pin extractor for that type of pin, what you did is as good as you can do, possibly better due to the TLC put into each pin.
Do you know anything about CPUTest? and how to use it?
This?
http://www.coolbook.se/CPUTest.html
From the Glucas page:
Glucas is a client to make primality test for Mersenne numbers. It is related to GIMPS effort to search the bigest primes ever found. In addition, it can be used as an excellent performance test.
Interesting, and No, I haven't had experience with that.
To test a processor, I just use OS X Chess. Run a Computer vs. Computer game and you will stress the CPU as hard as it can be stressed.
Run and complete 10 games and you are guaranteed a rock solid processor and stressed the cooling as much as it can be.
Very interesting regarding the Computer vs Computer chess match. I ran a few incrementally larger CPUTests yesterday and couldn't get the CPU temp above 60º. At that point the fans were 'thinking' about turning up.
I augmented my 'pin' repair with a twist in the bundle. It added some torque. Part of me doesn't like the idea but it is helping. There was a point where I couldn't close the machine and boot - could only boot it opened, then closing would 'short' and shut down.
It's been a bit of a road but well worth it, have learned a lot and I like my MDD again - 🙂. Thanks to you japamac and BDAqua.
MacOS Interchangeable From 1 PM to Another
