Latest Casualty ...

[GorfTheChosen]

Disabling Speculative Access did not solve the problems with burning with Toast with the optical drives.

However I was able to solve it for Sony CD-RW by removing the two other devices off the SCSI chain, running a single cable to the drive with a terminator installed on the other port, eliminating two old APS short cables.

I suspect it maybe a cabling issue more than anything ... although one of the eliminated devices is an OEM Apple 1GB HDD (vintage 1993 or 1994) that hasn't been showing up at all the last couple of times I turned it on. The HDD spins up but doesn't show up as being on the chain using a variety of SCSI utilities.

The DVR in the ADS Firewire case still locks up the computer when Toast attempts to start burning a disc. Could be a cabling issue or possibly the Firewire card. This particular device has worked reliably before on a number of different machines here ... but all of them had built-in Firewire.

[GorfTheChosen]

The repaired PSU's have arrived back from Andy.

I opened the package and looked just enough to tell that they were packed very well, will probably get around to completely unpacking and installing them here in the next hour or so.

[GorfTheChosen]

Joy !

The Power Mac G4 800 Dual Quicksilver is back up and running. (Big Thanks to Andy for his work on the PSU)

There was however a little excitement involved in getting it to happen: I had installed that SiL 3124 SATA card in the machine and it while the machine would power up/boot/give the start up chime, there was no video output (black screen)

As soon as I removed the SATA card it booted right up with video.

The G4 450 Dual is next ... right after I figure out how I want to partition the SSD and clone the present boot drive to it.

[GorfTheChosen]

Power Mac G4 450 Dual was obviously way overdue for a deep cleaning ... 

[GorfTheChosen]

And the Power Mac G4 450 Dual is up and running ... 

It initially booted into Open Firmware ... not sure what's up with that but seemed to boot and reboot ok after that.

I did stick that SiL 3124 SATA card in it, along with an Acard or ProMax ATA 66 card ... neither have any drives attached ... and ASP showed that the unit had NO PCI cards installed ... WTF ?

Booted off the SSD ... not terribly impressed with those (OWC Mercury Electra 3G's) on either the Dual 450 or the Dual 800 so far (much more noticeable difference when I installed one in the Mac mini - made it fly) ... boot times seem kinda long (25 - 30 seconds to get to the desktop) ... though on the 450 Dual maybe that's because of additional storage cards installed with nothing attached to them.

Plan on getting the LinkSys Gigabit switch installed and the 3 Power Macs wired up to it today, and connected to the modem/router ... probably do some network testing. And get all the exterior last reinstalled on the G4 450 Dual after it's washed.

Reconfigured the duplicator tower yesterday (found another 60GB drive in there that I forgot I had  ) and installed a Firewire bridge board connected to a couple of mobile racks in it as well. Had to use the plastic mobile racks rather than the nicer ALU ones, as the latter would not fit in the tower's slots.

The Firewire bridge board I think came from ProMax and uses an Initio chipset ... not sure if it supports more than one drive ... but I configured two trays with drives set to master, and two set to slave ... so we'll see what happens.

Might try to get the BIOS in the PC flashed if I have time.

[GorfTheChosen]

The caps for the Samsung LCD power board came in yesterday. Need to do a little more clean up on the work area though, before I attempt soldering on the replacements.

Last night I got around to finally installing a purty gold anodized ALU heatsink on the Northbridge chip of the PC and managed to get a good look at the cap situation on the mobo.

It ain't pretty ... it appears it's just full of those cheap green Teapo caps - I counted at least 22 of them (badcaps.net confirms 22) ... and a bunch of them are swollen up and leaking out of their tops.

After that I did a little sleuthing and found out via MSI's own support forum that it was a known issue with that product, with multiple folks mentioning that they RMA'ed their mobos to MSI for replacement or warranty work.

Of course ... 

[FBz]

Mac Abuse!

[GorfTheChosen]

Mac Abuse!

Clearly ...

[GorfTheChosen]

Ok, the recap on the power board for the Samsung display is largely done ... except for C317. (More on that in a moment)

I used one of those yellow 25w soldering irons ... mainly because I found out that the tip on the ECG is apparently not replaceable (easily anyways)

And all the screw-in tips I have fit the yellow irons and there were a couple of very small, fine point tips.

The 25w irons were more than enough and it all went pretty smoothly. I actually even impressed myself with being able to hold the iron without my hands shaking (much  )

It took a bit of practice but by the 8th pin I think had a decent looking solder joint.

 

[GorfTheChosen]

Here's the issue:

On C317, one of the pins was soldered into an exposed trace of solder on the board surface ... which had another exposed trace of solder right next to it.

I do not believe that the two traces were joined but it's possible that they were (Good case for: Take lots of pictures before you start dumbass !)

Either my hand slipped or I just got things a little too hot and the two traces co-mingled and may have formed a bridge between the two traces.

I'd prefer to clean that up and separate the two before putting power to it.

The thinking being that if they were connected originally the missing connection might not be as devastating as leaving them connected when they weren't connected originally and putting the juice to it.

The question is how to go about doing that.

Just use solder wick ?

[ssp3]

From what I can see from the limited area you've pictured, both of those "exposed" traces are just part of one larger trace. Don't worry about shorts between them.

[GorfTheChosen]

From what I can see from the limited area you've pictured, both of those "exposed" traces are just part of one larger trace. Don't worry about shorts between them.

Ahh ... so basically the darker area indicates a larger trace ... but on a layer lower in the board, below the surface.

I got so wrapped up with what was going on, on the surface, that I completely missed that.

Interesting.

Thanks !

[GorfTheChosen]

Finished bagging all the little screws and other assorted parts to clear off the workbench and then finished up the last cap on the power supply board for the Samsung display. Must have been 75 - 100 bags worth ... 

Hooked the Samsung up to the IMS TwinTurbo128 in the 9600 which has a Belkin video adapter on it and got nada as far as a desktop screen image was concerned, at whatever resolution or mode it was set for. The onscreen menu system did come up on the display however, so that was encouraging.

Moved it over and connected it to the G4 Dual 800 and the desktop image came up just fine, and the picture looked good ... no wonky anomalies.

However, the display went black just about the time I was going to shut it off and disconnect it.

I had touched the front of the display case in the lower righthand corner near the power button when that happened ... so I'm going to need to look into it further and see what the issue is if any ... possibility of a broken wire or bad power button perhaps.

At any rate I was encouraged that I was able to work on the power supply pcb and bring what had been an apparently dead display back to life.

Next up - after troubleshooting any remaining issues with the Samsung - will probably be to finish cleaning up the MacBook (A1181 2.4 Ghz Core2 Duo "Penryn") and putting it back together, replacing the CCFL for the display, installing a new SSD, maybe maxing out the RAM to 6GB.

[GorfTheChosen]

Making some progress ... have focused on the G4 800 Dual ... in terms of getting getting two partitions for Mac OS 9.2.2 and Mac OS X 10.4.11 fully setup with all the software software I own/use installed. Once I get that done and get any extraneous crap cleared off (a folder with over 35GB of "Software Updates" for starters) I'll have a better idea how much disk space is really needed for each partition. Will probably image or clone those ... so that I can just use the images for multiple machines.

I am having one issue with the G4 800 Dual though ... I think it's really more of annoyance than anything.

This has been going on for a long, long time (it was happening in the early 2000's) ... way before the PSU's died.

First boot from "Powered Off" I end up getting a kernal panic within the first few minutes after hitting the desktop - doesn't seem related to anything I'm doing (IOW: I can't do any number of things and it still occurs):

Thu Apr 16 12:03:21 2020
panic(cpu 0 caller 0x002FBE4C): A driver releasing a(n) IONetworkData has corrupted the registry

Latest stack backtrace for cpu 0:
      Backtrace:
         0x000954F8 0x00095A10 0x00026898 0x002FBE4C 0x002FFEB8 0x002FFD08 0x002FFE98 0x002FFD08
         0x00303630 0x002EB370 0x0008B6E8 0x00029234 0x000233F8 0x000ABEAC 0x028BA340
Proceeding back via exception chain:
   Exception state (sv=0x381C1500)
      PC=0x9000AF48; MSR=0x0200F030; DAR=0x0030A174; DSISR=0x0A000000; LR=0x9000AE9C; R1=0xF0080470; XCP=0x00000030 (0xC00 - System call)

Kernel version:
Darwin Kernel Version 8.11.0: Wed Oct 10 18:26:00 PDT 2007; root:xnu-792.24.17~1/RELEASE_PPC

Can anyone shed any light on this ?

Don't really recall it happening with the G4 450 Dual ... but I can't say for sure it hasn't ... just pretty sure I haven't seen it since I had the PSU's rebuilt for both machines.

[GorfTheChosen]

This is the Firewire Bridge board I'm using in the duplicator tower to access old PATA drives which I have in mobile racks ... uses an Initio chipset (they make a version with two IDE channels/connectors as well, which is supposed to support RAID configurations:

https://www.ebay.com/itm/FWB-IDE01A-Support-Two-IDE-ATA-ATAPI-devices-Initio-Chipset/133256102717

Anyone have experience with these ?

It appears to support two drives - Master and Slave.

I never used it that way previously ... but recently I thought it would be interesting to see if it would be functional in that configuration.

However, both drives which attempted to use as Slaves (in the correct mobile rack on the proper connector, and jumpered to "Slave") both ended up getting corrupted ... 

Fortunately, I was trying to use them as backups ... so I had the data elsewhere and didn't lose anything.