Alright, here we go again!! This is a first generation iMac G5 (the last one was a second gen). This was one of the donors given to me by a fellow forum member here. I wanted to get one cranked out right away, just so he doesn't think his contributions were for nothing! 
The changes in polymer caps from the previous one are that I used 1800uF 6.3v Nichicon LG series caps in place of the 1800uF 6.3v lytics that it originally had. I used 1500uF 6.3v UCC polymers in the last one, and it worked without a hitch, so I didn't foresee any reason why I would have any trouble here. I did one of the 20" ones first. My plan was to do both poly revisions on each size iMac. When the deal was worked out with the donor, I didn't have the other 20" I used in the other experiment, so he donated 2 of each. Now, I have a spare 20".....
Anyway, onto the project:
Getting ready to start:
Back removed. The motherboard had no bad caps, amazingly enough, there were actually some good Nichicon HN series caps in it. The date code was 0422 (made in the 22nd week of 2004), they should have failed by now! Pre-testing, this system actually would POST.
Crack open the Ac-Bel PSU (might as well start at the beginning), and tadaa!! there's the bad caps! I'm surprised it would even POST like this.
Recapped in Rubies:
Junk L-TEC caps:
Revived PSU:
Now onto the logic board. I was shocked to see the original caps were still good, even the rubycon's. I've seen these heaters cook rubycon and UCC lytics many times.....I was amazed beyond words that the HN's were still good as well, those would fail under normal operating conditions, forget about extremely heated conditions. All the caps on this board were original, it had not been recapped previously.
The 'miracle' nichicons:
Unbloated rubies:
I'm getting to the point I can have one of these apart in about 10 minutes....
...and the polymer goodness!
Nichicon LG series:
Some 330uF 16v UCC's replacing the 1000uF 16v HN's:
Back together:
It POST's:
...and OSX being installed:
it's up and running now, and another iMac polymodding experiment seems to be successful!!
Keep in mind, the primary purpose of these experiments is to see what will play nice in these systems as far as polymers go. Longevity testing is rather nill, as in a heat-stressed environment, polymer caps will outlast even the best lytics on an exponential scale, so that's not a mission objective here. By the time the polies would fail, we'd all have long white beards anyway. I just wanted to make sure that the poly mods and alternative capacitor values didn't compromise stability in any way.
The changes in polymer caps from the previous one are that I used 1800uF 6.3v Nichicon LG series caps in place of the 1800uF 6.3v lytics that it originally had. I used 1500uF 6.3v UCC polymers in the last one, and it worked without a hitch, so I didn't foresee any reason why I would have any trouble here. I did one of the 20" ones first. My plan was to do both poly revisions on each size iMac. When the deal was worked out with the donor, I didn't have the other 20" I used in the other experiment, so he donated 2 of each. Now, I have a spare 20".....
Anyway, onto the project:
Getting ready to start:
Back removed. The motherboard had no bad caps, amazingly enough, there were actually some good Nichicon HN series caps in it. The date code was 0422 (made in the 22nd week of 2004), they should have failed by now! Pre-testing, this system actually would POST.
Crack open the Ac-Bel PSU (might as well start at the beginning), and tadaa!! there's the bad caps! I'm surprised it would even POST like this.
Recapped in Rubies:
Junk L-TEC caps:
Revived PSU:
Now onto the logic board. I was shocked to see the original caps were still good, even the rubycon's. I've seen these heaters cook rubycon and UCC lytics many times.....I was amazed beyond words that the HN's were still good as well, those would fail under normal operating conditions, forget about extremely heated conditions. All the caps on this board were original, it had not been recapped previously.
The 'miracle' nichicons:
Unbloated rubies:
I'm getting to the point I can have one of these apart in about 10 minutes....
...and the polymer goodness!
Nichicon LG series:
Some 330uF 16v UCC's replacing the 1000uF 16v HN's:
Back together:
It POST's:
...and OSX being installed:
it's up and running now, and another iMac polymodding experiment seems to be successful!!
Keep in mind, the primary purpose of these experiments is to see what will play nice in these systems as far as polymers go. Longevity testing is rather nill, as in a heat-stressed environment, polymer caps will outlast even the best lytics on an exponential scale, so that's not a mission objective here. By the time the polies would fail, we'd all have long white beards anyway. I just wanted to make sure that the poly mods and alternative capacitor values didn't compromise stability in any way.
Attached Files
Comment