1
The Mac OS 9 Lives Forum is dedicated to...
Installing, Maintaining, and Extending the functionality of Mac OS 9
Each Section can be Expanded or Compressed by clicking the Title of the section
This will shrink the Forum to display Only the Sections you want
Maybe call it an "iSink"… after indibil’s original approach?
It isn’t pretty (yet) but the first “replicant” here is finally complete.
Could use a bit of cleaning (copper dust particles, etc.) and some emery cloth smoothing of the rough fin edges.
Used a metal hack saw blade to cut the 50 x 50mm standard size down to 40 x 50mm and then a palm sander to smooth the rough cut edges. indibil used a 4” grinder with abrasive blade to remove 10mm - but I didn’t trust myself to go that route. Even tested a very small mini table saw and a mini chop saw… but they were too under powered for the task (so much so that the hack saw blade seemed better for a more controlled cut between the fins.. For the next one - may try a metal jig saw (or sawzall) blade clamped and held with vice grips to cut that 10mm excess off (again, by hand). Smaller and more teeth than shown here.
You may note the black dots on the face of the middle one in the second image above. Those are the projected locations for holes. I even created a paper template by scanning an actual G4 mini heatsink @1:1 for this approach but finally settled on a variant of indibil’s mating of the old heatsink, plus the new one (clamped in a vise) and then drilled.
Also in the second picture above… red dots denote one fin that was later removed, which brought this total fin count to 19. Unknown whether fewer fins will effect overall cooling but that’s a test to be done soon. (indibil’s copper heatsink has 20 fins.)
Also, the green outlined area (also 2nd image, left side) denotes the thicker base of the copper heatsink compared to the lesser thickness of the original stock heatsink. Thought might need to countersink the holes on the copper to facilitate use of the original “spring-pins” length but indibil said unnecessary and that the original “spring pins” worked well even with the slightly greater thickness of the new heatsink’s copper base.
Trimming the fins. Where the real fun is!
Not really. Struggled and fretted over this more than anything and still have not arrived at a comfortable, set approach. I’ve enough finned copper stock here to make three more iSinks - so maybe by the last one I’ll have a better approach? For this first one, you might see where I drilled down through the copper to help locate the area where the spring-pins would enter the base in relation to the fins. This helped to guide how far back the fins would need to be trimmed. Then using very small metal snips, the fins were cut back as far as possible AND then used a Dremel tool with an abrasive bit to grind the fins back further. Also tried the Dremel with a circular abrasive disk to cut the fins, but definitely need more practice with that. Perhaps a combination of all the above would make it all easier / better?
CNC machine.
Yeah, I know.
From here, the next step for me will be to test what effect the copper heatsink might have on air exhaust temperatures compared to an original heatsink in a stock 1.5 GHz G4 Mac mini (if any at all). I think the temps may be hotter but those I have asked seem to think that: (1). The temps will be the same, or (2). Fan speeds will decrease.
AND all of this in prep for possibly overclocking a G4 mini to 2 GHz. My thanks to indibil and to sailorMH for their efforts in relation to the entire subject of overclocking the G4 minis. Might have to eventually send one of these finished heatsinks over to the Czechia (sailorMH).
For more in-depth, G4 mini overclocking info (warm up your Google website translator):
indibil…
https://rosysumenteinquieta.blogspot.com/
https://rosysumenteinquieta.blogspot.com/2023/12/mac-mini-g4-192ghz-con-disipador-de.html
sailorMH…
https://www.amigaportal.cz/node/156189
"fastest" flash, and using a pencil to overclock - MacRumors
Other info on overclocking the G4 Mac mini? There’s a thread over on MacRumors where ervus (in 2020) first mentions overclocking a mini to 2 GHz w/o a 7448 - but it still overheated. Read all (currently) 8 pages of that thread. (Perhaps this is what Jubadub previously referenced?) And although indibil’s resistor tweaks and current adjustments might be the final word in stable overclocking the G4 Mac minis to date.
Begin with ervus:
https://forums.macrumors.com/threads/fastest-flash-and-using-a-pencil-to-overclock.1973903/page-2?post=29415588#post-29415588
Random errata:
Drill bit size = 3.0mm
Harbor Freight (in the U.S.) sells a small set of metric bits for less than $5.00!
HSS Micro Drill Bit Set 3 / SKU 61526 / $2.99. (30 assorted small bits)
Hole template with fin trim-back info
You can either trim back 1/4” (6.35mm) of the fins in the front and back locations
(like indibil has done) OR… take a full 5/8” (15.875mm) from the back location.
See first image here at the very top.
Sourced the heatsink blanks, 50 x 50 x 20mm and the 100 x 50 x 20mm from:
YOLOWIN / AliExpress. 100 x 50 x 20mm can yield two for a few dollars more.
https://www.aliexpress.us/item/3256804076300199.html 50x50x20
https://www.aliexpress.us/item/3256804078907505.html 100x50x20
AND - used slow speed on the drill press as copper is very soft metal.
AND - AND… thanks to Philgood for first mentioning indibil!
But...talking about it just now I had an idea...it might be possible to build a cross compiler on another host...or something like that.
If using any of the 3 stock heatsinks, use copper shim(s) for the 7457 to take into account its slightly lower height compared to the stock 7455, or, in the case of the 7448 that has additional height due to the interposer board, use washers/spacers between the posts and heatsink to raise the height at which the heatsink stands.
My memory says it was already discussed, but the posts must have been deleted:
Heatsinks are never made of "stainless". It conducts heat extremely poorly. What is really at play is chrome-plated aluminum, which has a mirror-like finish.
So, performance wise by the numbers, the copper heat sink still takes it home, the metal heat sink comes in second and the aluminum heat sink comes in last. Common knowledge is that the metal heat sink sucks, is useless, should be avoided etc. but these temperature readings show otherwise. What does make the metal heat sink a poor choice is the system instability it somehow causes. - House of MothIt's been said before, but stuff gets lost or just unseen in a gazillion posts, so I'll repeat: