Page 6 of 20
Re: Fixing my Milan 040 (or not) blog
Posted: 23 Jun 2025 12:27
by peters
I was about to say, are there any tiny bits of metal in the ISA slots ?
Re: Fixing my Milan 040 (or not) blog
Posted: 23 Jun 2025 12:39
by peters
https://exxosforum.co.uk/atari/mirror/m ... lan060.htm
I had some terrible issues when upgrading the CPU that were an caused by the PSU. It was fine with the 68040 but not with the 68060.
Re: Fixing my Milan 040 (or not) blog
Posted: 23 Jun 2025 12:49
by exxos
dml wrote: 23 Jun 2025 12:04
...this is where you find a tiny coin jammed in the ISA slot :)
Yeah it looks clear other than dust.
Re: Fixing my Milan 040 (or not) blog
Posted: 23 Jun 2025 12:52
by exxos
I did wonder about the upgrade. But I already have a CT60 with 060 CPU. So I think I would prefer to keep the Milan as a 040 :)
I was looking for faster CPU, but lots of threads about the fakes etc similar issues with the 060 fakes in fact.
Re: Fixing my Milan 040 (or not) blog
Posted: 23 Jun 2025 13:36
by alexh
exxos wrote: 23 Jun 2025 12:52
I was looking for faster CPU, but lots of threads about the fakes etc similar issues with the 060 fakes in fact.
Yes sadly true. Someone unwittingly wrote a guide to 68040 parts stating that the 0.57um 68040@40MHz with mask L88M as being "the one to get". Ever since forum post (circa 2006) the resellers have been remarking 040 parts. It's practically impossible to get hold of a real one today.
It was an Amiga forum post but surprisingly the 040 market is fuelled by Mac owners.
The author of that forum post also used two throw-away non-technical terms "cool to the touch" and "too hot to touch" to describe the 0.57um L88M parts vs 0.8um D43B parts and everyone has been beating him up since. (
You can find Mac users doing thermal imaging just to disprove the claims). However I standby what I wrote :)
Re: Fixing my Milan 040 (or not) blog
Posted: 23 Jun 2025 13:42
by exxos
Out with the PCI slot. Will get a new one to replace it with. Then at least 1 PCI slot may actually work :lol:
IMG_3426.JPG
IMG_3427.JPG
I am going to heat this up from the bottom. Hence the tape to stop other stuff melting.
IMG_3428.JPG
Flux on top. Will heat with my gas soldering iron and hopefully it will come off without to much carnage.
IMG_3429.JPG
Re: Fixing my Milan 040 (or not) blog
Posted: 23 Jun 2025 13:50
by exxos
alexh wrote: 23 Jun 2025 13:36
Yes sadly true. Someone unwittingly wrote a guide to 68040 parts stating that the 0.57um 68040@40MHz with mask L88M as being "the one to get". Ever since forum post (circa 2006) the resellers have been remarking 040 parts. It's practically impossible to get hold of a real one today.
It was an Amiga forum post but surprisingly the 040 market is fuelled by Mac owners.
The author of that forum post also used two throw-away non-technical terms "cool to the touch" and "too hot to touch" to describe the 0.57um L88M parts vs 0.8um D43B parts and everyone has been beating him up since. (You can find Mac users doing thermal imaging just to disprove the claims). However I standby what I wrote :)
Yeah I read a big thread on another forum last night about it all (forget what forum) talking about it all :)
I can source genuine parts, but generally the price is way to high to really worth considering. But I could ask my suppliers. But they don't really deal with one offs anyway.
Re: Fixing my Milan 040 (or not) blog
Posted: 23 Jun 2025 14:02
by exxos
Set my hot plate to 300c. Literally took seconds and the IC came off! Easiest desoldering job ever!
IMG_3431.JPG
IMG_3433.JPG
My air filter also doubles up as a PCB cooler! :lol:
IMG_3434.JPG
Clean up next :) Then I will power on see if I get any beeps from the souder :)
Re: Fixing my Milan 040 (or not) blog
Posted: 23 Jun 2025 15:00
by exxos
Cleaned up..
No more odd shorts to VCC or GND (only did a quick test around) and they seem about 5K now. But no data output at all now. They data lines should be pulsing low after reset but now does less than before :shrug:
IMG_3442.JPG
IMG_3443.JPG
IMG_3444.JPG
Re: Fixing my Milan 040 (or not) blog
Posted: 23 Jun 2025 15:25
by exxos
Analysis of the address bus and 16,17,18 I don't think should be high..
Code: Select all
A0 0
A1 0
A2 0
A3 0
A4 0
A5 0
A6 0
A7 0
A12 0
A15 0
A16 1
A18 1
A17 1
A14 0
A13 0
A8 0
A9 0
A11 0
A10 0
There isn't much in the EPROM higher up, so likely as I am seeing D0-D7 all high, its probably correct with the output. But now to figure out why those 3 address lines are stuck high... :roll: