Hi,

Here is my recent Amiga 4000 adventure in overclocking...

After acquiring a nice Amiga 4000/040 system, I wanted to give it a speed boost from 25 MHz to 33 Mhz before I started to add expansion cards and OS 3.9 to it. I originally tried the Delay Line hack on my Amiga 4000's 3640 CPU card without any success. It would never clock higher than the stock 25 MHz (using a 50 MHz oscillator). All I would get is a yellow screen over and over again.

I purchased an 68040 at 40 MHz CPU and installed it in hopes if I could get a nice and stable 33 MHz, I would not need to worry about overheating. My 3640 is a revision 3.0 card. I assumed this might have something to do with my overclocking failure attempt. I purchased 2 newer GALs (U204 and U209) which would upgrade my card from 3.0 to 3.2. Eliminating possible Zorro III conflicts with Graphics cards, etc. Being a bit scared to de-solder the two chips and solder in the replacements (with sockets) I turned to the internet for solder tips. While searching I found another mod that can be done instead of resorting ti the GAL replacement (works in most cases of Zorro III lockups, etc). Being much simpler than replacing the U204 and U209, I gave it a shot. All that was required was de-soldering pin 6 on the U200 chip and lifting it from the board then bridge the pad at pin 6 to pin 7.

30 minutes later after close inspection of my modification and some praying, I installed the card back into the 4000 and it booted up fine. Out if curiosity I installed a 66 MHz oscillator in the socket I installed earlier and low and behold, it booted up as well with no errors. The system runs great and the passive heatsink on the 68040 doeen't get too hot.

Sysinfo reports a click speed of 33 Mhz, The CPU scores 24.6 MIPS, up from 18.7 and the FPU scored 6.7 MFLOPS, up from 4.65

It's a nice feeling when something works especially when it was not expected to solve a previous issue:)

Drewlio77

Last edited by Drewlio77 on 02-Sep-2017 at 03:52 PM.
Last edited by Drewlio77 on 02-Sep-2017 at 05:27 AM.
Last edited by Drewlio77 on 02-Sep-2017 at 05:25 AM.

@Drewlio77

I am impressed! That's exactly the sort of thing I would do then end up with a dead 3640!

_________________
Sam

@-Sam-

Thanks. It was 50% skill and 50% luck. When working on old electronics you never know for sure.

My luck continued...

I noticed the parallel port was not working yesterday when I tried to use ParBench between my A1200 Tower and the 4000. No connection would occur, ever. The motherboard I bought was recapped and the SIMM sockets replaced, so I looked over the board for anomalies. I discovered that the re-capper did 75% good work. Some traces were jumper-ed by the Audio and Parallel/Serial connectors due some bad cap replacement attempts. Luckily I had a dead board I could ohm out to check against. One of the jumper-ed wires was attached to the wrong component. Attempting to resolve this, the surface mount capacitor lifted completely off the motherboard. Apparently the solder pads packed it in from previous soldering mayhem. Luckily I managed to reattach the capacitor in-circuit on the back of the motherboard with a few hookup wires and using the exposed ground plane and solder pads that were available.

All went well. My DSS8+ Audio sampler and ParBench work as expected.

A word to the wise: Most professionally capped motherboards are not professional capped. If buying one, ask for close up, high resolution pictures of the board itself.

Drewlio77

Last edited by Drewlio77 on 03-Sep-2017 at 03:13 PM.

@Drewlio77


awesome! I love it when a plan comes together (of course a few prayers during and before turning power on always help)

_________________
____________________________
c64-2sids, A1000, A1200T-060@50(finally working!),A4000-CSMKIII
! My Master Miggies- Amiga 1000 & AmigaOne X1000 !
mancave-ramblings
X1000 I BELIEVE