A4000 freezing after a while

8bitbubsy · 01 November 2012, 18:09

My A4000D is acting weird! Let me first make a little summary of what's done to it:
- All caps replaced except for Picasso IV card (PSU, daughterboard, motherboard, accelerator). Replaced with same micro farad values and voltage ratings (or higher).
- No leak damage whatsoever on the A4000 mobo, it's in extremely good condition.
- PSU, on load, measures -11.8v, 11.9v, 4.8v
- The 68040 is running at stock 25MHz on the A6340 v3.2 card, and it has a heatsink *and* a fan.
- Super Buster 11 fitted.
- Cleaned the dautherboard & accelerator contacts and the Picasso IV contacts.
- Running OS 3.9 with all boing bags, at 1024x768 24bpp IIRC

Ok, now to the issue itself -- the A4000 works fine, but after 5-15 minutes the system freezes. Can't move the mouse or use the keyboard, it's all stopped. First thing that comes to mind is CPU overheating, but with a heatsink and a fan at 25MHz it should be rock stable.. I guess the problem lies elsewhere.

Any tips? Could it be RAM-related? Defective CPU? Too low 5v rail? I doubt it, but who knows..
EDIT: I have an idea. Maybe it's the Picasso IV overheating? I'll try to let the system run for a long time without the Picasso IV installed. I'll keep you updated.

roy bates · 01 November 2012, 18:19

try it without the p4 and check the 5 volt rail.

i used to think it was bad to go below 4.85 volts,is it the original supply?

actually does the supply drop for a sec during power on under full load?

was it ok before i assume it was.

8bitbubsy · 01 November 2012, 18:25

I'm too tired to test now, I'll do more testing tomorrow.
Yes, the PSU is original. All caps replaced, it's in very good condition.
Not sure about the voltage stuff, I'll do the tests tomorrow.

roy bates · 01 November 2012, 18:31

ahh ok,one other thing if this is different in any way to what it was before a recap what was the difference.

i mean what was added after if anything.ic's cards etc.

have you got a bare install of the os,just to eliminate a software problem you know minus the boing bags etc.

Damion · 01 November 2012, 20:08

4.8v measured at a molex is a little low IMHO. I've had to bump that up a few times in the past to keep them stable.

8bitbubsy · 02 November 2012, 12:38

Sadly I don't remember the voltages and stability of the unit before I recapped it, it was one of the first things I did to it when I got it.

Doing some more accurate PSU tests:
- Measuring +5v from molex when turning on measures around v4.95v to v4.92 (goes up and down a lot until the floppy drive seek before the OS boots, then it's stable at 4.91v)
- Measuring +12v from molex when turning on measures around v12.90v to v12.56 (goes up and down a lot until the floppy drive seek before the OS boots, then it's stable at 12.91v)

With the Picasso IV card removed, the voltages at stabled state are 4.92v and 11.89v
With the A6340 accelerator removed -as well-, the voltages at stabled state are 4.94v and 11.71v

Sometimes when booting, the floppy drive seeks like mad, at random intervals.

EDIT: Booted without the Picasso IV. The picture is jumping up and down sometimes, and it's getting some interference...I know it's not a problem with my Commodore 1084S monitor.

EDIT2: Pushing hard on the A3640 seems to make it stable sometimes, even though I *did* clean the accelerator connections... I'll take the A3640 to my lab and see if all the caps and network-resistors are fine, as well as take a close inspection.

This is weird. It won't always boot. One time I got a yellow screen, tried to remove all the fast RAM and reseat the chip RAM.

roy bates · 02 November 2012, 13:37

you will probably have to strip the whole system down and set it up outside of the case,pushing down on an accelorator also pushes down on the motherboard would you agree.
maybe theres an underlined problem with the motherboard bad via or just simply a solder joint somewhere near the card itself on one of the ic's

8bitbubsy · 02 November 2012, 15:14

So I did some flux add + solder reflow to the BGAs on the A6340 (some of them were kind of dried), and went through the board and reflowed some other stuff, and behold -- the weirdness is gone! It works just fine now, and I never get the weird floppy drive issues nor any strange signs/picture jumping.

roy bates · 02 November 2012, 15:51

thats great news mate,well done.

01 November 2012, 18:09	#1
8bitbubsy Registered User Join Date: Sep 2009 Location: Norway Posts: 1,710	A4000 freezing after a while My A4000D is acting weird! Let me first make a little summary of what's done to it: - All caps replaced except for Picasso IV card (PSU, daughterboard, motherboard, accelerator). Replaced with same micro farad values and voltage ratings (or higher). - No leak damage whatsoever on the A4000 mobo, it's in extremely good condition. - PSU, on load, measures -11.8v, 11.9v, 4.8v - The 68040 is running at stock 25MHz on the A6340 v3.2 card, and it has a heatsink and a fan. - Super Buster 11 fitted. - Cleaned the dautherboard & accelerator contacts and the Picasso IV contacts. - Running OS 3.9 with all boing bags, at 1024x768 24bpp IIRC Ok, now to the issue itself -- the A4000 works fine, but after 5-15 minutes the system freezes. Can't move the mouse or use the keyboard, it's all stopped. First thing that comes to mind is CPU overheating, but with a heatsink and a fan at 25MHz it should be rock stable.. I guess the problem lies elsewhere. Any tips? Could it be RAM-related? Defective CPU? Too low 5v rail? I doubt it, but who knows.. EDIT: I have an idea. Maybe it's the Picasso IV overheating? I'll try to let the system run for a long time without the Picasso IV installed. I'll keep you updated. Last edited by 8bitbubsy; 01 November 2012 at 18:17.

01 November 2012, 18:19	#2
roy bates Registered User Join Date: Apr 2011 Location: birmingham Age: 55 Posts: 2,827	try it without the p4 and check the 5 volt rail. i used to think it was bad to go below 4.85 volts,is it the original supply? actually does the supply drop for a sec during power on under full load? was it ok before i assume it was. Last edited by roy bates; 01 November 2012 at 18:26.

Similar Threads
Thread	Thread Starter	Forum	Replies	Last Post
WHDLoad freezing	Rabbit80	support.Hardware	12	10 June 2009 20:45
WINUAE 1.4.4 freezing	DDNI	support.WinUAE	3	24 September 2007 17:40
syndicate freezing	pbareges	support.Games	5	07 March 2007 17:41
Aweb Freezing up	jambomiga	support.WinUAE	3	18 September 2003 20:15
Freezing UP	Damage	support.WinUAE	2	14 January 2003 14:01

01 November 2012, 18:25	#3
8bitbubsy Registered User Join Date: Sep 2009 Location: Norway Posts: 1,710	I'm too tired to test now, I'll do more testing tomorrow. Yes, the PSU is original. All caps replaced, it's in very good condition. Not sure about the voltage stuff, I'll do the tests tomorrow.

01 November 2012, 18:31	#4
roy bates Registered User Join Date: Apr 2011 Location: birmingham Age: 55 Posts: 2,827	ahh ok,one other thing if this is different in any way to what it was before a recap what was the difference. i mean what was added after if anything.ic's cards etc. have you got a bare install of the os,just to eliminate a software problem you know minus the boing bags etc.

01 November 2012, 20:08	#5
Damion Registered User Join Date: Mar 2008 Location: US Posts: 315	4.8v measured at a molex is a little low IMHO. I've had to bump that up a few times in the past to keep them stable.

02 November 2012, 12:38	#6
8bitbubsy Registered User Join Date: Sep 2009 Location: Norway Posts: 1,710	Sadly I don't remember the voltages and stability of the unit before I recapped it, it was one of the first things I did to it when I got it. Doing some more accurate PSU tests: - Measuring +5v from molex when turning on measures around v4.95v to v4.92 (goes up and down a lot until the floppy drive seek before the OS boots, then it's stable at 4.91v) - Measuring +12v from molex when turning on measures around v12.90v to v12.56 (goes up and down a lot until the floppy drive seek before the OS boots, then it's stable at 12.91v) With the Picasso IV card removed, the voltages at stabled state are 4.92v and 11.89v With the A6340 accelerator removed -as well-, the voltages at stabled state are 4.94v and 11.71v Sometimes when booting, the floppy drive seeks like mad, at random intervals. EDIT: Booted without the Picasso IV. The picture is jumping up and down sometimes, and it's getting some interference...I know it's not a problem with my Commodore 1084S monitor. EDIT2: Pushing hard on the A3640 seems to make it stable sometimes, even though I did clean the accelerator connections... I'll take the A3640 to my lab and see if all the caps and network-resistors are fine, as well as take a close inspection. This is weird. It won't always boot. One time I got a yellow screen, tried to remove all the fast RAM and reseat the chip RAM.

02 November 2012, 13:37	#7
roy bates Registered User Join Date: Apr 2011 Location: birmingham Age: 55 Posts: 2,827	you will probably have to strip the whole system down and set it up outside of the case,pushing down on an accelorator also pushes down on the motherboard would you agree. maybe theres an underlined problem with the motherboard bad via or just simply a solder joint somewhere near the card itself on one of the ic's

02 November 2012, 15:14	#8
8bitbubsy Registered User Join Date: Sep 2009 Location: Norway Posts: 1,710	So I did some flux add + solder reflow to the BGAs on the A6340 (some of them were kind of dried), and went through the board and reflowed some other stuff, and behold -- the weirdness is gone! It works just fine now, and I never get the weird floppy drive issues nor any strange signs/picture jumping.

02 November 2012, 15:51	#9
roy bates Registered User Join Date: Apr 2011 Location: birmingham Age: 55 Posts: 2,827	thats great news mate,well done.

Currently Active Users Viewing This Thread: 1 (0 members and 1 guests)