Dead A4000, Need help diagnosing problem.

[alberonn]

I have an A4000 desktop motherboard in a Power Tower case. It has the older daughterboard. (not the one with the PCI slots.) It also has an A2091 with Guru ROM, Picasso IV, and X-Surf II sitting in the Zorro slots. I have two Hard drives hooked yo to the motherboard, a CD-ROM and a IDE ZIP connected to the X-Surf, and I had a old CD Burner hooked up to the A2091. (I now have a Jaz drive hooked up and the bad burner removed, but my computer still won't fire up.)

The problem is this, my A4000 is usually on all of the time. However we had guests here and I shut it down so I wouldn't bug them with the whine of the harddrives. When I try to fire it up I have a dim power light and the drive light goes on for several seconds (I think under a minute,) and nothing happens. I'm hearing the drives spin up, but nothing. The LEDs on the X-Surf II light up as well.

I have disconnected all of my expansion boards and still the same thing. After reconnecting the boards and unplugging my hard drives I still have the same problem.

I suspect possibly the Power supply. However, I'm, not sure how to tell if it's the PS or my motherboard.

Before this I would occasionally see my Amiga with the power light dim, but a reboot usually fixed that problem. When the light is dim, it isn't working.

Someone please help!   :cry:

[alberonn]

I don't know if this info is important, but it does have the stock CBM '040-25 mHz board in it.

[Matt_H]

So, it was working, you turned it off, and now it isn't?

I've found the top A4000 problems to be either battery damage or a loose CPU board.

[alberonn]

Matt_H wrote:
So, it was working, you turned it off, and now it isn't?

I've found the top A4000 problems to be either battery damage or a loose CPU board.

Correct. It was was working fine and now it isn't. If it's battery damage, would it kill the board entirely and so suddenly? As to the CPU board, it doesn't get moved or vibrated much, if at all. But I can try inspecting things after our guests leave.

[LoadWB]

My first instinct on this is power supply.  I see similar behaviour in PCs fairly often.  I had a Sun server do the same : damn thing ran for four years straight, had to shut down one night and it refused to come back on.  Replaced the power supply, came back up just fine.

Unfortunately, sometimes a bad PSU can take out an otherwise good board, too  :cry:

I am in the process of replacing my desktop Amiga PSUs with mini-ATX units.  Well, I would complete this process if AmigaKit would start selling 2000-ATX adapters!  ;-)

[zipper]

The pins of power connector to the mobo may have bad contact/corrosion - a known behaviour especially on expanded machines, and cause resistance to current flow.

[Hodgkinson]

It might be worthwhile purchasing a Digital multimeter and checking out the supply rails to the mainboard - Between black and yellow it should be no less than 11v and between black and red it should be no less than 4.5v. If theres any other colour wires, let us know.

Also, although the RTC battery shouldn't cause the problem since it doesn't power a BIOS like in PC's, it might be worthwhile checking that it hasn't gone below 2.7v.

[alberonn]

Well, my guests have left and I was finally able to open up my case. The one thing about opening my Amiga up that much is that I have a lot of disassembly to do. I have to open up my case, pull all of my expansion boards and and unplug a few cables (And try to remember where and how all of this goes back in.)

First the good news, my problem was caused by my processor board being loose. That bit of foam rubber that Software Hut had in my case was not sufficient to hold it together. After I was able to reassenble, my system came up. Though one of the HDs had protested at being off for a while. But it fired up after a couple of warm boots. I think it's time to back up my system.

For the bad news is that my battery was getting a little fuzzy. I'll have to get a tool to cut it off. Since I usually leave my system on and use FACTS to set my clock, that's not too much of an issue. (I'm looking for a better replacement for FACTS to set my clock though.) Getting to the battery did require me to remove the daughter board. That involved several screws and an extremely tight-fitting board. It'll have to opened up again soon before the battery eats my board.

[TjLaZer]

Yes remove the battery ASAP as it can and will ruine the RTC of the board.  It can also ruine the SIMM slots.  You may find if you replace the baord that it will not keep/hold time, this is due to the traces that were eaten from even the slightest battery leakage!  I had 2 do that so I know.

[Matt_H]

@ alberonn

Glad your system's back up.

I looked at FACTS back when I was battery-less, but it didn't appeal to me - I wanted something more transparent. I used a previous incarnation of this program.