X1000 broken (fixed)

[LyleHaze]

Just for the record: TRM 1.5 Page 18, Table 9 says there are 6 LEDs.. There are in fact 8.
Also, for those with less than perfect eyesight, With the board mounted in the factory case,
the LED's are numbered from LED8 on the left to LED1 on the right side.

LED8 and LED7 are driven from each core of Xena, and will glow slightly when Xena is in idle_reset state.

If you will look to the right..
LED1 is labeled "General Power Supply Fault"
LED2 is labeled "CPU Core Supply Fault"

These should NOT be lit on a good day.

Apparently you are not having a good day.

[Slayer]

I wanted to growl Kicko but I'm sure you've already growled at yourself in the mirror a few times by now

All things being equal you should be able to do at least one of these two things:

1) put your system back to the way it was before you decided to upgrade the firmware or insert the new graphics card and/or SSD - and try and boot that.

No good?

2) do as above but get hold of an appropriate AmigaOS4.1.x CDRom for the X1000 (do not worry about new drivers since we don't care about the new graphics card at this point and I don't want to kill 2 birds with one stone thank you ) and boot from the CDRom.

If this doesn't work then it's something more difficult, I'm hoping it is just corrupt firmware or similair since if you can't boot off a CDRom which is the fundamental baseline that the MB should be able to comply with then... anything is up for being the culprit

Next time we will know these two things

1) Do not try to do too many things at once (swapping out the graphic cards in itself would have informed you that it wouldn't work... but since you had added the SSD/firmware requirement as well etc it muddled up simple step by step processes and you ended up in a domino effect)

2) Do not post about this kind of stuff on Amigaworld, most of those replies in the thread are basically useless, they don't lack concern or genuine desire to help they just lack actual knowledge of the X1000 and how it all fits and works together (I know some real knowlegde will be mixed in there with all the rest of the advice but if you or anyone else could determine which was more correct and could be used then you wouldn't need help in the first place) you might get 6 replies in 6 mins but you want your X1000 in one piece don't you?

Ok, good luck and if I and everyone else on this site can help I/We will

[kicko]

and the test continues.....

Tried my radeon 4650, the 4gb ram and my original psu inside a PC computer. They all work so nothing wrong with them. Im typing right now with it. So something is definately wrong with the x1000.

I installed puTTy and connected null modem cable in the serial port PC (com1) to serial on the back of x1000 (have one installed). In puTTy i choose serial (com1) and speed 115200.

The only thing that appears is "ÿ" sign in the shell window when powering on x1000. Nothing more. Never used any serial debug before. Do i need to do anything more, any jumpers etc on the x1000 for this test or is this enough ?

This doesnt look any good Nothing appears on the monitor and led on the motherboard is only light 1 & 2 (psu & cpu). But the psu works on the pc.

I dont see anything burned or exploaded inside the x1000. However i cant see the back of the card as the housing.

About the rescue bios. Is there a hope or with all tests it shows that it wouldnt help ? Curious as i dont know how the bios works.

So i have harddrive (os4) in sata 0 and dvd sata 2. Radeon 4650 connected. Thats all.

How is it with "power sw" (red/white cable next to sata ports). Is it connected to J31 ? Dont remember how i had it in the beginning.

[zappa2009]

I dont see any J31 do you mean P31 ? I did a look in the Nemo-Technical.pdf and there are thease Pinouts i show below.

P30 is Drive-Led

P31 is something with the CPU. Let it free. It is for CPU GPIO2

P33 is the Power-Switch

P36 is the Reset Switch

With Putty use 115200 8 N 1 xON/xOFF cause the X1000 can not handle RTS/CTS with his 3 Pins.

I hope that helps.

[LyleHaze]

At this point, the risk may be that there are too many different opinions on how to proceed.
My first bit of advice would be to check with Chris BEFORE trying anything. AmigaKit is your
best source of reliable information!

Now more opinions:
If LED1 and LED2 are lit, then the processor will NOT run. Period.

Until LED1 and LED2 go out, there's not much sense in trying serial debug, or CFE jumpers, or other "soft" solutions.

Those LED's being lit mean that the power supply is not working. NOT NECESSARILY the power supply box in the bottom of your tower (though it's possible), but very possibly the power regulation system on the NEMO board.

This failure MAY be due to something shorting a power rail. But your previous post mentioned that with nothing but two drives plugged in, LED1 and LED2 were still lit.. so that reduces the possible causes of a short there.

Another post indicated that you might have things plugged in to the wrong connectors.. This could certainly cause a local short. So double checking, or running without any unneeded connectors might verify that. (it would truly be sad if all this is caused by something plugged in to the wrong place).

_IF_ it were me, and _IF_ I had Chris's OK, I would want to try powering up with ONLY a power supply and the power button connected.. We know it won't boot.. but this would remove all external causes of LED1 and LED2 being lit.
If ONLY the power supply and the power button are connected, AND LED1 and LED2 are lit, then you would quickly reduce the possibilities to NEMO or Power supply.

[edit] Yes, I am suggesting powering up with no RAM, no Radeon, no PCI Cards, No drives, No CaseLEDs, No Keyboard, No Mouse, No HUB, No kidding. Just a bare NEMO, a power supply, and the power button.

Good Luck, my friend.

[ddni]

I would also check that a loose screw or foreign body has not slipped behind the Nemo board causing a short.

[Hypex]

The only thing that appears is "ÿ" sign in the shell window when powering on x1000. Nothing more. Never used any serial debug before. Do i need to do anything more, any jumpers etc on the x1000 for this test or is this enough ?

That's a worry. That's what I got on the serial port after I broke my A1 CPU.

Another thing is to make sure the serial port is connected to the debug port. If your machine was factory built it certainly should be. But check the serial port is connected to the top serial header in the mainboard. Right next to the black chip.

This failure MAY be due to something shorting a power rail.

I take it any similar issues were resolved when the FC boards were sent out?

No kidding. Just a bare NEMO, a power supply, and the power button.

That's how I first tested my new board. I also didn't have a case at the time and MacGuyvered a power switch up but let's not go there.

[Slayer]

The only thing that appears is "ÿ" sign in the shell window when powering on x1000.

I do believe that is the first character you see on a serial feed but it is quickly followed by abbreviations in [] brackets

I fear the worst but lets just hope its a combination of elements causing this behavior

[abalaban]

and the test continues.....

The only thing that appears is "ÿ" sign in the shell window when powering on x1000. Nothing more. Never used any serial debug before. Do i need to do anything more, any jumpers etc on the x1000 for this test or is this enough ?

Seeing "ÿ" in the debug terminal can be caused by a misconfigured serial link: which speed? which parity? how many stop bits? How is the flow control?

IIRC on Nemo the configuration should be:
115200 bauds
8 bits
no parity
1 stop bit
no flow control (this last one I'm not sure anymore)

[ChrisH]

no flow control (this last one I'm not sure anymore)

I'm pretty sure that is correct. You don't want your Amiga freezing while it waits for a (potentially non-existant) machine at the other end to respond!