@Morphix

Hi,

Memory module:

Kingston ValueRam

KVR133X72RC3/512

Additional text, not so important:

9962254-008.A01
1049765-3,3V

Amazon:

http://www.amazon.com/Kingston-KVR133X72RC3-512-64MX72-133MHZ/dp/B00006HN6J

Ebay:

http://www.ebay.com/sch/i.html?_trksid=p5197.m570.l1313&_nkw=+KVR133X72RC3%2F512&_sacat=0&_from=R40

Battery is ordinary CR2032, so called clock-battery


In my SE it is located right down corner of mobo.


Before you start...

Be carefull with static electric.

Leave power cord to PSU. Switch PSU off from switch. Now your mobo is without current, but still grounded. ATX-mobos has 5V on, if PSU is on.

Before you touch inner parts of computer, touch computer case. That way, if you are electrically charged, it goes away to computer case, not to motherboard to destroy components.

Happy hacking... :)

@mlehto

Thanks a lot!!!!!!

YOU ARE THE MAN!

I have taken all those protection steps you mentioned! Thanks!

I opened the little lady, switched RAM to a different module. Rebooted.

Now the machine is on it is on irc it runs xmame, it plays a video, OWB is also open and.... STILL no crash -10 minutes now-

Lets see.


HOW the HELL do I pop out this battery -found where it is-?. Forgive the newbie question.

_________________
Peg1, Peg2, Efika,
Amiga 500, Amiga 1230,
Amiga 4000PPC,
Sam 440EP, AmigaOne XE,
MacMini PPC, PowerMac G4, G5

@Morphix

Thanks :)

I dont know ... I used little clockwork screwdriver. Carefully. If you have multimeter, you can measure it easily. Plus to battery and ground to mobo screw.

Psu off again.

If you switced to other memory slot, it could be, that simm is ok, but socket is broken. Or condensator(s) of other slot is bad.

@mlehto

Thanks man!

No it is frozen... again :(, SO... I am on my way to check battery and test a different memory module.

As I take some hours off work I managed to on the place where I have this machine stored and I am testing now. Ok on my way to ebay to order some memory.

Ah so I need a clockwork screwdriver to take the battery out. OK!

Last edited by Morphix on 26-Feb-2013 at 01:40 PM.

_________________
Peg1, Peg2, Efika,
Amiga 500, Amiga 1230,
Amiga 4000PPC,
Sam 440EP, AmigaOne XE,
MacMini PPC, PowerMac G4, G5

@mlehto

Are those electrolytic caps? If so, easy to see if they've boiled by the boob at the end (C-cup bad, A-cup suspect, flat as an airstrip ok). If those are the problem, replacing them is also rather easy, for a trained electronics technician

These boards are getting into the age domain where dried out caps should start appearing.

_________________
This weeks pet peeve:
Using "voltage" instead of "potential", which leads to inventing new words like "amperage" instead of "current" (I, measured in A) or possible "charge" (amperehours, Ah or Coulomb, C). Sometimes I don't even know what people mean.

@olegil

I think that they are aluminium cap. So they leaks, and it is not so easy to recognize. But correct, if activelly used, they start to be in age, where replacing is needed.

Did you got my mail ?

Last edited by mlehto on 26-Feb-2013 at 02:34 PM.

Lets hope that after changing battery and memory the problem will be fixed.

_________________
Peg1, Peg2, Efika,
Amiga 500, Amiga 1230,
Amiga 4000PPC,
Sam 440EP, AmigaOne XE,
MacMini PPC, PowerMac G4, G5

Some updates:

This is a post from another forum:

Quote:

i had terrible stability issues with my SAM and my peg after update 3 and onwards once USB2 support was added to the system. under heavy CPU loading -- such as browsing using MUIOWB or decoding a high-resolution video -- i would get random freezes. turns out the issue was the USB stack not getting enough CPU time (or something) and not 'seeing' my USB keyboard/mouse for certain periods. it then deleted the devices from its tree, thinking they had been disconnected when they weren't. others have reported similar problems with USB2 support in conjunction with USB2 PCI cards using the VIA VT6212L chipset. the bottom line is that i disabled EHCI support in my kicklayout file. this means reverting to USB1.1 speeds, but i got stability back. try editing your SYS:Kickstart/Kicklayout file and commenting out the line 'MODULE Kickstart/ehci.usbhcd' and see if that helps. it sure helped me. -- eliyahu

Following those steps I managed to have an almost stable system again. Still doing some testing but I guess USB stack is the big deal.

I will of course change the battery and test with a different RAM module as well.

Keep you posted!

Last edited by Morphix on 27-Feb-2013 at 08:54 AM.

_________________
Peg1, Peg2, Efika,
Amiga 500, Amiga 1230,
Amiga 4000PPC,
Sam 440EP, AmigaOne XE,
MacMini PPC, PowerMac G4, G5

@Morphix

Awesome if this fixes things, but rather horrible that a USB stack would stop working under heavy load. This USB thingie is so much worse than the CPU cost thingie that I don't even see a way out. Until AOS4 actually gets a usable USB implementation, we're screwed.

_________________
This weeks pet peeve:
Using "voltage" instead of "potential", which leads to inventing new words like "amperage" instead of "current" (I, measured in A) or possible "charge" (amperehours, Ah or Coulomb, C). Sometimes I don't even know what people mean.

@mlehto

Got it, thanks.

Seems you have a later revision which didn't need the PCI GNT/REQ swap then. I'll have to sit down and recreate that myself it seems.

Other than screw up your IDE, what does that mod you have do?

And does the first wire go TO the screw hole? Why not just wire it to GND closer to the GAL? To me it looks like it goes through a hole drilled in the side of the screw hole, possibly to something at the back.

Last edited by olegil on 27-Feb-2013 at 09:05 AM.
Last edited by olegil on 27-Feb-2013 at 09:01 AM.
Last edited by olegil on 27-Feb-2013 at 09:00 AM.

_________________
This weeks pet peeve:
Using "voltage" instead of "potential", which leads to inventing new words like "amperage" instead of "current" (I, measured in A) or possible "charge" (amperehours, Ah or Coulomb, C). Sometimes I don't even know what people mean.

@olegil

Quote:

olegil wrote: @mlehto Got it, thanks. Seems you have a later revision which didn't need the PCI GNT/REQ swap then. I'll have to sit down and recreate that myself it seems.

So you have the same A1SE revision as I have (without the GAL)?

Quote:

And does the first wire go TO the screw hole? Why not just wire it to GND closer to the GAL? To me it looks like it goes through a hole drilled in the side of the screw hole, possibly to something at the back.

I guess it goes to another signal on southbridge, but pictures of the bottom side of the board would be really helpful here (in case you don't have them). This fix looks like an early version of the uA1 DMA fix, which comes in the form of an ispGAL, IIRC.

BTW: I think we are in the wrong thread.

_________________
A1SE: G3@600MHz, 2GB, 1GBit network card
A1XE: G4@933MHz, 2GB, refitted AC'97 codec
microA1: G3@800MHz, 1GB

- A1 Linux support -

@Geri

Yes, we are. Do you have the mod I need, though?

_________________
This weeks pet peeve:
Using "voltage" instead of "potential", which leads to inventing new words like "amperage" instead of "current" (I, measured in A) or possible "charge" (amperehours, Ah or Coulomb, C). Sometimes I don't even know what people mean.

@olegil

I have the board with only two patch wires on the bottom site and without the GAL. But I yet have to take some (high-resolution) pictures. I can do that on the 9./10. of March.

_________________
A1SE: G3@600MHz, 2GB, 1GBit network card
A1XE: G4@933MHz, 2GB, refitted AC'97 codec
microA1: G3@800MHz, 1GB

- A1 Linux support -

@Geri

No rush, just take your time. I'll set up a mass spamming system on your PM inbox to kick in on the 11th, then?

_________________
This weeks pet peeve:
Using "voltage" instead of "potential", which leads to inventing new words like "amperage" instead of "current" (I, measured in A) or possible "charge" (amperehours, Ah or Coulomb, C). Sometimes I don't even know what people mean.

@olegil

Quote:

olegil wrote: @Geri No rush, just take your time. I'll set up a mass spamming system on your PM inbox to kick in on the 11th, then?

_________________
A1SE: G3@600MHz, 2GB, 1GBit network card
A1XE: G4@933MHz, 2GB, refitted AC'97 codec
microA1: G3@800MHz, 1GB

- A1 Linux support -

Some quick updates together with a huge thanks to everyone helped. The system now runs 100% stable. Removing the defective usb stack improved stability but what brought the system back to normal for good was the replacing of the old prism2.device that I was ussing on my wireless pci network card, with the standard Amigaoneethernet driver plus an ethernet cable to the ethernet port of the lady :)

I have faced similar crushes on other Amiga systems ussing this old driver, so this was a good food for thought and a chance for some more testing, baving still some crushes while ussing AmiFTP made me relise tht the problem could be also located to somewhere else and not only on the usb stack. I have not however tested the newer prism2.device, will do it soon, lerhaps no crushes in this version.

I will of ourse change the battery just in case and keep you posted.

Thanks again.

_________________
Peg1, Peg2, Efika,
Amiga 500, Amiga 1230,
Amiga 4000PPC,
Sam 440EP, AmigaOne XE,
MacMini PPC, PowerMac G4, G5