SGI: Hardware

o.k. so i got another octane here.

switch on:

white > piano > red > spinup > white.

all good you say.

no video output, it seems to be generating sync pulses but no picture.
also - NO SERIAL CONSOLE OUTPUT!!!!!

after about 10 attempts the console outputed "diagnostic failure - press any key to continue" or something like that.

pressing any key of course did nothing.


IDEAS?

another thing, when i power it down there is a tiny sizzling type sound coming from inside somewhere.
does the octane keep one of the fans running for a while at powerdown or is something *very* wrong somewhere.

this last problem has me worried - do i need to keep the extinguisher around it? i'd rather take it back to the kitchen.

It wouldnt happen to have the 747watt Cherokee power supply would it? mine makes the same sizzling noise when the system is off and so does both my O200 towers.

Ummm...neither of my O200 towers or my 747watt Cherokee P/S make sizzling noises. The have a low hum from the transformers, but nothing that I would describe has a sizzling sound.

it is a cheroke, an 060-0035-001-b.

so now i have to strip it and look for tracking on the board - great

still dousnt explain the other stuff, maybe a total strip'n'clean inc xbow will fix it.

Intel-OUTSIDE wrote: after about 10 attempts the console outputed "diagnostic failure - press any key to continue" or something like that.

pressing any key of course did nothing.


IDEAS?

I guess you tried the reseating game already? Also, what do the diagnostic leds behind the front door say? They aren't too helpful in most cases, but still..



where QA-QD are the four XIO slots, Heart and BaseIO are IP30 and Frontplane - so those two should always be lit.

Mare wrote: It wouldnt happen to have the 747watt Cherokee power supply would it? mine makes the same sizzling noise when the system is off and so does both my O200 towers.

yeah, my o200 PSU does that too. I was quite irritated when I first switched the PSU on after I had received the machine Both my cherokee Octane PSUs are silent when the fan is not spinning though..

diagnostic leds were the first thing i went for, no problem there.
heart,i/o and qa all on.

i am going to rip this thing right down at the weekend, the sizzle noise is not aceptable.
i consider it dangerous because it sounds like ht tracking.

if i find the problem i'll post the details for everybody.

Btw, whats the difference between SE and SSE apart from the size and tram sockets, they seem to be very expensive now (SSE)

SSE in simple terms has 2 SEs deviding the picture in 2 and both doing half. SSE is therefor a lot faster. SSE can also do a lot more double buffering since both Geomitry Engines have RAM of their own.

On your PSU, most switching power supplies have this "sizzling" sound, it's usually a
loose part in the transformer, but could also indicate leaking high voltage. If you have
another PSU you could try if your machine will power up with that, on peecees a lot
of the unstability of older machines is due to bad PSUs.

WhizzMan wrote: On your PSU, most switching power supplies have this "sizzling" sound,

no they dont, i am an electronics engineer working with "switchers" since the mid 80's,
they either run silent, or the chokes can hum if the windings are loose.
*they never sizzle*

the sizzle is a high-voltage leak-over on the pcb.

it is going to be a lot more common in europe because our rectified mains is around 380vdc whereas in america and japan it's only around 180-200.

the problem is when a spark tracks across a pcb it leaves a tiny burn.
as this continues the layers build-up until you have a strip of burned carbon that is conductive.
then things get dangerous.

Intel-OUTSIDE wrote: it is going to be a lot more common in europe because our rectified mains is around 380vdc whereas in america and japan it's only around 180-200.

I always wonder why Europe goes with 220v. It's so much more dangerous than 110 for household use. I *know* why we do it in China - people are cheap ! And they're also cheap to replace ... but Europe should know better.

a higher voltage can use a thinner and longer cable.
also, you have to step the power up less in things like tv's

brownouts are almost unheardof in europe also, because a smalll (20-30v) dropout is less noticed.

if you look at country's with 100/110v they have transformers *everywhere*, we dont.
thats because you have to keep the cables short at 110v to avoid ecessive voltage-drop.

Intel-OUTSIDE wrote: brownouts are almost unheard of in europe also, because a smalll (20-30v) dropout is less noticed.

if you look at country's with 100/110v they have transformers *everywhere*, we dont.
thats because you have to keep the cables short at 110v to avoid ecessive voltage-drop.

Yeah, sure ... but when you're lying there dead on the floor all these cost-saving measures aren't that kewl, ya know what I mean ? The bunny used to chew the cords around here so that I've zapped myself really good a few times (china is 220v 50hz also but we do have brownouts) ... with 110 it's just a mild tingle. Or even if it's a good shock, it's not normal to die from it. Which brings up something I've been wondering about - are bunnies impervious to electricity ? that little doe must have chewed through most of the extension cords in the house without showing any obvious damage. Not that you could tell much, with a bunny, but still ....

btw, it's 220 to the drop at the house, so the "long 110v lines" part doesn't hold water. We could discuss the same safety question with the 50hz - 60hz variation, too. You guys just like to live dangerously, I think

SHIT!!!
shitshitshitshitshit

serial diagnostic decides to report this:


Starting diagnostic program...

Press <Esc> to return to the menu.


Exception: <vector=Normal>
Status register: 0x34005083<CU1,CU0,FR,IM7,IM5,IPL=???,KX,MODE=KERNEL,EXL,IE>
Cause register: 0x801c<CE=0,IP8,EXC=DBE>
Exception PC: 0xa8000000208c79d8, Exception RA: 0xa8000000208ce370
Data Bus error
HEART ISR : 0x204000000000000<BASEIO,TIMER>
HEART IMSR: 0x0
Bridge(id=0xf) ISR: 0x800<PCI_MST_TIMEOUT>
PCI/GIO error: 0x80001c<DEV_NUM=0x0,ADDR=0x80001c>
VID 0's saved user regs in hex (gpda=0xa80000)
Cannot connect to keyboard -- check the cable.
Cannot open keyboard() for input
Cannot connect to keyboard -- check the cable.
Cannot open keyboard() for input


System Maintenance Menu

1) Start System
2) Install System Software
3) Run Diagnostics
4) Recover System
5) Enter Command Monitor

Option? 5
Command Monitor. Type "exit" to return to the menu.
>>

>> hinv -v
System: IP30
Processor: 300 Mhz R12000, with FPU
Primary I-cache size: 32 Kbytes
Primary D-cache size: 32 Kbytes
Secondary cache size: 2 Mbytes
Memory size: 128 Mbytes
Network: ef0 ethernet (100/10 base-T)
SCSI Disk: scsi(0)disk(1)
Audio: RAD Audio Processor
Graphics: ESI
>>
>> exit


i was sure it had 256megs of dram, maybe one is bad, but i dont have any spares here - SHIT!

an update,

i cant find whats making the noise in the psu yet,

i stripped and cleaned *everything*.

i found that the fan on the xbow was sticking - WATCH FOR THAT ONE - IT MAY EFFECT THEM ALL.
oil seems to have fixed that problem.

the original problem remains - i am 99% sure it's the IP30 so i am going to re-heat all the balls on the heart with some smt-flux.

wish me luck.

o.k. it's dead.

either the ip30 or the xbox is finished.

dont have another to swap parts so it has to wait.

Intel-OUTSIDE wrote: the xbox is finished.

That's just too funny for somebody that goes by the handle Intel-OUTSIDE
I guess you meant XBOW. And, sorry for your loss.