Falcon030 dead after Centuriontech CT60/CT60PSU incident - investigation begins

[alexh]

Have the prices of Falcon chipsets been affected by the reFalcon project yet? Last time I looked :

Combel (C302284) $45
Videl (C302829) $45
DSP (C303002) $45
SDMA (C302996) $45
SCSI (C302897) $45
Codec (C303170) $40
GALS $100

And that's only about half the chipset, you've got the ST chips YM, LMC, MFP etc

[exxos]

Expensive yeah.. I've been down that road before with people. They just keep throwing cash at the thing but eventually I just decline to work on it after awhile. Things fast turn into a money pit and I don't like taking people's money if it's not getting any results. If the main chips have to be replaced, I'd probably not want to spend the time on it at that point.

[Steve]

Have the prices of Falcon chipsets been affected by the reFalcon project yet? Last time I looked :

Last time I checked a few years ago they were those prices, so it seems they haven't changed.

[peters]

I did have one thought as I drove back from CL5.

Would the switch design be slightly safer with usa 120V than it is with 240V ?

[exxos]

Would the switch design be slightly safer with usa 120V than it is with 240V

Probably a bit, but I can't prove it, so please read this as reasoning and not a finding.

Lower voltage puts less stress on anything that can break down: the gaps between contacts, the gap between the two poles, and any dirty or carbonised path on the board. Peak voltage on 120V mains is about 170V, against about 325V on 230V mains, so it takes a lot less on 230V to make a spark jump a gap or a surface track start to conduct.

The catch is current. For the same power, a supply on 120V draws roughly double the current it does on 230V. The label on the switch says 2A at 250VAC and 5A at 125VAC, so on paper it has more current headroom on 120V. Those are resistive ratings though, so they don't tell us how it copes with the switch-on inrush of a switch-mode supply on either voltage.

What we don't know yet is what actually happened inside the switch, how much current the supply draws, or what the inrush looks like. Measuring the real current and inrush at the switch would settle that. Until then I can't say whether 120V would have made any difference here. Not that either voltage should end in fireworks.

[exxos]

Serial adapter came just. Verified it worked on my own Falcon.. So it seems a Falcon without a Videl doesn't appear to want to do much So thats super annoying !

Got the scope on XAS and ROM CE and does at least seem to be some signs of life..

IMG_5325.JPG (80.88 KiB) Viewed 55 times

IMG_5326.JPG (76.15 KiB) Viewed 55 times

RESET doesn't seem to go low for a second time, so it doesn't look like its getting as far as issuing the reset instruction from ROM, which is pretty much the first instruction it runs from ROM.

[PhilC]

Good that its got some activity, like its waiting for its next link in the boot up chain.

[exxos]

Good that its got some activity, like its waiting for its next link in the boot up chain.

RESET doesn't seem to go low for a second time, so it doesn't look like its getting as far as issuing the reset instruction from ROM, which is pretty much the first instruction it runs from ROM.

[PhilC]

Good that its got some activity, like its waiting for its next link in the boot up chain.

RESET doesn't seem to go low for a second time, so it doesn't look like its getting as far as issuing the reset instruction from ROM, which is pretty much the first instruction it runs from ROM.

I guess you'll need to hook up logic analyser to see, or do the boring bit of probing all the address and data lines

[exxos]

Well thats odd. Unplugged the diag-cart, then nothing on XAS at all.. plugged the cart back in , still nothing.

EDIT:

Now theres more activity

IMG_5327.JPG (65.75 KiB) Viewed 31 times