Entering the 68k ecosystem with a Centris 650 - what to check?

[David Cook]

I just picked up a Centris 650. There isn't enough notable to start a new thread, so I'll just piggyback here.

The CPU an HRC variant, which is interesting. (The factory heatsink and solder paste was removed prior to photographing.)



Upon powering up, the computer emitted *most* of the happy boot chime followed by a 'clipping' and distorted end to the chime. After that the computer did not show video nor did it make a sad Mac sound. About 10 seconds later, a burning component smell filled the air. Removing the cover did not show any damage. Rebooting with a Flir infrared camera targeted the following location. The camera was unnecessary, as smoke appeared at power up around these parts. Can you guess what's wrong?



Capacitor C34 is a factory tantalum. Yet, apparently waking up from such a long slumber killed it. It doesn't appear burned or burst, because it failed only 'mostly short'. There's a big difference between 'mostly short' and 'all short'. It measures about 16 ohm in resistance.

Replacing the capacitor fixed the problem and the computer starts up and passes all the tests. I'm surprised that the amount of current present to emit smoke and internally burn a component did not damage the nearby traces.

 

[joshc]

I just picked up a Centris 650. There isn't enough notable to start a new thread, so I'll just piggyback here.

The CPU an HRC variant, which is interesting. (The factory heatsink and solder paste was removed prior to photographing.)

View attachment 69649

Upon powering up, the computer emitted *most* of the happy boot chime followed by a 'clipping' and distorted end to the chime. After that the computer did not show video nor did it make a sad Mac sound. About 10 seconds later, a burning component smell filled the air. Removing the cover did not show any damage. Rebooting with a Flir infrared camera targeted the following location. The camera was unnecessary, as smoke appeared at power up around these parts. Can you guess what's wrong?

View attachment 69650

Capacitor C34 is a factory tantalum. Yet, apparently waking up from such a long slumber killed it. It doesn't appear burned or burst, because it failed only 'mostly short'. There's a big difference between 'mostly short' and 'all short'. It measures about 16 ohm in resistance.

Replacing the capacitor fixed the problem and the computer starts up and passes all the tests. I'm surprised that the amount of current present to emit smoke and internally burn a component did not damage the nearby traces.

I’ve had a similar problem with a 610 before. Did you also have to replace the resistors next to it?

 

[David Cook]

Did you also have to replace the resistors next to it?

No. I was concerned that they suffered some heat damage. However, given how large they are, and since the machine worked and passed all the tests, I figured the exposure must have been brief enough to avoid any significant effect.

 

[cheesestraws]

The CPU an HRC variant, which is interesting.

I can only see HRC as "High Rupture Capacity" which, while a desirable property of a CPU probably isn't what is going on here. What does HRC mean here?

 

[finkmac]

Supposedly it runs cooler than an XC68040RC25

 

[David Cook]

What does HRC mean here?

@MrFahrenheit and @Fizzbinn also have a 25MHz 02E31F, which they also found in a 650. There is speculation that this runs at higher temperatures or produces less heat. I have not found a primary source (i.e. Motorola datasheet) that defines 'HRC'.

 

[MrFahrenheit]

@MrFahrenheit and @Fizzbinn also have a 25MHz 02E31F, which they also found in a 650. There is speculation that this runs at higher temperatures or produces less heat. I have not found a primary source (i.e. Motorola datasheet) that defines 'HRC'.

I found information in the past that H variants were intended for government / military so perhaps it means “hardened” meaning it’s certified for use in space?