IIci with potentially dead 50MHz oscillator.

[tattar8]

I'm trying to resurrect a IIci which had a Maxell bomb go off, which I caught before the damage was too bad, and only really ended up eating a couple of traces which I patched; I then recapped the entire board.  Using the ATX power supply mod, I'm able to get power to the board, but I have no signs of life other than a small pop from the speaker.  I tried probing random address+data lines with an oscilloscope hoping to see some activity, but no dice.  I'm now suspecting the mildly rusted 50MHz oscillator on the bottom right corner of the board, near where the battery was.  Using the oscilloscope, I got nothing meaningful from its output pin, despite getting correct-frequency waves from the other three oscillators.  Is that oscillator (I believe it's Y6) the main CPU clock?  If it is, then I'm guessing I'd need to replace it.  Does anyone know where I can get a replacement?

 

[Fizzbinn]

I have a IIci in almost the exact same condition (Maxell bomb, recapped, etc), I found replacement 50MHz oscillators on eBay:

https://www.ebay.com/itm/Vishay-Dale-XO-53B-50-000M-Oscillator-50MHz-5V-TTL-DIP-14-Qty-4/181289637274

Unfortunately mine's still not booting. The power-on signal from the keyboard works, I get the "pop" from the speaker, the monitor comes to life (with square not rounded corners), but no start up chime or flashing disk icon. I figure something else is broken and/or I need to look closer as the traces...

FYI I Installed the replacement oscillator in socket, I used the same oscillator and sockets to overclock a IIsi successfully. I swapped oscillators with that IIsi to check the replacement oscillator wasn't bad. 

https://www.amazon.com/Valuepro-2001-2X7-4CP-Machine-Full-Can-Oscillator/dp/B00B88A9TA

 

[tattar8]

I got the replacement oscillator and now I have a 50MHz wave, but like the above poster, the IIci still doesn't boot.  I tried probing pin 6 of the 68030, which is supposed to be the clock input, but there's nothing there.  Does anyone know where that's supposed to connect?  I looked at the schematic, but it just says it connects to "MPU CK", with no other mention of that label.  I'm guessing it's getting divided by 2 from the 50MHz crystal that died, but what's supposed to be doing that division?

 

[Bolle]

The CPU clock signal is coming from the MDU at pin 106. It goes through R89 which has 47ohms so your meter might not pick this up in continuity mode.

The 50MHz signal from the oscillator should go to pin 107 on the MDU.

You can find all that on the "clock" page (sheet 7/7) in the schematics.

 

[tattar8]

Found it; I'm actually reading an open circuit across R89 as well as R87.  Continuity between MDU-106 and R89, and between R89 and CPU-6 are both fine, and I can see the 25MHz clock on the MDU side of R89.  That would probably explain it.  What size SMD resistor would I need to replace it with?  I don't have a ruler handy to measure it myself with.

 

[Bolle]

Just remove and bridge them with a blob of solder for testing purposes.

The resistors are there to suppress any noise or ringing that might occur on the clock line.

 

[tattar8]

Alright, did that, and now the 68030 is getting a clock signal (it should be a sin wave, right?).  But I'm still not getting any signs of life out of it -- no chimes of any kind, no activity on the address lines.  If I hit the reset button, all the address lines, as well as the AS signal, go high as long as I hold the button, and then go back low.  Is it possible that the 68030 itself is dead?  Is there something else I should check for?

 

[Bolle]

it should be a sin wave, right?

I would imagine that cutting the 50MHz signal in half is more of a digital operation going on inside of the MDU so I would imagine it should actually be more of a square wave.

Is the signal you are measuring actually 25MHz-ish? Otherwise the MDU might be broken if it just passes through the sine wave signal from the oscillator. Not sure if and how that would happen though. Overvoltage on the clock input pin?

Somehow the resistors must have died as well if they are indeed open circuit so this might be an option that something bad happened there.

Sidenote: I once had a IIci board with a defective CPU. The board would work with a PowerCache or Turbo040 installed but wouldn't do anything without them.

No idea how you would easily detect a dead CPU though... remove ROMs and see if it walks the address bus maybe?

Did you check if the system is actually coming out of reset correctly? This is handled by one of the Sony sound chips. What's the state of the reset pin on the CPU when it's just sitting there?

Edit: btw did you check all address and data (and enable) lines going to the ROMs? Might have missed one of them and they tend to get damaged if the battery goes off in the IIci.

Alternatively you could stick in a ROM SIMM if you have one to see if that changes anything.

 

[tattar8]

I'm measuring a perfect 25MHz sine wave on the clock pin and a perfect 50MHz sine wave on the oscillator output.  ARST idles high and goes low when I hit the reset button.

I already had to jump all three address lines to the ROMs that ran under where the battery was.  

I'll check the address/data lines, but shouldn't I see some activity on them even if one or two are disconnected?

 

[tattar8]

I don't know if this is normal, but I'm seeing DC offsets on all the clocks, not just the 50MHz. 

Here's a shot of the output of the 50MHz oscillator:  

And here's a shot of the clock input to the 68030:  https://i.imgur.com/yeeY1FF.jpg