tf530 build issues

[gumorr]

I've been building a tf530, and i've been having issues getting it to boot.
Using diagrom, it seems to at least execute instructions, but just hangs really early on. It's really inconsistent where it hangs: sometimes i just get a black screen, other times it manages to brighten the power led, and occasionally i get a purple or cyan screen.
I think my soldering is fine, i can't find any shorts. I've also swapped out the 33 mhz oscillator i was originally using for a 25 mhz one, but it still doesn't boot. This thing has me stumped. Anyone have any ideas what could be going wrong?

[terriblefire]

I'd suggest back to the 33Mhz osc to start with. Then see what you get on the serial port with Diagrom.

[gumorr]

At both 25 and 33 mhz, there is absolutely no text output on diagrom.

A few other details that should probably be included:
I am testing this in an amiga 2000 eatx board, which is supposed to be equivalent to a rev 6.2 board.
The board is attached to a cpu slot relocator board. I'm an idiot and soldered it directly to the board because i didnt have any machined pin headers on me.
There is also no FPU installed on the board. I've heard that the FPU apparently helps with clock load.

[terriblefire]

At both 25 and 33 mhz, there is absolutely no text output on diagrom.

I'm just checking you're looking on the serial port with 9600 baud. If you get no output at all the CPU isnt even able to read from ROM.

A few other details that should probably be included:
I am testing this in an amiga 2000 eatx board, which is supposed to be equivalent to a rev 6.2 board.
The board is attached to a cpu slot relocator board. I'm an idiot and soldered it directly to the board because i didnt have any machined pin headers on me.
There is also no FPU installed on the board. I've heard that the FPU apparently helps with clock load.

FPU will help but this is completely dead. Check clocks, power then control lines. check AS to the CPU wiggles. Check DTACK replies.

[gumorr]

I'm just checking you're looking on the serial port with 9600 baud. If you get no output at all the CPU isnt even able to read from ROM.

Yup, i got putty configured correctly on my pc, and I've tested it with just the regular 68000, so serial isn't the issue at all.

FPU will help but this is completely dead. Check clocks, power then control lines. check AS to the CPU wiggles. Check DTACK replies.

I've done some probing with a logic probe, and here's what I've found:
The most glaring thing I've noticed is that it seems like there's no activity on the clock line. The 7 mhz one is good, though.
RESET and HALT don't seem to be getting asserted when the computer freezes.
AS has some activity, along with DTACK, whenever I reset the computer. Usually they both end up freezing high, although sometimes they end up still having activity on them when the computer freezes. They both seem to do the same thing, so I'm assuming DTACK is replying correctly.

[terriblefire]

Have you installed the clock select jumper? JP1? If you have check there is some clock there.

[gumorr]

Have you installed the clock select jumper? JP1? If you have check there is some clock there.

I have the jumper set to the leftmost position, it looks like the clock is just stuck low.

[terriblefire]

If you dont get any clock at the jumper then your crystal is dead. Or your jumper is bad.

[GadgetUK164]

Worth watching my TF530 video - I had a stuck low clock, it was the solder mask not in the right place (PCB manufacturer issue) where it meant pins around the 68K socket (and the crystal) were joining to ground. Check your crystal output is not grounded, and then check all the pins on the 68K socket to make sure only the grounds are connected to ground.

[gumorr]

Well, after a bit of looking around I realized something. I put a capacitor instead of a resistor for the clock load resistor. :headbang: