OSI 600D Superboard Build

[rubber_jonnie]

Well, if the clock isn't getting up to TTL transition voltages then you'll not get anything out of the other side of the 74LS chip.

As it goes, I am getting the correct frequencies at the LS00 (Actually it's an S00, but should be fine), though voltages are off but I need to trace it across the board as well.

My schematic is sorted now:

PXL_20240119_134848242.jpg

Retro computers for scale :)

[stephen_usher]

I've seen larger ones. :-)

So, I'm guessing that the clock pulse doesn't get out of the 00?

[rubber_jonnie]

I've seen larger ones. :-)

So, I'm guessing that the clock pulse doesn't get out of the 00?

Hmmmmm. I'm not sure my schematic matches 100%

So I did a bit of a quick look and there is a 74LS92, and that has a direct connection on pin 11 to pin 37 of the CPU, so PHI0 or clock in. That doesn't quite match the schematic, but OK.

This IC has an input from the clock circuit and it does match the 4v P2P expected, and looks OK compared to the image in the service manual, so there are some differences between what I have and the schematic.

However, the PHI0 output on pin 11 is under a volt, and though the IC tests fine, I'm wondering if it is marginal, as it's getting a good clock in, but not sending a good one out.

[PhilC]

@rubber_jonnie pull the cpu and see if the clock pulses improve, could also be a dodgy cpu.

[rubber_jonnie]

@rubber_jonnie pull the cpu and see if the clock pulses improve, could also be a dodgy cpu.

Yep I already tried that and it makes no difference, plus I swapped out the CPU twice. What I am going to do is double check that pin 11 on the LS92 isn't being pulled down for some reason or other. That will also by extension check pin 37 of the CPU.

That'll be tomorrow though.

[rubber_jonnie]

I posted about my problems on the OSIWeb forum and got a super helpful reply.

It turns out that having a good signal of 4v P2P on the input of U29, the 74LS92, won't help me because it's supposed to be a 74LS93!!!!

I ordered what was on the BoM, which shows a 74LS92, but on the schematic in the assembly instructions, which looks like somebody drew on wet blotting paper with a felt tip pen, it says '93'.

Also, the service manual shows a '7492' at U29. I suppose if you were buying a kit BITD, there wouldn't be a 74LS92 included, so you wouldn't make the mistake, but it seems the person who replied over on OSIWeb ran into the same problem and also ordered a 74LS92, so it's easily done.

Given the CPU feeds a clock signal elsewhere via PHI2, it's not really surprising nothing works with the wrong IC in place!

Parts are now ordered, and I'll report back when they get here. Also looks like the rest of the RAM will get here on Monday too.

Fingers crossed everyone!

[PhilC]

Why is it when building these things or fixing something you always end up needing that one logic chip you don’t have in stock?

[rubber_jonnie]

Why is it when building these things or fixing something you always end up needing that one logic chip you don’t have in stock?

This is the way...

[stephen_usher]

What a pain, and the 74LS92/93 truth tables look very similar too.

Is it acting as a clock divider?

[rubber_jonnie]

What a pain, and the 74LS92/93 truth tables look very similar too.

Is it acting as a clock divider?

They do indeed. AFAIK it is acting as a divider, taking the 4Mhz down to 1Mhz for the CPU and a few other things besides.