My OSI 610 build - Expansion for the OSI 600

[rubber_jonnie]

What's interesting for me is that it wasn't immediately apparent from buzzing the cable out.

Pin to equivalent pin from connector to connector worked fine, and checking for shorts on pins next to each other in the same row was fine, however when I checked from pins on the left row to pins on the right, that was when it showed up.

Don't normally have issues like this with IDC connectors, but there you go.

[rubber_jonnie]

OK, so I have 2K of RAM in the board now, but it's not being seen. Very annoying.

Sadly I was working today, and had no time to look at it, and by the time I do, I should have some more RAM to fit, but I'll need to test it in the main board first, one of the perils of getting IC's from Ali Express, but to be fair the last lot I bought all worked just fine.

I did scope the RAM on Friday quickly and noted that address line look OK and so do the data lines (They are all a bit marginal so there may be more to it), but the _CE and _WE lines are not looking good at all, so I need to start looking in the area of the 74LS138 that drives _CE0 and _CE1, U15.

_WE or BR/W as it is labelled on the schematic seems to come via pin 15 of the mainboard to 610 connector, so I need to figure out if that is coming through. I originally thought it was for some reason coming from the 6850 on the 610 board, but that was a misread of old schematics that are split across multiple pages.

If it's not there then that might be a clue as to wether it's an issue with the signal path for the mainboard to the 610, or on the 610 itself.

[rubber_jonnie]

OK, so I'm a bit stumped. This is the decode logic for the 610 RAM:

DECODE.png

U18 is what controls the /CE lines, but at U15, U16 & U17, there is no input at pin 4. This is provided by U18, and pins 12, 13 and 14 are sitting like this:

IMG20250429133934.jpg

These measure the same at U15, U16 & U17 pin 4.

Pins 1-3 of U18 look like this:

IMG20250429134004.jpg

And according to the datasheet the LS138 requires 2v for a High and 0.8v for low, which this seems to show is the case.

The other inputs for U18 are 4,5 & 6. 4 & 5 are tied to GND, and 6 is tied to 5v. All of these buzz out correctly. So the inputs are there, but no outputs meaning no /CE :(

I lifted the 3 output legs to see if that made a difference, and it doesn't, so what the heck is going on, the LS138 at U18 is not outputting as expected. BTW I also tried other LS138's and got the same...

And I have tested the LS138s, both in my TL866 Pro and also in the 600, they seem to run just fine.

[stephen_usher]

You've not got a mislabelled CMOS part have you? They need higher trigger voltages after all and 2V wouldn't cut it.

[rubber_jonnie]

You've not got a mislabelled CMOS part have you? They need higher trigger voltages after all and 2V wouldn't cut it.

Yeah, well, I have been wondering if they are relabelled parts, but I got them from a supplier I trust They are labelled 74LS138. They do test OK in my TL866 and on the main computer board, but that may be because the voltages are higher there.

I will order some more 138's but not until sometime next week when I'm in France, otherwise they run the risk of being left on the doorstep whilst I'm away. If I order them mid week I'll be home by the time they arrive I hope.

The BoM states 74S138 or 74LS138, and the schematic also states LS. Bloody annoying.

[rubber_jonnie]

OK, so I'm going to have to park this whilst I'm away.

It seems clear that U18 isn't providing any output to U15, despite the inputs being good and the IC testing fine.

No /CE0 or /CE1 from U15 means no RAM, and since U16 and U17 require the same signals then /CE is a bust across the whole 610.

I see it as being either re-labelled LS138s that require a higher voltage to work or that the 8T95s are bad as was suggested on another forum. Either way I can't order replacements as they won't arrive before I go on holiday.

What's coming out of the 8T95s looks fine, but I have no way to test and my bad cable could have damaged them.

So I'll order some before I come home and hopefully they'll be here the week after next.

[HigashiJun]

Lot of unexpected troubles...

Hope you'll finish this project soon.

[rubber_jonnie]

Lot of unexpected troubles...

Hope you'll finish this project soon.

Well, once the RAM is sorted I'll have to get the FDD interface working...

A while to go, but yeah, it's be nice to finish it.

[rubber_jonnie]

Well, I have N8T28's, N8T95's and some Fairchild 74S138's on order for when I get back to the UK at the weekend, we'll see how they get on.

I couldn't find LS's by Fairchild, and the BoM says S's are OK, so let's see what happens when they come.

I was advised by another builder Fairchild ICs worked well, hence the choice, and there is a chance that my faulty cable could have damaged at least the N8T95's, so the same builder advised I changed them too. They do have output though, just not sure if it's enough to drive the LS138s I have.

Not sure how I could test the NT28's and NT95's to see how they are doing one way or the other, the TL866 doesn't allow for it.

[rubber_jonnie]

OK then, it's still not working and I decided to have a look at what happens to U18 when I his reset and found that there were actually signals.

Pin 12:

IMG20250524155551.jpg

Pin 13:

IMG20250524155637.jpg

And pin 14:

IMG20250524155713.jpg

So there are actually signals during reset. So I went back to sheet one of the schematics and I see that everything looks fine but BR/W, which comes from U2 pin 13 of U2, with the input being pin 14.

Capturing on both pins at once I see this:

IMG20250524162930.jpg

Yellow is pin 13, blue is pin 14. So I lifted pin 13 (Output) and checked again with the scope:

IMG20250524162730.jpg

Lo and behold the output on pin 13 is now correct.

So it looks like something is interfering with BR/W so I need to dig further to see just what that is. I had my concerns about BR/W before and it seems they've come full circle.

I did note that BR/W on the RAM IC's is the same as pin 13, so I need to trace it across the board and see what I can find. I suspect U18 is doing nothing because BR/W isn't working so it's simply not bothering with the 610 RAM, as it thinks there's none there.