RML380Z – One step forward…

In my previous post, things were looking up. I had replaced a couple of faulty chips, re-seated all of the chips on the CPU and Video cards and swapped out the RAM from the CPU card with the expansion cards and now my 380Z was getting to the COS prompt.

All I needed was a key press. Specifically, I needed a ‘B’. Sadly, I don’t have a keyboard.

Screenshot from 2015-10-25 20:08:23Looking at one of the technical documents shows that the keyboard interface is a simple 7 bit parallel interface with a strobe. How hard can it be?

Arduino to the rescue?

151025-IMG_20151025_182536405_HDRI had a supply of Arduino pro mini compatible boards and so I set to building an interface and a sketch to drive the circuit. The idea is to catch characters from the ‘duino’s serial port and squirt them out of the parallel port the way the 380Z likes them.

Not so Fast.

I plugged in the board and nothing happened. I restarted the machine and nothing happened again. Just a blank screen, no COS prompt. Doh!!.

I went over the circuit for the interface again and a again and couldn’t find anything wrong with it. It does appear to have toasted the 380Z though.

A clue?

By coincidence I have just been to the Centre for Computing History in Cambridge and they have a 380Z on a table. I quick word with Jason who runs the place and I have a look inside. Their machine has a different keyboard cable to mine. Mine has an additional connector in-line with the cable. Theirs doesn’t. Sadly, I didn’t take a picture of theirs but it definitely only had one connector at each end.

151025-IMG_20151025_182559575You can see from the photo here that mine has the additional connector. After that connector, wires go off in the direction of the PSU. I’m wondering if mine has been modified for some reason and that has made it incompatible with my relatively simple interface.

I’ll have to trace where all of the wire go and see if that’s significant. I could also do with confirmation that the keyboard ribbon should be straight through.

 

RML380Z – Signs of life.

151014-IMG_20151014_213240378OK. Getting excited now πŸ™‚

I swapped the 4116 DRAMS from the CPU board with those on the RAM expansion card. And now… Ta Daa. I haven’t’ tried the RAM expansion card yet and so I don’t know if the chips are faulty or just dirty and re-seating them has done the trick but so far so bloody brilliant. I’ve waited 35 years for this.

πŸ™‚

RML380Z – Arrival of the new chips.

A small envelope arrived in the post Today. A tube of 74LS00 and 74LS393 to replace the ones identified as broken by my chip tester. The chip tester says that the new 393s are also faulty so I suspect the tester is a little flaky. Anyway, on with the show.

151014-IMG_20151014_180418795There has been some improvement but not a fix. I also reseated the chips that were not tested (ROMs, DRAMs etc.). I moved the 4116s around too to try and flush out the problem.

Back to square one?

It looks a bit like it. I think this might be a clue.

151014-IMG_20151014_181530814_HDR

 

380Z Repair – Send for supplies.

151009-IMG_20151009_162919625_HDRToday I received an IC tester through the post. Yay.

this is an important bit of kit for me. All I have to do is remove a chip from its socket, dorp it in the tester and press ENTER. This little gadget identifies and tests the chip in a fraction of a second. Crikey.

I’ve worked through the VDU and CPU boards and found one faulty 74LS393 counter on the VDO board and a 74LS00 NAND gate on the CPU board. Hopefully these are the source of my problems. If not, I’ll keep looking.

Supplies are on order. I’ll keep you posted.

 

380Z Repair – Another try.

After a break of a couple of years I’ve decided its high time I had another go at repairing my RML-380Z. You can read about this machine and my previous attempts here.

A quick recap… 53e525645f5aa2c829d66d1d878e69f6

When I power the machine on I get a a fixed, pseudo random image on the screen but no other signs of life. I’ve re-seated chips in their sockets and swapped the RAM chips around the board but that’s about it.

It has been suggested that I swap matching logic chip to see if that gives any clues…