Hello fellow Cromemco fans...

[David Roberts]

Hi fellow Cromemco fans.

I have just joined the group - so I thought I would say a brief hello.

I live in the UK (Worcester to be accurate) and own 'bits' of a Cromemco Z-2D.

I have 3 sets of Cromemco S-100 cards that I obtained a few years ago from a guy who had thrown the 3 chassis away but kept the cards (weep). Mainly 16FDC, 64KZ and ZPUs. I have a number of TUART cards, PRI cards and a glut of D+7A cards. The odd 64FDC thrown in there for good measure...

The Z-2D chassis, power supply and Blitz Bus was obtained from Marcus a good few years ago also. I had a couple of faults with the bridge rectifiers on the power supply (all fixed now) but was having problems successfully bringing the system up. I have recently been helping someone over on VCFED with their system - so that has rekindled my interest in the Cromemco, and I have started to work on it again.

I see a few names on here I recognise from VCFED!

I had a basic system of a 16FDC, 64KZ and ZPU. On power-up I see a brief flash of activity on the 16FDC transmit line and then nothing. I have replaced the 3 boards with another set - and observed exactly the same issue. I have also tried with an RPi and minicom connected via a USB/RS232 converter. Nothing.

I have an RS232 line monitor on the 16FDC console, and the transmit line is not actually active when powered up (apart from the initial activity). It is almost as though the 16FDC is sending out a continuous <BREAK>.

As there is no hard RESET on the TMS5501 UART, I am beginning to suspect that the ZPU is not running properly to initialise things. However, that could point to any of the 3 cards I have in the system.

I have written some test firmware to replace RDOS on the FDC that doesn't require a RAM card to operate - but (unfortunately) I don't have the immediate means to burn a 2708 EPROM available. I will see if a friend of mine has an EPROM programmer that I can use.

The DC power rails look OK - so I am beginning to suspect the Blitz Bus backplane. I shall remove it and see if there is any damage to it anywhere obvious. One of the S-100 connectors did have a bit of corrosion on it - but I did try to rescue it. Perhaps I only think I did a decent job and some of the S-100 lines are cross-coupled in someway?

If that doesn't give any joy - I might have to break out the oscilloscope and logic analyser...

Dave

[randy482]

My favorite answer to bringing up a Z80 computer:

builderpages:gary_kaufman:z80ice [RetroBrew Computers Wiki]

Randy

[Udo Munk]

RDOS is trying to find out the baud rate of the terminal, you need to enter a few CR on the terminal.

[David Roberts]

Udo,

Aware of the autobaud feature and have tried that.

I was using a RPi with a USB/RS232 converter and minicom software.

I have tested the serial configuration on my RPi using an RS232 breakout box and looped back TX and RX. 

When I connect my breakout box to the RPi I get the TX light illuminating. This is the normal observable result for a quiescent RS232 serial line.

When I connect the breakout box to the 16FDC, I see a brief flash on the transmit line when I power on, and then the light goes out. This indicates to me that the transmit line is sending a BREAK and is not at the normal quiescent level.

However, the 16FDC may not obey the ‘standard’ rules for the RS232 transmit line. After I have checked the Blitz Bus out (it looks fine - I have just cleaned it up with a tiny brush) I will recheck with CRs again to see if I screwed this process up the first time through. Did I set the correct number of data bits, parity and stop bits I ask myself now? If I didn’t, that would be a schoolboy error!

Cheers,

Dave

[randy482]

Cromemco's need to have pins 2,3, and 7 ONLY connected. 

They do not support other stand pins plus the have current loop on some pins.

Randy

[Mike Arnold]

Agree with Randy on the simplicity of the serial interface. When RDOS starts with the auto baud setting it sends out a break on pin 3 to which the 3102 terminal responds with CRs on pin 2 to allow RDOS to set the baud rate. If it is sending out the break that is encouraging news as it is half working. I suggest that you confirm this by observing the Tx line after reset rather than just power up. On the subject of character settings, I always use 8 data bits, 1 stop bit, no parity. The 5501 is quite forgiving. Another test is to set the 16FDC switch 5 to ON and it should start at 300 baud with a character stream on pin 3.

As you know, normal RS232 is to have -12V (or thereabouts) on the Tx pin when idle which then toggles to +12V when actually transmitting.

However, I can check out your cards when you visit Manchester and give you some diagnostic software. Most of Cromemco's board faults relate to chips not contacting properly in the IC sockets.

Mike

[David Roberts]

Hi,

So, I have got a little further.

Good news regarding my RS232 serial connection - everything was OK in the end.

I changed the 16FDC to a fixed serial speed of 300 bps and I started to get signs of activity. However, I didn't observe the customary "Cromemco RDOS ..." message; but I did get the ';' prompt. After a little playing around (but I didn't have my RDOS command crib sheet to hand) I switched the machine off and had something else to do. On my return, nothing on the transmit line again...

I 'tweaked' the 16FDC a bit in the slot (I had the lower card guides in - but the upper ones not in preparation for poking it with an oscilloscope) and everything burst into life!!!

I was able to enter a few RDOS commands and things were partially working. The machine would crash/hang after a short while - but I could recover the situation with a hard reset. Power on / power off - all fine. Shut the machine down for the night and then the following morning ... nothing again.

Sometimes the transmit light would be OFF (the equivalent of a <BREAK> condition) and at other times it would be ON (normal RS232 inactive state) but with no transmit data on it (the initial sign-on message and prompt).

So, it would seem nothing fatal - but most likely poor connections. So, I will go with Mike's suggestion of the IC socket connections to start with. I shall look to remove the ICs from their sockets, squirt a bit of contact cleaner in the socket and check the IC pins for tarnishing. I will see where that gets me.

On another note, the Blitz Bus has a partially corroded S-100 connector in slot 1. I have measured the resistance between adjacent signal lines and found it to be 2 MOhms. The further I move away from the corrosion, the higher the resistance gets between adjacent signal lines - until the resistance is virtually infinite where there is no corrosion. Thoughts on whether this would be a problem or not?

I would hazard a guess as to no. Unused inputs on the S-100 bus should have pull-ups in the order of 1-10 kOhms to +5V so vastly swamp the 2 MOhms. Driven outputs will pretty much drive the signal line irrespective of the 2 MOhms.

That may still be an area to look at - either remove the S-100 connector from Slot 1, or cut the signal traces on the backplane to it.

Thoughts?

Dave

--
You received this message because you are subscribed to the Google Groups "Cromemco" group.
To unsubscribe from this group and stop receiving emails from it, send an email to cromemco+u...@googlegroups.com.
To view this discussion on the web visit https://groups.google.com/d/msgid/cromemco/86c6b402-3810-400f-98cd-84395cd9756cn%40googlegroups.com.

[Mike Arnold]

When I used to maintain Cromemcos we would take out the chips and gently clean them with a shaped ink rubber. Extremely laborious but effective. 

[MikeS]

Hi Dave,

 

fancy meeting you here after all the fun we've had on VCF, especially the recent AIM65 marathon ;-)

 

Hope you made some more progress over the holidays; regarding the corroded connector I'd at least remove it, clean up anything you might find underneath it, and maybe even replace it just for completeness.

 

Maybe we can chat about your box a bit more next year ;-)

 

All the best of the season to you and everyone else in the group; let's hope 2022 is a damn sight better than the last two years!

 

mike

To view this discussion on the web visit https://groups.google.com/d/msgid/cromemco/ced2dca7-2044-4154-9bde-dd98a587dd58n%40googlegroups.com.