Another Nascom

[acollins22]

Hello Folks,

Having seen the progress made over the last few weeks with a Nascom-1 on the other thread I thought I dust mine off and try, with a little help from my friends, to get it going.

I started working on this as part of retrochallenge2017/10 and you can read about it on my website here.

To cut a long story short the machine powers up but displays a screen of garbage.



I have the following voltages from the PSU...
+11.94
+4.94
-5.07
-12.06
So I think they are all OK. (I did recently change the 5V regulator because it was putting out only 4.79ish but it hasn't made any difference to the screen output)

Using my 'scope I can see a 2mhz clock on pin 6 of the CPU. The MREQ and RD pins are moving about as I would expect so I think that the processor is running.

I have also tried my own ROMs that hit the IO and I saw activity on the IOREQ and a ROM that just accesses RAM and saw no IOREQ activity suggesting the ROM is being accessed OK.

Please can someone suggest a plan of attack as I'd really like to get this 1K marvel back up and running again.


Thanks very much,

Andy.

 

[daver2]

The more the merrier Andy!

So, it looks like you have a well-formed and synchronised 48 by 16 VDU display - which is good.

What monitor ROM have you got and how are the N1 links configured before we start?

When you say "you have tried your own ROMs" does this mean you can program 2708's to try in the ROM socket at address 0000? If so, you could write a simple Z80 program (which I would be quite happy to do for you) to directly write characters to the screen so we can see if the problem is with the Z80 processor side or with accessing the video RAM.

Dave

 

[acollins22]

Hello Dave,

Thanks for the reply.

I have Nassys3 in at the moment.
I'm struggling to read the pin designations on the jumper but there positions are as follows...
LK3 = Top left to bottom left.
LK6 = Top right to bottom right.
LK4 = Bottom right to top left.


I can't program 2708s but I can blow 2716s and have a couple of adapters that let me put them into 2708 sockets and just use the bottom half of the EPROM.

I did blow a couple of ROMs to try and see if the screen RAM was OK. One used 00h the other 2Ah ('*'). I've just found one, I _think_ the '*' and tried it.

Each time I power on I always get a screen similar to the one I showed above. With repeated presses of the reset button I sometimes get the screen shown below. I've tried shorting the pins at the back of the reset switch and that isn't a lot better. It feels as if the machine isn't coming up "clean".

The character is the character for 05h.


Cheers,

Andy.

 

[falter]

I believe on mine it was replacing one of the 81ls97s that got me past the frozen garbage screen.

 

[daver2]

LK3 = Not interested in.
LK6 = 2 MHz CPU operation.
LK4 = Not interested in.

I assume this is a NASCOM 1 on its own (i.e. no buffer board or expansion) and, therefore, LK1 and LK3 are both set to "INT"?

It is also evident from what you are saying is that with a test EPROM in place - and multiple 'goes' at hitting reset - that the video memory can be initialised (thus indicating that the Z80 CPU can write to the video RAM under certain circumstances.

This leads me to suspect that operation is somewhat problematic and intermittent as opposed to a hard fault.

In circumstances like these I would consider the following:

1. Do you have a chip or a chip socket with dirty legs or making poor contact?
2. Do you have a 'slow' device (e.g. an EPROM that is too slow compared to the timing requirements of the CPU at a specific clock speed).
3. Do you have a 'noisy' power supply?
4. Is the reset circuitry somewhat 'slow' to recover?
5. Will the machine run reliably at a slower clock speed?
6. Do you have something that is temperature dependent?

Assuming you have just turned the machine on - and the machine is not working properly then - I would rule out item 6. Machines tend to work better cold than hot. However, you can get a hair dryer and a can of freezer spray and try both on the N1 board in areas to see if they have any effect.

The easiest thing to do first is to reduce the clock speed to 1 MHz (by adjusting LK6) and see if that has any effect.

Can you report the exact part numbers on the NAS-SYS 3 EPROMS and your test EPROM(S) so that I can look them up in a data sheet. I assume your test EPROM doesn't use any RAM (other than the VIDEO RAM)?

I can't see that (4) would be a problem - as there are no capacitors or active devices in the N1 RESET circuitry. Just a pull-up resistor, a swicth and a carbon based lifeform (e.g. you).

These are the worst faults to find - but I think you know that already...

Dave

 

[acollins22]

Hello Folks,

Thanks for the suggestions. I'm working through them at the moment but current progress is...

The power supply is very clean on all four rails.
Running the machine at 1Mhz makes no difference.

It's interesting to me that the reset switch connects every time but the machine doesn't like to reset. I'm sure that's a clue but to what I'm not sure

I can't read the part number of the NasSys as the ink has blurred over the years. The Fill program is of my own devising and doesn't use any system RAM. It runs from ROM and writes to the screen RAM.


More to follow


Cheers,

Andy.