Bringing up the RC2014/WOPRjr for the first time

[Michelle Lawson]

Yep, I named my 'server' app after the movie. Anyway, first time powered on and nothing on the terminal. I have the CPU and SIO/2 clocks set for 2.4576MHz and the Port A interface set for 38400 baud, N81, and no flow control; which I assume is correct. 

I brought out the scope and I notice that none of the upper four decode signals on the 74HCT138 ever go low. That would imply that the A2 input never goes high and is stuck low. When I first tried to scope that pin, while is did see pulses, then were never high enough to be a logic 1. All of the other logic signals on everything I scoped out were perfect square waves. A reseat of the chip didn't help, so my next step will be to reflow the joints, try a different chip (an 'LS' should work for testing), and maybe a bodge wire to see if there is a trace issue.

But, if my terminal settings or wrong, or if anyone has any ideas so I'm not barking up the wrong tree, please shout them out. Thanks

[Ed Silky]

When I see a signal that should be going to a logic high, but is only slightly going up, I look for a short to another logic line or to ground by way of a resistor. I don't have a schematic and board layout to help more than that.

-Ed

--
You received this message because you are subscribed to the Google Groups "RC2014-Z80" group.
To unsubscribe from this group and stop receiving emails from it, send an email to rc2014-z80+...@googlegroups.com.
To view this discussion, visit https://groups.google.com/d/msgid/rc2014-z80/9f78279b-3374-4ec6-9fc7-3d6aeb6c18c9n%40googlegroups.com.

[Michelle Lawson]

Okay, I think I may have made 'some' progress. The solder joints associated with address bit 7 on the SIO/2 card looked suspicious, so I reflowed them. And, while the RS232 serial interface is so simplistic, it always finds a way not to be; along with having zero info on the dongles, short of the silkscreen, I did swap around the receive, transmit, and moved the RTS line to the CTS line on the dongle. Now I at least get gibberish on the screen initially, so I suspect I have a protocol error left to tackle. I'd like to get some confirmation as to what handshaking is required, as opposed to my shotgunning approach, besides the 38400, N, 8, & 1 I have set. Thanks

[Michelle Lawson]

Well, I bodged in a little 8 pin strip header that I hot glued to the top of the SIO-2 card so I could easily scope the RX, TX, CE, M1, RD, IORQ, C/D, and B/A signals, right on the SIO-2 chip. They all had good looks TTL level signals, and while I have no method of looking at more that two at a time, they were at least doing something. Except for the TX and IORQ signal. I expected TX would show nothing since nothing shows up on the screen. The RX line shows activity consistent with a key being depressed. Oddly though, the IORQ line is always high.

I scoped that back to the CPU and it never changes, even at the CPU; a good solid high all the time. I would expect that there should be some initial activity of the IORQ line when power is applied and the software starts to initialize the chip. So, I trying to figure out what I can kludge together next to keep troubleshooting this thing......I sure could use some ideas right about now...... Thanks

[Ed Silky]

As I mentioned, I don't have the same setup as you (as I use a Z80 SBC of my own design), but I can give you some general comments/suggestions.

The IORQ- line won't do anything unless interrupts are enabled on the SIO and something occurs that causes the SIO to generate an interrupt. Just starting things up and initializing things will not (necessarily) cause the IORQ- line to go active (low).

On the signals that you indicate that you brought out to be able to 'scope', you don't mention WR-. I would scope the SIO CE- and WR- together, sync on CE-.

With that ready (if you have a storage scope, do a single trace) - reset/restart the system. You need to see CE- active (low) and WR- active (low) at the same time. There should be about a dozen occurences of that. That would be the start-up initializing the SIO. If you don't see that, then you'll need to track things closer to the bus. For example, does the address decode generate the CE- and does the WR- signal come through from the bus.

If you do see those writes, then it could be that the initialization doesn't enable interrupts (it uses polling), or it is only enabling them for the port you aren't sending data to. It could also be that the data coming in on the RX pin isn't being recognized as valid serial data. If the SIO isn't configured to interrupt on errors, it won't generate an interrupt unless valid data is received.

If the SIO does get the CE- and WR- signals (indicating that it is probably getting initialized), I would think that you would see something on one of the TX lines (TXA or TXB), as most boot-ups will try to print something. If you can see signals on either of those, then (with the scope) you can try to figure out what communication parameters it is using (baud, data bits, etc.).

I'm sorry that I can't be more specific. Hopefully the folks that actually have a setup like yours will come back from their holiday soon and help you out.

Let me know if that helps at all.

-Ed

To view this discussion, visit https://groups.google.com/d/msgid/rc2014-z80/de743fb6-dccc-40b0-afe9-249f551600fan%40googlegroups.com.