CPU280 - No console [message #5542] |
Fri, 16 November 2018 05:13 |
trick-1
Messages: 164 Registered: September 2016
|
Senior Member |
|
|
Hi Folks
So I have built up the CPU280 board but have been unable to get anything out of the serial port.
When I power on the board the
- D1 goes out quickly (reset)
- D2 goes out quickly (clock frequency)
- D3 takes some time but eventually goes out (memory check)
So it looks like the CPU it talking to the ROM and also able to check the RAM.
the only other thing of note is that the CPU gets a little warm on the corner near the LED's but that's about it.
Have checked the following
- used known working null modem cable (and tried without)
- confirmed the laptop still talks to another serial device
- solder joints
- looked for solder bridges
- checked for bent pins on IC's
- Reset the DS1287 by connecting the jumper J10 with the power off.
- confirmed all parts are right location
- confirmed all parts are right value
- pulled and reseated the CPU280 - all looked OK
I am basically at the point where I am considering replacing the following
- LT1134ACN
- Z280
One other thing I might try tomorrow is obtaining a USB to Serial module and with the LT1134ACN pulled and wire it directly into pins 17/18/19/20 (TX/RX/RTS/CTS).
Anyway I thought I would see if anyone had any other suggested areas to take a look at.
Thanks in advance
Richard
|
|
|
|
|
Re: CPU280 - No console [message #5548 is a reply to message #5546] |
Sat, 17 November 2018 13:00 |
|
lowen
Messages: 226 Registered: August 2016 Location: Western NC USA
|
Senior Member |
|
|
The pin descriptions are correct in the documentation, they just don't map 1:1 to a standard 9-pin plug. Here's the pin mapping I use, with the left side being the CPU280 pin header and the right hand side a male DE9 plug:
To use as a DTE (PC 9 pin male standard pinout):
CPU280 Pin -> DE9
1 -> NC
2 -> NC
3 -> 3
4 -> 5
5 -> 2
6 -> NC
7 -> 7
8 -> NC
9 -> 8
And then I use a null modem along with that to properly loop back all the handshaking lines. I would post a photo of the cable, but that might not help you much. It is definitely not a 1 to 1 pin correspondence, though!
--
Bughlt: Sckmud
Shut her down Scotty, she's sucking mud again!
[Updated on: Sat, 17 November 2018 13:01] Report message to a moderator
|
|
|
|
|
|
|
Re: CPU280 - No console [message #5553 is a reply to message #5552] |
Sun, 18 November 2018 09:46 |
tingo
Messages: 111 Registered: February 2017 Location: Oslo, Norway
|
Senior Member |
|
|
Also note that a usb to TTL adapter might not work if this is a "real" serial port (12V levels, MAX232 compatible). Benn there, done that.
Torfinn
|
|
|
Re: CPU280 - No console [message #5554 is a reply to message #5553] |
Sun, 18 November 2018 10:30 |
|
lowen
Messages: 226 Registered: August 2016 Location: Western NC USA
|
Senior Member |
|
|
Also, while I read that you had reset the DS1287, it would be worth trying a new DS12887A. The regular DS1287 or DS12887 without the A suffix isn't jumper resetable.
The fact that your LEDS are cycling correctly leads me to believe that the remainder of the board is working ok.
--
Bughlt: Sckmud
Shut her down Scotty, she's sucking mud again!
[Updated on: Sun, 18 November 2018 10:32] Report message to a moderator
|
|
|
|
|
|
|
|
Re: CPU280 - No console [message #5814 is a reply to message #5813] |
Wed, 02 January 2019 20:10 |
|
lowen
Messages: 226 Registered: August 2016 Location: Western NC USA
|
Senior Member |
|
|
Hmm, 19.6608 MHz.... ah, I see in the hardware manual that being listed as an alternative frequency. If you have a 12MHz rated Z280 (which all of the ones I sent out are) then try a 24.576 MHZ crystal instead. Your trace looks like good serial data. Now, where are you checking the clock, right at the crystal or the Z280 clock output? The Z280 has an internal divide by two on the clock.
So, if you system is scaled on its bit rate, I would try a bit rate of either 7680, 5378, or 4303 bits per second ( those are the ratios of 9600bps to the possible clock ratios of 24.576:19.6608, 19.6608:11.016, and 24.576:11.016). The easiest to try is to put in a 24.576MHz crystal and check for 12.288MHz at the Z280 clock out on pin 47 of the Z280.
Do check to make sure the crystal can is not touching the board or has an insulator between the bottom of the can and the board, as if I recall correctly there is a via under the can.
The current CPU280 system software may not compensate for odd clock speeds, but I haven't dug deeply enough to be sure of that, since the DS1287 RTC chip can be used to figure out the clock speed.
EDIT: If you're checking the frequency right at the crystal it's possible the probe is adding enough parallel capacitance to change the oscillator frequency.
--
Bughlt: Sckmud
Shut her down Scotty, she's sucking mud again!
[Updated on: Wed, 02 January 2019 20:14] Report message to a moderator
|
|
|
|
|
|
|
|
Re: CPU280 - No console [message #8163 is a reply to message #8162] |
Thu, 31 December 2020 16:57 |
cmacarthur
Messages: 26 Registered: April 2018 Location: Alberta / Montana
|
Junior Member |
|
|
Hello Richard
THX for the replay
"Make sure you are using the right speed GAL's and double check that your GAL's are programmed ;-)"
I am using PALCE 16V8-5 (5-ns propagation delay), is that to fast ?
I could change to -7 (7.5-ns propagation delay) as I have a few hundred of each...
See attached
I checked the programming BUT will try the slower ones next
I am using ROM 1.20.2, I will also try changing to 1.20.1
I hope to get back to the CUP280 next week...
THX
CM
|
|
|