English Amiga Board

English Amiga Board (http://eab.abime.net/index.php)
-   support.Hardware (http://eab.abime.net/forumdisplay.php?f=20)
-   -   A1000 serial issues, and crashes when sending files to SER: (http://eab.abime.net/showthread.php?t=99582)

Turrican_3 09 November 2019 13:54

A1000 serial issues, and crashes when sending files to SER:
 
Hey there, it seems my A1000 woes haven't ended yet...

After replacing the faulty RAM chip (there's another thread for that, no need to talk more about that) it seemed everything was fine, but while trying to debug issues with an ACA500+ it was clear I needed to use the serial port.

Unfortunately, all I got from the serial log was a bunch of garbled characters. Same thing happens disconnecting the accelerator.

Still having issues at the receiver side even after having swapped CIAs. :-\

MAJOR (?) UPDATE:

Quote:

Originally Posted by Turrican_3 (Post 1357955)
Ok since this seemed to go nowhere I decided to connect my (only and cheap!) oscilloscope to the signals coming from my Amiga 500 and 1000.

Unless I'm doing something very wrong with the probes (connected it to TXD and GND, pin 2 and 7 respectively) I see signals going as high as roughly 16V on the A1000 to about 25V on the A500.

Shouldn't it read much, much lower than that?!

I guess this might also explain why the PC adaptor was reading garbage, as I understand it accepts about 5V max.

=======================

THIS PART OF THE ISSUE WASN'T APPARENTLY AN ISSUE AT ALL (WRONG SETTINGS) SO I'M STRIKING IT OUT

I thought, let's make things as simple as possible: physically unplugged the ACA accelerator, and decided to run Workbench 1.3

Opened an Amiga Shell, went to S: and entered "copy #? to SER:"

As soon as the DPAT file copy starts, system hangs.
(this happened on WB3.1 as well for the record with the ACA plugged in, but as I said I wanted to be absolutely sure it wasn't the accelerator that was creating problems - and yes I know it's an unsupported configuration on paper so that was one more reason for me to test the system without installing it)

Please note Amiga Test Kit with the null modem dongle says there's something wrong with the RI signal (pin 22), but keeps running tests as long as I want.


Thoughts?

Toni Wilen 09 November 2019 18:19

Try swapping CIAs. Serial control signals are handled by CIAs. Paula handles RX and TX.

th4t1guy 09 November 2019 18:44

Does the a1000 serial port have the same pinout as other amigas? I thought it was different for the parallel port at least.

Turrican_3 09 November 2019 19:42

@Toni
Yeah I understand swapping CIAs is among the first things to do with serial issues, but do you believe it could lead to system crashes?


@th4t1guy
Pinout should be the same, but gender of the connector is different from other Amiga models (IIRC all of them)

No idea about parallel to be honest though.

Toni Wilen 09 November 2019 19:50

Quote:

Originally Posted by Turrican_3 (Post 1357778)
@Toni
Yeah I understand swapping CIAs is among the first things to do with serial issues, but do you believe it could lead to system crashes?

Anything is possible..

Because CIA swap is so easy, it should be always done before trying anything else :)

Quote:

Does the a1000 serial port have the same pinout as other amigas? I thought it was different for the parallel port at least.
Pinouts are mostly same but some power pins have been moved around. Full 1:1 gender changer is not good idea when using A1000.

Turrican_3 09 November 2019 21:29

Got it Toni, thanks.
I will try swapping CIAs and will report back as soon as possible.

Turrican_3 09 November 2019 23:32

CIAs swapped, Amiga Test Kit keeps signalling RI (pin 22) error but it doesn't crash system - same behaviour as before.

Workbench 1.3 though doesn't seem to crash anymore: I found out what was blocking the copy operation (forgot to set handshake to "none") but that's all, as I said Amiga is simply unable to complete command but remains responsive.

Weird perhaps, but I guess I'm fine with that as long as system is stable (here's assuming RI signal fault has nothing to do with previous crashes)

Turrican_3 10 November 2019 00:36

I see conflicting reports about the presence of the RI signal on the A1000 serial port.

Maybe there's no issue at all about this signal, but I'm still unable to get proper, readable data at the receiver - i.e. PC - end. :confused

Turrican_3 10 November 2019 15:45

Ok, I officially declare myself totally, utterly confused.

Tried the "send S: stuff to SER:" on an A500 KS1.2 + Workbench 1.3 and the Amiga instantly crashed at DPAT (the first file that it tries to send)

This is getting weirder and weirder. :confused

I'm going to try a Workbench 1.2 disk now just to be on the safe side.

Turrican_3 10 November 2019 17:43

Ok since this seemed to go nowhere I decided to connect my (only and cheap!) oscilloscope to the signals coming from my Amiga 500 and 1000.

Unless I'm doing something very wrong with the probes (connected it to TXD and GND, pin 2 and 7 respectively) I see signals going as high as roughly 16V on the A1000 to about 25V on the A500.

Shouldn't it read much, much lower than that?!

I guess this might also explain why the PC adaptor was reading garbage, as I understand it accepts about 5V max.

Wrangler 10 November 2019 19:08

I believe the normal signalling range is between +-15v but equipment must be able to cope with +-25v. Sounds like your PC adaptor isn't properly RS232 compatible.

Turrican_3 11 November 2019 14:50

Definitely!

I really should have turned on the scope before, and most of all looked at RS232 specifications, it would have been clear there was no way I could make the two machines talk to each other.

I think a MAX232 or something like that should do the trick, right? I happen to have a spare ICL232 at home, seems a good fit.


I'm still confused about the crashes though, I'd like to try the same procedure with some other real Amiga I own and/or even with WinUAE.


All times are GMT +2. The time now is 19:33.

Powered by vBulletin® Version 3.8.11
Copyright ©2000 - 2020, vBulletin Solutions Inc.

Page generated in 0.04169 seconds with 11 queries