mdk1@cblpf.ATT.COM (Michael King) (02/01/88)
I have been having problems with my Radio Shack RS-232 Program Pak. When I
try to connect to the Amdahl here at AT&T, I get the carrier signal, then
when I switch the modem from talk to data, I get garbage on my screen. I'm
using the pak with a Coco3, an upgraded Multipak Interface, and Datapack
III+ communications software. I've checked the baud (1200), word length
(8), stop bit (1), and parity (even) and everything is set correctly. The
tin lands on the pak look like they've oxidized a little. I used an eraser to
try to get the oxidation off and it works sometimes. Other times, it has no
affect. Usually, after rubbing and rubbing, I can get the pak to work,
but if I shut it down for even a few seconds, I get the same problem. Is
anyone else having this problem? What have you done to correct it?
Any help is appreciated.
Mike King
..!ihnp4!cbosgd!cblpf!mdk1
--
~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
UUCP:..!ihnp4!cbosgd!cblpf!mdk1 |"What's the point in being grown-up |- The
| if you can't be childish at times?"| Doctor
~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~pete@wlbr.EATON.COM (Pete Lyall) (02/02/88)
In article <1017@cblpf.ATT.COM> mdk1@cblpf.ATT.COM (41450-Michael King) writes: > >I have been having problems with my Radio Shack RS-232 Program Pak..... Mike - The RS-232 pak under LII has problems with dropped interrupts. This is primarily because of the GIME, and the way interrupts are managed (edge vice level sensing is employed). Many/most folks have gotten around this by running a PHYSICAL IRQ jumper out of the pak and into the CPU. You find the little blue jumper wire that currently connects to pin 8 on the RS-232 pak.. it connects to a point about 1/4" to the left. Remove this jumper, and connect a wire to what was the left connection point. Run this out the side of the pak, and connect a plug to it (I use a subminiature phone plug). On the coco end, connect a line to the CPU's IRQ pin (pin 3). The easiest place to do this is on the pull up resistor for the IRQ line... I believe it's R2, but I'm not near a coco or schematic at the moment. Best bet is to check with an ohm meter from PIN 3 of the 6809 to that bunch of resistors in the upper right hand corner of the board (from the front). Connect this to a suitable jack and mount it on the case. Voila - instant REAL IRQ buss. (I also use this same IRQ bus for other paks, such as my Hard Disk HCA, and in the past the PBJ pcpak/rtc). This will make the interrupt operation MUCH better re: the ACIAPAK devices (I use T2 and T3). Another thing is that the UART used in the RS-232 pak - a 6551 - has some design flaws... it'll stop a character IN MID BIT if the Clear to send line is disabled. Best bet is to jumper pins 4 to 5 (RTS to CTS) on the RS-232 pak connector. Also, unless you NEED carrier sensing, jumper 6,8, and 20 also (DSR/DCD/DTR).. The 6551 also loses transmit capability if DCD is not present. If none of the above helps, try a different pak. Also, have you tried a different terminal package? -- Pete Lyall (OS9 Users Group VP)| DELPHI: OS9UGVP | Eaton Corp.(818)-706-5693 Compuserve: 76703,4230 (OS9 Sysop) OS9 (home): (805)-985-0632 (24hr./1200 baud) Internet: pete@wlbr.eaton.com UUCP: {ihnp4,scgvax,jplgodo,voder}!wlbr!pete
koonce@ronzoni (tim koonce) (02/02/88)
In article <1017@cblpf.ATT.COM> mdk1@cblpf.ATT.COM (41450-Michael King) writes: > >I have been having problems with my Radio Shack RS-232 Program Pak. > I don't know whether or not Datapack III+ operates in the 2mhz mode, but I suspect it does, and if so, that may be the culprit. I've heard a number of complaints about the RS-232 pak which were eventually traced to the 6551. Many of the older paks had 6551's which won't work with the 2 mhz clock rate. If yours is one of these, the simplest fix is to replace the whole pak (if you can find one!), or at least the 6551. I'm not sure if that's your problem, but I have heard about this happening. If it works fine under programs such as Mikeyterm, or the built-in comm program, then that is probably the problem. Try using the program built-in to the pak, (or any other comm program which doesn't know about the 2 mhz clock. Greg-e-Term, for example, may use the fast clock on the 3) just to see if everything works. If so, then it may well be the clock speed that's hurting you. - Tim koonce@bosco.berkeley.edu "Graduate assistant: it's not just a job, it's an indenture." +-----------------------------------------------------------------------------+ |Bitnet: koonce%bosco.Berkeley.edu@jade.berkeley.bitnet | |ARPA: koonce@bosco.berkeley.edu Delphi: TIMKOONCE CIS:72276,1135 | +-----------------------------------------------------------------------------+