Note: This web page was automatically created from a PalmOS "pedit32" memo.

RS-232 notes

RS232 notes

Alex tells me that 50' cables should be no problem, and that UCI used to have rs-232 cables across campus: EG from ICS to the Administrative building
Looks like an interesting "breakout box" based on *ix and two cables:
most basic RS-232 cable is 3 wires The 3 minimum lines are: 1) ground 2) transmit and 3) receive ground must be on the correct pin straight through is "modem" flipped is "null-modem" with more than 3: rts/cts are ready/clear to send, the original form of hardware handshake: am i ok to send? Yes, go ahead ...or... dtr/dsr data terminal/set ready, normally on as soon as powered on, comprises another form of hardware flow control, and a less standard one. Some people felt that the DTR/DSR lines were wasted, so repurposed them for flow control. handshake means "flow control" serial module sm16 is the cyclades normally pins 4 and 5 on an rj whatever, 8 pin serial connector will be ground (the middle two), so you're grounded even if wired up in reverse. If you touch a voltmeter to pins 4 and 5, the value should be near 0. 2.x is high enough to indicate a problem, maybe lower too you can use rj* serial for ethernet, but only for very short distances. You can use ethernet cables for serial, but: 1) only gigabit-capable ethernet cable works (because GigE requires more lines?) 2) ethernet cables are harder to work with (harder to add connectors?) 3) ethernet cables are more expensive Alex has two kinds of breakout boxes: one for 25 pin connectors (with led's?) and one for rjmumble, without led's. The latter makes it easy to test using a voltmeter. The lines with higher values (when connected to ground?) are likely data lines, or does this just mean an active line? I thought everything might be 0, 5, -5, 12, or -12, rather than using magnitude to indicate "data line"
db9 almost a std 5 gnd 2 xmt 3 rcv
cyclades str thru
If you google up the pinouts for the equipment at both ends of the connection, as well as whether each end thinks it's "a computer" or "not a computer", then you can probably wire it up with that. You may also need to know what kind of flow control each end is expecting, but this is often configurable. Basically, if one end thinks it's a computer and the other doesn't, then you wire up transfer to transfer, receive to receive, and ground to ground, and maybe some flow control lines (RTS, CTS - or DTR, DSR) if one or both ends won't do software flow control. If both ends think they're computers, or both ends think they're not computers, then you need to cross the transmit and receive lines - IE wire up one machine's transmit to the other machine's receive, and vice versa. Oh, and if you can get away with using a parallel or USB cable or something else instead, you probably want to! It'll probably yield both faster communication as well as a faster hookup in terms of admin time.
From AIX-L: here is the standard pin outs for a RJ-45 Table 3-1 lists the pinouts and signals for the RJ-45 port. Table 3-1 RJ-45 Pinout for networks {Ethernet} Pin Description 1 Receive Data + (RxD+) 2 RxD- 3 Transmit Data + (TxD+) 6 TxD- RS-232C serial connections (primarily for modem data transfer) ** Note: direction of signals are relative to the respective each connector Legend: <-- = incoming signal --> = out-going signal DB-9 female connector end RJ-45 female connector end Pin# Signal code Direction Signal Description from Pin# Signal Code Direction Signal Description 1 CD <-- Carrier Detect 3 DTR --> Data Terminal Ready 2 RxD <-- Receive Data 6 TxD --> Transmit Data 3 TxD --> Transmit Data 5 RxD <-- Receive Data 4 DTR --> Data Terminal Ready 2 CD <-- Carrier Detect 5 GND --- System Ground 4 GND --- System Ground 6 DSR <-- Data Set Ready 1 RI <-- Ring Indicator 7 RTS --> Request to Send 7 CTS <-- Clear to Send 8 CTS <-- Clear to Send 8 RTS --> Request to Send 9 RI <-- Ring Indicator HTH Jim Galloway ... I forgot that there should be a jumper wire from pin# 1 to pin# 6 on the DB-9 end as well
Great URL about RS232:
A newer serial terminal concentrator than conserver:

Back to Dan's palm memos