Two devices listening to same serial port 1

[JoeAtWork]

I'm trying to support a customer but I'm a programmer and this is a hardware issue. It's probably very self-evident for some of you.

He has a lab instrument that is sending data via RS232. He has two physical devices that need to pick up the data at the same time:
1. A serial printer that will print it out
2. A computer running software that is listening to the port and parsing the data

He has connected the instrument to a BlackBox Modem Splitter (

http://www.blackbox.com/Store/Detail.aspx/TL074A-R4)

to the "Master Port", the printer is connected to the BlackBox on Port1, and the computer to Port2.

When the software is not running (not listening to the port), the printer picks up the data fine. When the software is running at the same time, the printer picks up some of the data, but misses a lot of it.

I suspect he is using the wrong equipment (i.e. the BlackBox), except for the fact that the printer seems to work while the software is not running. The software always picks up the data fine, it is not affected by the printer running. If it is of any relevancy, the software can be configured to set the protocal to "None" or "ASTM 1381".

Can anyone offer an explanation of what is wrong, and what might be the solution?

Joe Schwarz
Custom Software Developer

http://www.plandev.net

 

[edfair]

My suspicion is that the black box is getting confused by signals coming from the computer and the printer. Would question whether the computer running the program is pushing commands to the lab instrument and grabbing control of the port at the same time as the printer is supposed to be getting data.

If this is strictly a data capture issue there is no need for the black box. The lab instrument is capable of putting out enough voltage/current to run 2 line receivers (chip that receives the signal). In fact line drivers (sending devices) put out enough signal to drive 10 receivers.

If the printer works OK by itself, data capture on the computer would be by using just the signal common and rx lines. Any buffer full problem on the printer would be handled by the printer and lab instrument.

The alternate would be to use the computer to control the lab equipment and set the printer to no protocol and only use two wires, rx and common. But the problem there is any response to the control program will get printed.

Can the control program be set up on the lab instrument and then the computer put into passive capture? That would allow both to receive and with no software dumping to the lab instrument and no printer handshaking both should receive correctly.

As I see it. Hope you can understand.

Ed Fair
Give the wrong symptoms, get the wrong solutions.