Share RS232 DB9 1

[Snoop168]

Trying to send one physical DB9 port off of phone system to 2 devices. One is a buffer box for SMDR, the other is an audio recorder which also needs the smdr... Can this just be done by grabbing the RX and GND pins and sending them to the second device that needs to receive the data?? Neither the Buffer box nor the audio recorder need to TX on the rs232 connection...

Or can this only be done with a device such as

http://www.bb-elec.com/Products/Serial-Connectivity/Serial-Data-Tools-Adapters/Port-Splitters.aspx

Thanks!

 

[robertjo24]

Yes, you should have no problem just using the RX and GND pins off he serial connection if you are certain no TX will be required. They are effectively shorted across in the splitter device. One thing to be careful of is cable distance though.

 

[edfair]

Works fine. Have done it for years looking at data moving between systems.

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

 

[Snoop168]

Thanks Guys... I'm wondering if the length is the culprit then. I'm converting from DB9 to RJ45 for an extended run and mostly because the wires were already in place as spares. I then switch back to DB9 for conncetion to the second device. Both devices work fine on their own, even the one with the long run but as soon as they are both hooked they both dont work. Trying to test by removing the pins from the RJ45 since its a bit easier than trying to remove the pins from the actual DB9 connector.. Is there any actual limit on the length when you go to RJ45. I have heard all different, most I see say Serial is limited to 50 ft. Does the fact that I am now running on twisted pairs give me any more length? Aside from any recommended length constraints, it still works on its own so i'm not totally convinced that length is the problem except if now with the 2 devices hooked up theres too much resistance on the circuit...??

Either way when i remove one wire which I suspect to be the TX wire the first device "appears" to be working, but its not a device I can get into to confirm so I am just going by the lights on the port... But the second device still does not seem to work and since I am suspecting that I am removing the TX wire I can't try to send an enter key to see if I get a response. Only just hope that I catch an automatic line that gets sent from the phone system..

 

[edfair]

Length issues with 232 are "Alice in Wonderland" related. When they are good they are very, very good. When they are bad, they are very, very bad. I've had issues at 25ft and no problems at 250ft. Some depends on the line drivers and receivers if you are lucky enough to have discrete silicon rather than "all in one processors".

Got to assume that you don't have Tx to Tx and Rx to Rx. Might be well to get a cheap VOM to check the transmitted and received voltage levels at both ends. You may also need to dummy the handshaking signals at both ends.

If you can slow down the communications speed it might help.

Cat 3 or 5 sholdn't adversly affect your signals. But it won't enhance them either. One issue is the additional resistance through the additional jacks. I've also run across issues where the cat5 was direct connect to the wrong kind of keystone jack.

Under similar circunstances I have used a scope to see what the data bits look like. At long distances there is a lot of rounding of what generally is a sharp transition.

Line drivers are of some help. They have a different way of signalling that is not as affected by distance.


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

 

[mforrence]

Remember that serial uses ground reference - if the two devices don't share a ground, you'll have a problem. They also need to have identical ground potential as compared to the transmitting device. Use powered short-haul modems to overcome the trouble.
Mike

 

[Snoop168]

Yeah, essentially all 3 devices are sharing a ground and I am thinking thats where the trouble is coming into play?? IDK. I tested by limiting both of the receiving devices to only getting the TX and GND pin from the source and putting it in on the RX and GND pins on each of the receiving devices... No matter what the length of the cable is it doesnt work so I think that proves that the length is not an issue...

 

[wires]

I'm assuming that all devices are using the same baud/bit protocol and work individually when just connected with grnd and the TX pin of the phone system to grnd and the RX pin on either the SMDR or audio recorder. Step one is to verify this. You might have a handshake problem that requires extra wiring to get things talking and/or listening. A Serial Break-out Box is very helpful in troubleshooting this sort of thing

If all is good then you likely have a signal distortion problem. This can be "rounding" of the signal transitions due to line capacitance/inductance or grounding issues. This sort of thing is fairly easy to troubleshoot with a oscilloscope. Lacking that some short haul modems will likely do the trick as suggested by mforrence. Something like this would likely do the trick. Giving B&B a call would likely be helpful.

 

[robertjo24]

Make sure that both devices in parallel are either DCE or DTE. If one is DCE and the other DTE, the TX & RX pins will be swapped. Check the RX and TX for all devices and make sure which pins are which. You can do this easily by taking a volt meter and measuring the potential from GND to pin 2 & pin 3 (of the DB9). The TX pin will show voltage, while the RX pin will not.

 

[Snoop168]

Just want to thank everyone. The way I had it wired should have worked. The problem is that the PBX vendor when on site put a new DB9 to RJ45 connector on the PBX thereby UN-doing the cross that the old one had. I was wired fine but since the call accounting box was wired straighthrough there were 2 TX's on the same wire causing all to fail. Thanks Again!