Expected Ground Current in RS-232

[BrownTown]

So anyways, I am designing a PCB for my senior project and the board is pretty packed, so there isn't lots of room and therefore some of the traces are getting pretty thin, so the question I have is does anyone know what sort of current the Gnd pin should be expected to sink? Since it is just a reference ground it seems like not much, but there is always the possibility that for whatever reason the computer on the other end might dump more current in than is expected and mess up the signaling. Right now the trace is only really rated for ~250mA I would say which seems just fine for a signal pin, but who knows?

 

[So]

Although I've never tested it, I seriously doubt that you're going to see more than 250mA in the ref. ground for a serial connection. I generally use RS232 for interfacing with micro controllers, and I did up the size of my entire ground mesh (node, whatever -- I'm not completely sober at the moment) in when doing my last PCB, but I am taking the return current for some 190V nixie tubes - on the order of 100mA.

 

[Nik]

NEEERRRRRRRRRRDS!!!!

kidding. I have nothing constructive to add to this thread. Goodnight.

 

[Qacer]

I think that if your PCB's only ground reference is via the RS-232 port, then your maximum sink current going thru your trace should just be whatever current consumption your circuit system requires. I'm not sure what the standard is for maximum current draw for RS-232, but I did find this:

"RS-232 drivers and receivers must be able to withstand indefinite short circuit to ground or to any voltage level up to +/-25 volts."

I think what you should do is just power up your circuit and test it with a 25V signal level and measure the current coming out of your ground pin.

 

[Colt45]

damn close to zero.

tx is limited to something on the order of 10mA

as long as you can see the trace it should be fine

 

[edro]

Is the board externally powered?

 

[PottedMeat]

Originally posted by: Colt45
damn close to zero.

tx is limited to something on the order of 10mA

as long as you can see the trace it should be fine

Yeah, thats what I would do. Hey if the current surges you've got a 'fuse'. And that cable is just for signaling, not any significant power transfer.

Is the board externally powered?

I would guess so if it does anything really useful. After all how many things are powered only though the RS232 port need more than a few mA.

 

[Analog]

We usually connect the ground pin to the ground plane of the PC board. You didn't mention whether you had a ground or power plane or how many layers.

 

[Modelworks]

Are you connecting directly to a device that uses default pc level rs-232 or a device using a Max232 type chip to convert to TTL?

If its a Max232 , it only requires 10ma Max.

 

[BrownTown]

I am using the Max232 chip to do the connection. As for the PCB ground plane, this is only a 2 layer board so there aren't really dedicated ground and power planes. As for power source, the board is powered by 120V AC from the wall and has transformer, capacitors and voltage regulators on board for +5V and +18V.

Anyways, thanks for the advice I think it will all be fine the way it is, maybe if there were a fault on the line there would be a problem but TBH it only really has to work long enough for me to demonstrate it and such

 

[Pepsi90919]

Originally posted by: GuideBot
NEEERRRRRRRRRRDS!!!!

kidding. I have nothing constructive to add to this thread. Goodnight.

USER WAS PUT ON PROBATION FOR THIS POST