Opto Coupler on a RS-232 Signal Line

[rockbassist90]

I would like to make a shield on my PCB isolating the input/output side of the RS232 signals fiber optically...though I am unsure of the current requirement for use of RS232 and an optocoupler to work effectively in this application.

Any pointers?I would assume it's less than 5mA?

[Andreas]

Any pointers?I would assume it's less than 5mA?

Yes.  (5mA@5V is the mouse driveability spec).

what was the question?

With best regards

Andreas

[eneuro]

I would assume it's less than 5mA?

Was able to make optoisolated RS-232 <-> I2C inaterface and power two optos on PC side.
I2C MPU6050 Gyro & acceleroemter was tested and calibrated on Linux PC using such interface 
PC side is powered ONLY by RS-232 pins.
Of course MPU side two optos are powered from its own power supply, so it is galvanic insulated.
There is another advantage, while this MPU needed 3V PS it was easy power it up while PC side of course was a few kV insulated from MPU side.

Below working old prototype-proof of concept only, while now PC USB and CP2104 dongle with 4 GPIO and few fast optos is better solution for me than messing with RS-232 

[suicidaleggroll]

Do you want your interface to conform to RS232 specs, or do you just want to rig something together that works for this particular device and might not work for any others?

RS232 is a very "tolerant" specification.  If you want your device to truly conform to RS232, then you have to be able to handle a big range of voltages.

The latest specification is that valid voltages are in the range of +3 to +15v and -3 to -15v relative to ground, and -3 to 3 is invalid.  The device should also be tolerant of any voltage up to +/- 25v, and the driver and receiver must be able to withstand a short circuit to any voltage level up to +/- 25v indefinitely.

From what I understand, the current needs to be adequate to drive 2500pF of capacitance at 30V/us, which comes out to 75mA.  I'm not sure how close to this number off the shelf RS232 drivers come, but I have found a few references to "at least 45mA".

Now that's not to say that if your interface fails to meet one or more of these limits that it flat out won't work or will blow up, but that's the spec you should aim to hit.

[Andreas]

The latest specification is that valid voltages are in the range of +3 to +15v and -3 to -15v relative to ground, and -3 to 3 is invalid.  The device should also be tolerant of any voltage up to +/- 25v, and the driver and receiver must be able to withstand a short circuit to any voltage level up to +/- 25v indefinitely.

From what I understand, the current needs to be adequate to drive 2500pF of capacitance at 30V/us, which comes out to 75mA.  I'm not sure how close to this number off the shelf RS232 drivers come, but I have found a few references to "at least 45mA".

Sorry but I think that you should specify it clearer:

the +/-3 V is the forbidden voltage level range at the receiver.
The transmitter has to supply at least +/-5V into the 3-7kOhm load of the receiver Input.

And I do not think that the 30V/us are valid for the full capacitive load at least for standard RS232 drivers.
The SN75188 is somewhere around 3V/us at 2500 pF and limited current below 10mA according to datasheet.

http://www.ti.com/product/sn75188

With best regards

Andreas