EMI generated by RS232 charge pump?

No, but if it diminishes materially when a cable is connected I'd suggest it isn't the charge pump. Maybe loading the receivers with a say 22k (or so) to ground would help.

Doubt it. They don't determine the pump frequency.

Well, I didn't like what they were doing to the power supply on some of our boards, so if there was analog circuitry on them we started using a tantalum cap as a bypass instead of the monolithic ceramic. It took out a lot of the noise. Might be worth a quick test.

Gary Peek Industrologic, Inc.

Thanks for the feedback Gary, I'll give it a go.

Thanks, I'll investigate this further.

Greetings, Mark.

I had a very similar problem very recently with a design. The power supply line was very noisy due to the MAX3222. I did 2 things which greatly quieted down the problem.

And always remember the golden rule:

Note that the source (the MAX3222 IC) is still screaming as loudly, but it now has no antenna to radiate its noise (i.e. the power supply line). Note also that this does not address radiation problems from the RS-232 lines, when connected. That is another issue.

Go to it!

Mario C.

markp wrote:

the

that

We had a problem with radiation from the RS-232 lines where I used to work. Filtering with ferrite beads and capacitors solved the problem.

the

and check the dielectric of the capacitor: dont use Z5U/Y5V, use X7R/X5R/XB

that

Thanks Mario. There's a PCB update coming up so I'm going to try to put on some components as you suggested.

Mark.

How about useing DTR/DSR to switch on or off the power supply to MAX3222E ..? (via some standalone level converter)