SV1GAP: Comparison of 2 RS-232 programming circuits

Update 2017_3_28

Sources:

http://www.sm5nvc.se/erak/mod/index.php?sort=date&path=rig/YAESU/kabel/

http://www.sm5nvc.se/erak/mod/rig/YAESU/kabel/icq7-int.gif

http://www.cqham.ru/un7gm_icom.htm

http://www.cqham.ru/opc478.gif

http://image.prntscr.com/image/dc424cad890843f79765360a4fd1cde7.gif

2017_3_28 Start of update

The comparison concerns 2 circuits for connecting a transceiver to a RS-232 port. Recently someone reported a problem when his transceiver (Yaesu VX-7) was sending data to the RS-232 using the circuit below:

I searched for similar circuits and I noticed that in all the other circuits that I found, a different topology with a PNP transistor (instead of a NPN one) was used to connect the transceiver to the RxD (pin 2) port. Looking at the above circuit I suspected that it provides insufficient voltage and insufficient output resistance when connected to a load (RxD port).

http://www.ti.com/lit/an/snla037b/snla037b.pdf: "Receiver Input 3 kΩ to 7 kΩ"

http://www.arcelect.com/rs232.htm: "Receiver Input Resistance (Ohms) 3k to 7k, 4k min."

So I added R_Load to the outputs. As it is obvious from the following images, the effect of the R_Load on the circuit on the left (Label Net: Out_NPN) is very high. I tried collector resistors of lower value than the original 4.7kΩ and I chose a 1.5kΩ resistor. With the 1.5kΩ resistor, even in the worst-case scenario of the 3kΩ receiver input resistance, a voltage above 3V occurs in RxD port during logic level 1.