Maker Pro
Maker Pro

RS422 to RS232 converter.

HellasTechn

Apr 14, 2013
1,579
Joined
Apr 14, 2013
Messages
1,579
Is the uC for just monitoring the the "working line"?
Yes.

I know that the ic's are the best and probably the only way to go (for ttl output).

I am new to this so i do need guidance. could you provide me with the datasheet for the dual input rs422 chip ? it looks like i cant find any. (uA/MC9637).

I would like to proceed with a high quality chip because i had a bad experience with fake max3232 chips and i would not like to have similar problems.
 
Last edited:

HellasTechn

Apr 14, 2013
1,579
Joined
Apr 14, 2013
Messages
1,579
I have been testing the ua9637 chips and i came across a little strange behavior.

I use rs422 signal comeing from two GPS receivers. I feed GPS A TO Input A and GPS B to input B (the - and + of each gps output connected correctly to the inputs)

now it works just fine but when i disconnect GPS A and leave GPS B, i would expect the rs232 1 output to remain high (idle state), instead the chip just passes whatever signal from rs323 2 output also to 1 output.

The same happens when i remove GPS B and leave GPS A connected. Instead of haveing an idling (high) RS232 B output the signal from RS232 1 output also appears on both 1 and 2 outputs.

If i remove both GPS A and GPS B inputs only then both rs232 outputs stay high.

Why is that a problem to me ? because the RS232 outputs are then connected to a Picaxe mucrocontroller that reads each rs232 gps signal and determines if the GPS receiver is active or not. If i remove one of the rs422 inputs i want the coresponding output to go high "remain at idle state" so that the picaxe can detect that there are no data geing transmitted from the coresponding gps device.
 
Last edited:

HellasTechn

Apr 14, 2013
1,579
Joined
Apr 14, 2013
Messages
1,579
I guess i could utilise two ua9637 chips (one for each receiver) but that would require more space on the pcb...
 

dorke

Jun 20, 2015
2,342
Joined
Jun 20, 2015
Messages
2,342
That shouldn't be like that !

When you say "disconnect" ,what exactly do you do?
Where do you disconnect?

Are you using line termination?
Are you using twisted pairs?
What length are the wires?

Please show schematics and photos of the board and connection.
 

HellasTechn

Apr 14, 2013
1,579
Joined
Apr 14, 2013
Messages
1,579
When you say "disconnect" ,what exactly do you do?
On the breadboard i remove both A+ and A- (or B+ and B-) wires from the chip's input pins.

Are you using line termination?
Now i do not. I thought it was not needed.

What length are the wires?

Not more than 20cm total distance from the RS422 transmitter device.

Are you using twisted pairs?
No just straight cables.

Please show schematics and photos of the board and connection.
I thought i had uploaded the schematic... sorry, I will upload in a couple of hours because i dont have it with me at the moment.
I will take pictures of the breadboard circuit and post them !
 
Last edited:

dorke

Jun 20, 2015
2,342
Joined
Jun 20, 2015
Messages
2,342
Well,
You need to use twisted pair(you can twist them yourself for short distance).
Use a termination resistor between A- and A+ same for B,value should be about 100 ohm.
waiting to see schematics and photo.
 

HellasTechn

Apr 14, 2013
1,579
Joined
Apr 14, 2013
Messages
1,579
Here is the diagram. I dont have a picture of the breadboard right now but there is nothing really to see there. Its all wired up just like in the diagram.

Note that while i was probing the outputs with the Oscope, i had nothing connected on them.
 

Attachments

  • UA9637.png
    UA9637.png
    13.2 KB · Views: 97

dorke

Jun 20, 2015
2,342
Joined
Jun 20, 2015
Messages
2,342
I see a few things:

1. RS422 needs no return ground,it should be disconnected.
2. Put termination resistors as close as possible to the IC -pins(shortest legs).
3. Use a de-coupling cap ,as close as possible to the pins(shortest legs).
4. Use twisted pair wires for the RS422 lines.
5. Why the name "RS232"? it is TTL level there.
6. A photo is always very important,even if you think it is "just like the diagram".
7. You should probe the outputs and the inputs at the same time,again a photo of what you found may be very helpful.

UA9637.png
 

HellasTechn

Apr 14, 2013
1,579
Joined
Apr 14, 2013
Messages
1,579
I read you...

i will breadboard the circuit as shown in your diagram and take pictures and also scopings. The problem is that i will not have the gps devices in hand for about a week. I will get back here as soon as i have something.

Thank you !

P.S.
The cap between VCC and GND is a decoupling cap right ? does it have to be electrolytic or it does not matter ?
 

dorke

Jun 20, 2015
2,342
Joined
Jun 20, 2015
Messages
2,342
Yes,de-coupling.
It should be ceramic type, very short leads ,X or Y type(i.e X5R,X6R,X7R etc.) but not Z!
 

HellasTechn

Apr 14, 2013
1,579
Joined
Apr 14, 2013
Messages
1,579
I have been testing the ua9637 chips and i came across a little strange behavior.

I use rs422 signal comeing from two GPS receivers. I feed GPS A TO Input A and GPS B to input B (the - and + of each gps output connected correctly to the inputs)

now it works just fine but when i disconnect GPS A and leave GPS B, i would expect the rs232 1 output to remain high (idle state), instead the chip just passes whatever signal from rs323 2 output also to 1 output.

The same happens when i remove GPS B and leave GPS A connected. Instead of haveing an idling (high) RS232 B output the signal from RS232 1 output also appears on both 1 and 2 outputs.

If i remove both GPS A and GPS B inputs only then both rs232 outputs stay high.

Why is that a problem to me ? because the RS232 outputs are then connected to a Picaxe mucrocontroller that reads each rs232 gps signal and determines if the GPS receiver is active or not. If i remove one of the rs422 inputs i want the coresponding output to go high "remain at idle state" so that the picaxe can detect that there are no data geing transmitted from the coresponding gps device.


Back again digging this long dead thread. Like dorke suggested in #30 but a little differently...I pulled the inputs to ground through a 10k resistor. the problem went away and the UA chip works fine !

Is correct to pull the + and - inputs to ground through 10k resistor ? Or should i stick to Dorke's reccomendation to put a 100 ohm resistor between + and - inputs ?

Thankyou !
 
Last edited:
Top