RS485 serial data send problem[SOLVED]

erhane · Joined: 01 Jul 2014 Posts: 41

When i want to transmit data for example send 'u' with putc('u'); command it sends ' ??u'

one blank two question marks than u.

there is codes and outputs.
//////////////////////////////////////
Code 1:

oxo · Joined: 13 Nov 2012 Posts: 219 Location: France

Proteus or ISIS not allowed here

http://www.ccsinfo.com/forum/viewtopic.php?t=47549

erhane · Joined: 01 Jul 2014 Posts: 41

this topic is not about proteus or isis

dyeatman · Joined: 06 Sep 2003 Posts: 1933 Location: Norman, OK

Why did you start a new thread?
Go back and continue with the one you started on this subject.
_________________
Google and Forum Search are some of your best tools!!!!

oxo · Joined: 13 Nov 2012 Posts: 219 Location: France

erhane · Joined: 01 Jul 2014 Posts: 41

erhane · Joined: 01 Jul 2014 Posts: 41

ezflyr · Joined: 25 Oct 2010 Posts: 1019 Location: Tewksbury, MA

Rename,

Ugh! Please come back when you have real hardware, as we don't dianose Proteus related problems. See the sticky at the top of the forum!

Also, it's obvious that you have spent NO time researching this issue, as your wiring for the RS-485 IC is contrary to that which is recommended.

There was a recent thread about this.....l.

John

erhane · Joined: 01 Jul 2014 Posts: 41

ok i will send real hardware photos in 5 mins wait. I cant beleive that you are not beleiving me tis is ridiculous

erhane · Joined: 01 Jul 2014 Posts: 41

RF_Developer · Joined: 07 Feb 2011 Posts: 839

Probably biassing resistors. RS485 has termination - the 120R resistor that MUST be at both ends of the bus, but NOT in the middle - and often biassing resistors - the pull-up and pull-down resistors that should only be in ONE place on the bus.

RS485 in half-duplex ("two-wire") configurations - the commonest arrangement - needs the biassing resistors. Full-duplex ("four-wire") configuration may not. In general any configuration where the drivers/transmitters are disabled must have biassing resistors to ensure the bus defaults to a proper state when nothing is driving.

The biassing resistors must be connected so that when not being driven the lower going signal is pulled low, and the higher going one is pulled up. This means when the bus is not driven it will be pulled to the inactive state. If the biassing is wrong the receivers will see spurious - "wrong" and unexpected - characters whenever a driver turns off.

RS485 suffers from confusion over which bus signal is which. Manufacturers of driver/receiver/transceiver ICs have never managed to agree on which signal is which and what they should be called. Perhaps the commonest naming is A and B, but some manufacturers, LT for example, call A, B and vice-versa, so requiring designers to connect their A pin to the B line on the bus. Others use other naming schemes such as + and -.

Here, TI have decided to use A and B for receivers and Y and Z for their transmitter. However, if you look carefully at the datasheet you'll see that TI use A as the non-inverting input and B as the inverting, and Y as the non-inverting output and Z as the inverting. However, you have A pulled up, BUT Y pulled down, and B pulled down, while Z is pulled up! I think, but I cannot be sure, that A and Z should be pulled up, and B and Y pulled down. I have spent a lot of time trying to troubleshoot a half-duplex RS485 network where the biassing had been swapped - it was very difficult to work out what the trouble was, I just got spurious characters.

Also there really should only be one pair of biassing resistors on an RS485 line pair. Where depends on the system configuration: on point-to-point and multi-driver, one receiver networks, it would make sense to put them on the receiver, while for single-driver, multi-receiver lines they make most sense on the driver. For half-duplex multi-drop peer to peer networks there's no clear choice, but like "highlanders", there can only be one. Short networks running at moderate speeds with more than one might work but the impedances will be messed up and won't work at high speeds or over long distances. The same goes for terminations - the 120Rs. They should only be at the ends of the lines, not in the middle. The clue's in the name "termination" - they terminate or "end" the transmission line.

So, to sum up: check the 560R biassing resistors. They don't look right.

erhane · Joined: 01 Jul 2014 Posts: 41

Thank You,

I will work on it and i will share results.

erhane · Joined: 01 Jul 2014 Posts: 41

Dear RF_Developer,

Thank you very much! i solved my problem with removing R21 and R22. Y should be pulled up and Z should be pulled down. it was vise versa. Now everything perfect =).