USB to RS485 Converter Design troubleshooting

Thread Starter

mmbot

Joined Jul 10, 2019
3
Hi, I made this USB to RS485 Conveter and have been trying to get it to work but have had little success. Wondering if anyone can help me out?

I have been communicating with a PID controller using an off the shelf USB to RS485 converter and thought it would be fun to try to make my own.
The circuit uses the FT230X and MAX13082, as far as I can tell I have the schematic correct and have the FT230X programmed correctly. Using a scope I can see that there is information being sent and that the levels appear correct compared to the off the shelf version but I don't have a scope that can read the info being sent.

The schematic I used is basically right from the FT230X datasheet. I omitted the pull up resistors but have since tried adding them in but that did not change anything.

Any help/ideas would be appreciated
 

Attachments

MrChips

Joined Oct 2, 2009
20,215
Depends on how the device communicates via USB.
USB is not UART.

The purpose of a USB-to-UART bridge is to take UART signalling and send it to a USB host and vice-versa.
The USB host requires a USB stack and device handler which makes the USB port mimic a serial COM port.
 

Thread Starter

mmbot

Joined Jul 10, 2019
3
the FT230X appears as a COM port and can have its settings changed with the FT_Prog utility so I'm fairly certain that end is working correctly.

I am currently using a piece of software that communicates with the PID (Fuji PXR3) though a purchased USB to RS485 converter that utilizes the FT232R. This chip also appears as a COM port and can be programmed through the FT_Prog utility.

I can only assume that something is wrong with the FT230X sending signals to the MAX13082 but from what I can tell it should be correct

The programming of the FT230X chip is as indicated on the schematic
 

MrChips

Joined Oct 2, 2009
20,215
Since you have a scope, check the polarity of the RS-485 signal and confirm that the A & B connection is the correct polarity.

Next, check the biasing on the RS-485 A and B lines. One side should be pulled up to +5V and the other pulled down to GND (1kΩ resistors should be fine). Usually A is pulled up and B is pulled down. However some people label A and B reversed.

Next, check the bit rate, i.e. measure the shortest bit width of a typical data transmission. Confirm that it matches the required baud setting.
 
Yeah, the schematic looks OK. I agree you should check you RS-485 polarity. This is a common problem. The Maxim part is set up for B to be more positive than A when the line is "marking". This is the polarity of a "1" bit and also the polarity of the line between characters. You often see "A" and "B" assigned the other way.

To bias this chip, you would pull "B" up and "A" down.
 

Picbuster

Joined Dec 2, 2013
990
Hi, I made this USB to RS485 Conveter and have been trying to get it to work but have had little success. Wondering if anyone can help me out?

I have been communicating with a PID controller using an off the shelf USB to RS485 converter and thought it would be fun to try to make my own.
The circuit uses the FT230X and MAX13082, as far as I can tell I have the schematic correct and have the FT230X programmed correctly. Using a scope I can see that there is information being sent and that the levels appear correct compared to the off the shelf version but I don't have a scope that can read the info being sent.

The schematic I used is basically right from the FT230X datasheet. I omitted the pull up resistors but have since tried adding them in but that did not change anything.

Any help/ideas would be appreciated
RS485 1/2 duplex I normally connect pin 2 and 3 at the max ( it's send or receive)
Did you load the 1/2 duplex rs485 settings into the ft230?
Full duplex won't work as pin2-3 are not controlled @ full duplex.
Same applies to the virtual port usb driver ( using the control lines to switch between send and receive) (it's rts I think).

Picbuster
 

Thread Starter

mmbot

Joined Jul 10, 2019
3
Hi, Thanks for everyone responses. I haven't had much time to work on this until recently. I added some biasing to the lines, B up and A down and now in a terminal I can see that it is receiving information. I'm just hitting the wires together and seeing random characters which is better than before. I can see on the scope that it is sending information as well however it is still unable to connect to the Fuji controller in the same manner as my purchased USB to RS485 converter.

The next thing I can think of to try is to have this converter talk to with purchased one to see in terminal if it is sending information correctly.

Does anyone have any other ideas of what I should look for now?

Thanks.
 
Do you have termination at both ends of the bus?

Check the number of data, start, stop bits for RS485.

Addressing.

baud rate.

RS232 was easy. RS485 was a nightmare to get set up. Can't stress the polarity issue.

What changes the mode from transmit to receive? Converters generally use a time-out, I believe.
 
Top