RS-422 Trouble

[Twlighting]

Hey There,

Banging my head against a wall trying to get 'some' response out of an RS-422 device. In this case it's an ultrasonic anemometer which will ultimately be connected to an AMX controller to provide windspeed/direction readings.

The anemometer is capable of full duplex 422 and half duplex 485 at various common baud rates, however it has no external means to configure these settings, they must be modified over whatever the current serial interface is. Being this was a used unit, it's not clear what the current settings are.

The AMX device has a number of configurable 422/485 ports, so I've spent the evening swapping around TX/RX, flipping polarity, and changing port speed settings with no success in receiving data. So I've turned to the DMM to start measuring voltages on the controller and anemometer.

I've found the controller TX is usually held around +/- 4V between TXA and TXB, and RXA and RXB are around +/-300mV. This seems correct. Performed a few loopback tests on the controller, ports are working as expected.

Interestingly the anemometer has 0V between TXA and TXB, and also 0V between RXA and RXB. I am able to measure about +373mV from RXA to Gnd and RXB to Gnd, and +9mV from TXA/B to Gnd. This voltage disappears when I disconnect power to the anemometer, so that leads me to believe it's at least turning on, though there are no LEDs or other indicators that is the case.

Any guess if this would this point to the connection being setup for RS-485? Or to some sort of a failure on the unit?

As far as the units themselves:
Anemometer: WindObserver II | Gill Instruments
Controller: https://www.avace.com/pdf/NXI_installation.pdf

Thanks for any insight,
-Griffin

 

[RT48]

There's no way to reset the device to some known factory default?

 

[Twlighting]

There's a 'safe mode' where you power down the anemometer, set up a 19200 baud 422 connection and then power up the device while holding down * in a terminal application. In theory that should allow you to get into it and modify settings, however I haven't had an success getting a response this way either.

I'm picking up a 232>422/485 converter later to remove the AMX gear from the equation and see if I have any luck going direct to the PC... we'll see.

Thanks,
-Griffin

 

[bharvey2]

It's been a long time since I've messed with RS422/485 but the RS232/422 adapter would be a good place to start. Also, I recall using terminating resistors (120ohm ?)

 

[Twlighting]

Sadly didn't have much luck with the new converter.

Ended up speaking with the manufacturer and putting it on a scope, seems there's an issue with the power supply circuitry in the device.

It's pulling around 25mA where nominally it should pull 50mA. Looks like the photos from the scope are too large to attach, but that basically confirmed what I was seeing on the DMM, a clean signal coming from AMX and the converters around +/- 5V and just a few hundred mV coming from the anemometer.

So the next challenge is finding a 7mm socket/nut driver with an OD of 9.5mm to remove the recessed bolts from the unit to access the PCB inside and perform a visual inspection on that. Have a feeling I'm going to have to grind down a normal 7mm socket to fit.

 

[bharvey2]

Wow, that's going to be a thin walled socket. I've modified a number of tools myself. Buy a cheap socket, chuck it up in a lathe or drill press and cut away!