Communicating with MKS MFC

[Jacob]

Hi,

I have a 1179A MKS MFC with RS-485 serial connection. I am trying to control this MFC through labview. I have a NI PCIe-8431/8, 8 Port, RS485/RS422 Serial Interface that I trying to communicate through. The MFC uses an RJ-11 cable for its digital RS485 communication. There is ground, D+, and D- pins that I have isolated coming out of the RJ11 and have a converter for a DB9 connector. I have tried a number of different pin connections in an attempt to communicate with the MFC all of which do not seem to work, mostly focusing on combination's of connecting the D+ to RXD+ or TXD+ and the D- to RXD- or TXD-. Using VISA I try to send a command to the MFC and I read what appears to be totally random responses. Not sure what I am doing wrong but MKS does not seem willing to provide much support so any help would be great! There is the correct power going to the MFC as well. Thanks

Jacob

[Tim_S]

I have a 1179A MKS MFC with RS-485 serial connection. I am trying to control this MFC through labview. I have a NI PCIe-8431/8, 8 Port, RS485/RS422 Serial Interface that I trying to communicate through. The MFC uses an RJ-11 cable for its digital RS485 communication. There is ground, D+, and D- pins that I have isolated coming out of the RJ11 and have a converter for a DB9 connector. I have tried a number of different pin connections in an attempt to communicate with the MFC all of which do not seem to work, mostly focusing on combination's of connecting the D+ to RXD+ or TXD+ and the D- to RXD- or TXD-. Using VISA I try to send a command to the MFC and I read what appears to be totally random responses. Not sure what I am doing wrong but MKS does not seem willing to provide much support so any help would be great! There is the correct power going to the MFC as well. Thanks

Have you looked into the port settings in MAX? It sounds like you have a half-duplex communication; you will need to set the port up in MAX the same way.

Tim

[Jacob]

Have you looked into the port settings in MAX? It sounds like you have a half-duplex communication; you will need to set the port up in MAX the same way.

Tim

Thanks Tim for responding. I have looked into that and although I am not sure I understand it fully I have set MAX for two leads on auto. I get something back but it is not what I put in or close

[asbo]

That sounds a little bit like the behavior you get when you choose the wrong port speed. Have you confirmed what baud rate the device is communicating at?

Edited November 2, 2010 by asbo

[Tim_S]

Thanks Tim for responding. I have looked into that and although I am not sure I understand it fully I have set MAX for two leads on auto. I get something back but it is not what I put in or close

I'd suggest not using automatic detection; I've not had good luck with it. You should certainly perform Asbo's point in it's not just the duplex settings that you need to check.

Tim

[Jacob]

Thanks you guys I will try both of those and get back to you about how it goes!

[Jacob]

Everything is set correctly and still does not work...

[asbo]

Are you sure the data coming back is intended to be human-readable?

[ShaunR]

Posting the command reference manual would be helpful

[EricLarsen]

When serial communication doesn't work, the first suspect is always the cable. Double and triple check you have the connections done right. But the fact that you are getting some data back might indicate the connections are correct, but triple check them anyway.

I've used quite a few 1179As in the past and have had really good luck with them. Somewhere around I have a cable diagram, if I can scrounge that up I'll send that to you. In the meantime, here is a small library of VIs I wrote to communicate with the 1179As. Sorry, they aren't documented very well, but they should be pretty self-explanitory. Just open up MKS 1179A Tree.vi for access to the top level Vis.

Each 1179 has to be assigned a unique address, and I don't remember how to do that, but you'll need to know the address to communicate with the proper unit. It seems to me that all units start with a default address (255?). Something like that. You then use the change address VI to set it the address you want. You also need to calculate the proper checksum to transmit to the instrument, my code will show you how that's done.

The Wink.vi command is a very handy utility to run. It will cause an LED on the MFC to blink, which will tell you if you have communication established.

MKS 1179A.zip

[Tim_S]

Everything is set correctly and still does not work...

Assuming you're still having problems... You have a multiport card; have you tried to loop the port back into another in the card? Are you sure you have a good cable? Plugged into the port you think you are? Have the connector pushed on all the way? (That last one bit me again a few minutes ago.)

Tim

[Jacob]

When serial communication doesn't work, the first suspect is always the cable. Double and triple check you have the connections done right. But the fact that you are getting some data back might indicate the connections are correct, but triple check them anyway.

I've used quite a few 1179As in the past and have had really good luck with them. Somewhere around I have a cable diagram, if I can scrounge that up I'll send that to you. In the meantime, here is a small library of VIs I wrote to communicate with the 1179As. Sorry, they aren't documented very well, but they should be pretty self-explanitory. Just open up MKS 1179A Tree.vi for access to the top level Vis.

Each 1179 has to be assigned a unique address, and I don't remember how to do that, but you'll need to know the address to communicate with the proper unit. It seems to me that all units start with a default address (255?). Something like that. You then use the change address VI to set it the address you want. You also need to calculate the proper checksum to transmit to the instrument, my code will show you how that's done.

The Wink.vi command is a very handy utility to run. It will cause an LED on the MFC to blink, which will tell you if you have communication established.

Still not having much luck so if you could find that wiring diagram that would take away a large part of the variable we still have. Thanks for the help though!