Does LabVIEW really work for serial communications above 9600 baud?

[bstedwel]

I have tried to use LabVIEW to communicate over a serial port at speeds higher than 9600 baud with no success.  I have a vi that works fine at 9600 or below, but fails to initialize the serial port at any higher speed.   I know the hardware supports the higher speeds, but it seems like LabVIEW has a built in limiatation of 9600.  (I seem to remember reading this about LabVIEW somewhere before).   

I see a lot of posts from people explaining what should be possible, but my question is this ->  Has anyone actually been successful at using serial communications in a vi at speeds higher than 9600 (for a serial port).    I'm trying to verify if this is a limitation of LabVIEW or not.  Thank you.

[smercurio_fc]

There is no inherent limitation in LabVIEW. What do you mean that it "fails to initialize the serial port"? Are you actually getting an error message? Which version of LabVIEW are you using? Which version of VISA? Do you have the right data size and parity set?

When you refer to the hardware are you referring to the target device? Have you confirmed this communication with HyperTeminal? I ask this because at higher speeds you usually have to do flow control and you need a good cable.

[TNeilson]

I'm writing an app write now that works up to 115200.  I'm using Modbus protocol, and using the NIModbus vi's (well, I tweaked them a little, but didn't do anything to the baud rate portion) for most of the coms.

[daijoubu]

I have used LabVIEW 7.1.1 at a baud rate of 115200. 

[TNeilson]

For faster baud rates, make sure your control is set for U32.

[Brian_Beal]

I have used Labview at 500K, without problems.  You just need to make sure that your hardware can do what you want it to.

[Ben]

"...and you need a good cable.'

and

"make sure that your hardware can do what you want it to. "

The important thing was mentioned by others but let expound.

At higher baud rates your cable can make a big difference. A small spike (picked from the environment) or reflections (due to impeadance mis-matches) can be large enough and long enough to look like valid line states and fill a bit cell time.

At lower baud rates, the logic in the chips can distinguish between the noise and the data because the glitches can get "over-sampled" out and rise times are not degraded.

Ben

Message Edited by Ben on 04-04-2006 12:51 PM

[bstedwel]

I'm using LabVIEW 8.01 and PDA 8.01.  The target device that fails to initialize is a Pocket PC PDA (Dell Axim X50).  I have verified communications at higher speeds using a laptop and hyperterminal between the PDA and the laptop.   My baud variable is U32.    Running the vi on the PDA with anything greater than 9600 throws errors during the serial init vi.  I have other commerical apps on the PDA that prove the PDA is capable of higher speeds and that the cable is good.   Changing the baud rate back to 9600 or lower in the vi makes it all work properly.  

[Ben]

"...Running the vi on the PDA with anything greater than 9600 throws errors "

"Now that is a horse of different color!"

I suggest you start a new post and make it clear you are talking about a PDA. I (unfortunately) can not speak for the PDA's.

Ben

Message Edited by Ben on 04-04-2006 12:59 PM

[TNeilson]

I did some work with a PDA and I think 9600 was the best I could get.