Hello. I'm using analog input 0-10v on a U6 Labjack. The system works but I have lack of cosistancy/accuracy. Measurements of a linear pot return values that are close but vary about .02 inches.
I have (maybe incorrectly) determined that the speed of the USB is causing the issue. I have searched the Profilab, and the Labjack details and cant figure out if the USB rate is changeable.
If this is not feasible, would I be worthwhile going to the Ethernet connection?
Can you help me with this?
.png%3Fitok=PHGBxf6M "EEV Blog Reviews LabJack T7-Pro & Software Tools")
.jpg%3Fitok=B1YcuDFK)
.png%3Fitok=Ybt-OlgU "LabJack + JEDI One Can Create a Powerful Distributed DAQ Ecosystem")
Are you thinking that the U6 is reading the correct value at the time it takes its reading, but the time of that reading is not when you think it should be and that leads to a problem in your measurement?
That is, if you just read a voltage that is not changing, the U6 reading agrees with a high-accuracy DMM connected to the same AINx/GND at the same time?
Thanks for the response - The actuator is moving, and I'm determining when to stop it based on the analog voltage. Moving slower gets more accuracy, so I believe that the comms speed being slow creates a lag in accuracy. Not sure really, I suppose it could be the computer speed?
I don't know how to determine the USB speed hence my question regarding baud rate?
I believe that the Labjack is reading the "stop" voltage but since the actuator is moving, it would be possible that the reaction time would be at fault.
I tried a battery in place of the power supply to remove the possibilty that the ripple was causing an issue.
I would confirm that when your system is not moving you get the proper reading. And then further if you can look at the data from when it is moving and confirm that the readings seem to lag behind the expected values, you would then think the problem is either:
1. The acquired reading is the proper value at the time it is acquired, but the time when it is acquired is later than assumed.
2. The reading is acquired at the proper assumed time, but due to some sort of analog signal speed problem the signal is not changing as fast as needed. The U6 acquires the actual voltage present at the assumed time, but that voltage is to slow to change and has not changed to the proper value.
Seems at this point you are assuming #1. We can't speak to the timing details of Profilab, but the typical times of the U6 for command-response reads are in Section 3.1:
https://labjack.com/support/datasheets/u6/operation/command-response
At Range=10 & ResolutionIndex=1 it should take about 0.6 + 0.05 milliseconds per read. For a test, can you do a simple untimed loop in Profilab reading a single AIN and note how many times that loop iterates per second?
I ran a test on my machine using a LabVIEW exampled called "U6 CR Speed Test.vi" and see about 0.29 ms per read. I attached an executable so you can try it. Set #Iterations to something bigger like 2000 and set Resolution=1, then click the run arrow. We provide similar examples (called AllIO) in most of our UD programming example archives.
Note that we are getting messages from Office365 that it is blocking the notification emails the forum is sending you. See attached PDF with bounce message.
I see - I will change my email address to fix that. Thanks for the response. I'm getting close because of your note [Thanks] - I have no idea how to change the range or resolution /sample rate or the analog input using profilab. Looked everywhere.
Thanks for the detail - The I have no idea how to use the Profilab to change analog conversion rates/speeds etc. Also, this message board is confusin - I press save and my email goes away - No feedback that anything has been delivered?
After you add a new comment to a forum topic, you should see a warning message saying that the post will not be visible until we approve it. We have to do that to keep all sorts of spam from getting posted.
You will have to contact Profilab, but it is possible they just use the defaults for Range & ResolutionIndex and do not allow you to control it.
If you want to see how the U6 is performing with default values, do the test in post #4 above but use ResolutionIndex = 8 or 9 (U6 or U6-Pro). At Range=10 & ResolutionIndex=1 it should take about 0.6 + 0.05 milliseconds per read, whereas at At Range=10 & ResolutionIndex=8 it should take about 0.6 + 1.33 milliseconds per read.