Jump to content

As of July 17, 2015, the LabJack forums here at forums.labjack.com are shut down. New registrations, topics, and replies are disabled. All forums are in a read-only state for archive purposes.

Please visit our current forums at labjack.com/forums to view and make new posts. To post on the current forums, use your labjack.com login account. Your old LabJack forums login credentials have been retired. There are no longer separate logins for labjack.com and LabJack forums.


Photo

U6 Input impeadance


  • Please log in to reply
3 replies to this topic

#1 JCS5920

JCS5920
  • Members
  • 2 posts

Posted 22 August 2013 - 03:51 PM

OK here goes.. I have 8 identical circuits that feed 8 input channels of a MUX4051 (74HC4051). Using the extending mapping channels I can access these channels by the extended channels list where the LJ changes the CIO-0,1,2 lines to address the multiplexer and switch in the proper channel. Now let me preface this by saying the UE9 does not have this issue it is only seen in the U6 (2 separate devices tested). My circuit is an op amp (2252A) setup in trans-impeadance mode) op amp positive input wired to a 1.25V reference (lm4140-1.2) and the circuit is a R1 = 20 ohm R2 = 33000 ohm.  The output of the circuit is fed into an RC filter with 1M .1uF cap, simple low pass filter. When the MUX is addressing channel 0 (CIO 0,1,2=LOW) the voltage read is approximately 3mv lower than expected. When I jumper the 1M (inline resistor) this effect goes away and all channels read as expected. This only happens on MUX select channel 0. So I figure this is an input impedance issue?? but how? My question is what is different about the 74HC4051 when it is selecting a channel other than 0? Are channels 1,2,3,4,5,6,7 somehow buffered internaly, an d channel 0 is fed through? and if so why is this an issue with the U6 only? the UE9 works perfectly. Maybe a different MUX would work?? I hope I have given enough info to get this ball rolling.  I really want to use the U6 but until I get a resolution to this I simply have to use the more expensive UE9. 



#2 LabJack Support

LabJack Support
  • Admin
  • 8677 posts

Posted 23 August 2013 - 07:22 AM

1M input impedance is beyond specifications for the UE9 and U6.  Perhaps all the readings are wrong from both devices?

 

Try changing the series resistor to 1k.  You can use higher resistance in some situations, but all ranges & resolutions are specified at 1k or less.

 

To confirm it is a settling related issue, change the order and timing of your acquisition.  What are you using for software?  You can test with LJLogUD or LJStreamUD.  Do a test where you acquire channels 0, 1, 2, and 3.  Then do a test where you acquire 3, 2, 1, and 0.

 

How fast are you scanning your inputs?

 

What range and resolution are you using, or do you want to use?



#3 JCS5920

JCS5920
  • Members
  • 2 posts

Posted 26 August 2013 - 08:45 AM

Yes 1k resolves the effect unfortunately I have hundreds of fielded boards which makes this a painful solution. I think I will make a buffer circuit to go in between and offer it as a free update. Thanks for the help 



#4 LabJack Support

LabJack Support
  • Admin
  • 8677 posts

Posted 26 August 2013 - 11:29 AM

Static error due to bias current:  Are you seeing this?  Both the U6 & UE9 have a typical static bias current of around 20nA, so I would expect the 1M resistor to cause a 20mV shift.  If you compare a channel with the 1M versus a short do you see a 20mV difference?  Perhaps this is not a problem for you because they all have the same shift.

 

 

Dynamic error due to mux switching:  This seemed to be what you were noticing where 1 channel was 3mV different than the others on the U6 only.  This type of settling error is dependent on the scan order, scan speed, resolution, and range, so perhaps you can just make some software tweaks to eliminate this.  Let us know what you are using for scan order, scan speed, resolution, and range.

 

Seems like you are saying that you scan 0, 1, 2, 3, 4, 5, 6, and 7, in that order.  You see a 3mV shift on 0 compared to 1-7 which all read the same.  A quick thing to try is to simply add an extra read of 0 to the beginning of your scan list, so you are doing 0, 0, 1, 2, 3, 4, 5, 6, and 7.




0 user(s) are reading this topic

0 members, 0 guests, 0 anonymous users