Actual range and resolution confusion.

[Rogerio]

Sorry, I'm new to all this and have 2 questions. First, I checked everywhere but only got more confused aboud the analog inputs.
There are single ended and differential inputs, they both say +/-10 volts. I understand that on a differential input one side would go down to -10 volts and the other side to +10 volts (20v range?), but on a single ended one I don't understand this, I thought single ended meant 0-10 volts for example?
If I have a 0 to +10 volt signal connected to a single ended input (+ and gnd) do I get the full range of the input? 0v is 0 counts and 10v is 65,536 counts?

Also, this is more of an application question, if I need a reference 10v signal could I use a DC-DC converter on an analog output and monitor this signal with an analog input so it could be self calibrating? It would only use a few mA.

Thanks!

[LabJack Support]

You are confusing differential & single-ended with bipolar & unipolar. See the "Differential Analog Inputs" app note:

http://labjack.com/support/app-notes/diffe...l-analog-inputs

The U6 has 4 input ranges: +/-10V, +/-1V, +/-0.1V, or +/-0.01V. These are the same whether differential or single-ended, and mean that the difference between the positive channel and negative channel must be in that range. For a differential input the negative channel is another analog input, while for a single-ended input the negative channel is ground.

http://labjack.com/support/u6/users-guide/2.6

You can see that the actual nominal input range at +/-10V is -10.6V to +10.1V:

http://labjack.com/support/u6/users-guide/2.6.2

Appendix A has more specifications about the analog inputs, and note 7 points you to the signal range tables in Section 2.6.5:

http://labjack.com/support/u6/users-guide/appendix-a

http://labjack.com/support/u6/users-guide/2.6.5


The easiest way for you to get 10V at <=3mA would probably be the LJTick-DAC:

http://labjack.com/support/ljtick-dac/datasheet

Besides the LJTDAC, likely solutions would involve 2 parts:

-First, you would need a power-supply of perhaps +12V. You could use a DC-DC converter to convert 5V to 12V such as the 445-2448 from Digikey, or you could just use a wall-wart with 12V output.

-Then, you have something to provide a regulated 10V output from the 12V power-supply. One option would be to search Digikey for a "voltage reference" that would do the job (perhaps the AD581KH). Another option would be to put an LT1490A op-amp on an analog output (see Section 2.7.1.2).

[Rogerio]

Yes, I was thinking as if they where bipolar!
Think I got it now, I can connect my 0-10v signal single-ended and would cover the whole range of the input.
I could have a signal of 1-11v in a differential setting (As long as I do not exceed 11.9v) and it would give me the same "counts" or raw output as the above configuration.

Right?

And the LJTick-DAC does seems like the easiest way to solve the reference voltage.

Thank you!

[LabJack Support]

Yes, I was thinking as if they where bipolar!

Just to be clear, they are always bipolar, whether differential or single-ended. I think you were originally thinking that differential reading are bipolar and single-ended readings are unipolar.

Think I got it now, I can connect my 0-10v signal single-ended and would cover the whole range of the input. I could have a signal of 1-11v in a differential setting (As long as I do not exceed 11.9v) and it would give me the same "counts" or raw output as the above configuration.

I think what you are suggesting here is that if you do a differential reading of AIN0-AIN1, where AIN0 is 11V compared to ground and AIN1 is 1V compared to ground, the differential reading will be 10V. That is correct, but there a few things to consider:

• At Gain=x1 you will have Vout=10.0V. That is just within the 10.1V limit from Appendix A, so good.
• Both terminals are within +/-11.5V of ground. That is good according to Appendix A.
• You have Vcm=6.0V and Vout=10.0V. That is good according to Section 2.6.5.

[eka]

Yes, I was thinking as if they where bipolar!

Just to be clear, they are always bipolar, whether differential or single-ended. I think you were originally thinking that differential reading are bipolar and single-ended readings are unipolar.

Think I got it now, I can connect my 0-10v signal single-ended and would cover the whole range of the input. I could have a signal of 1-11v in a differential setting (As long as I do not exceed 11.9v) and it would give me the same "counts" or raw output as the above configuration.

I think what you are suggesting here is that if you do a differential reading of AIN0-AIN1, where AIN0 is 11V compared to ground and AIN1 is 1V compared to ground, the differential reading will be 10V. That is correct, but there a few things to consider:

• At Gain=x1 you will have Vout=10.0V. That is just within the 10.1V limit from Appendix A, so good.
• Both terminals are within +/-11.5V of ground. That is good according to Appendix A.
• You have Vcm=6.0V and Vout=10.0V. That is good according to Section 2.6.5.

Look at TI's REF102. I have been using it as "current source" thru 12K ohm series resistor and 1200 ohm reference for measuring a PT1000 probe's resistance. I power it with U6's 12V terminal.

Attached File(s)

 volt_ref102.pdf ( 583.57K ) Number of downloads: 4

 

[LabJack Support]

I believe eka is suggesting you use Vm+ to get the higher voltage power supply. That is a useful source, but you can't draw much current:

http://labjack.com/support/u6/users-guide/2.11

http://labjack.com/support/u6/users-guide/appendix-a

[Rogerio]

I was looking for at least 5 mA to be on the safe side, so VM+ is not an option. I´m testing a variable DC-DC booster that´s supposed to regulate within 1%, will see how it turns out.
Meanwhile I have a load of software issues to resolve :)

Thank you both for your replies!

[eka]

Sorry about not indicating Vm+ as source of 12V for Ref102.


Ref102 is rated for 10ma, so wall-wart driving it would work.