Simultaneous multichannel acquisition

[manos]

Hello everyone!

Due to my recent post about the simultaneous acquisition i'm unable to preform the task. I'm using the DAQ-700 card and LabVIEW 6.1.

Also, i'm aware of the fact that the 700 does not support simultaneous sampling. But, i also read a post where it is said that it is possible to switch between multiple channels over a MUX implemented into the daq700 (i wonder how).

I tried with 4 'AI Sample Channel' and 4 charts, but on all the charts i get the same signal. The signal connected to one of the channels gets to all the charts.
I also tried with 'Intermediate AI', and the same thing happens.

Has anyone an idea, or an .vi ?!

Every help/post is welcome. Thank you.

[TCPlomp]

I don't know the 700, but I would try to ground your other signals, what you most likely see is a residue of the signal in the multiplexer.

Ton

[manos]

Yes, but i need the other channels, i cannot ground them.

I'm measuring V(t), amplitude(y axes) and time(x axes) are both important to me. The thing is, all the channels will not be (let's say) active at the same time - they will not be connected at the same time.

For example:

 c1 is connected: 4V,
 c2 is connected: 5V,
    and the c3 and c4 are not connected - and there comes the problem - the c3 and c4 are then showing 4 or 5V, and that is not actually what i need(that should not happen). I need the     c3 and c4 at this situation in highZ maybe, or something that is VERY different from the standard values i'm measuring (at c1 and c2) - so i can differ connected channels from the          others...

[TCPlomp]

So you measure them, but you're not interested in them.
I don't see what's the problem.

You can (selectively) multiply them with 'NaN' to remove the data

Ton

[manos]

I could multiply the others, but i don't know the exact time when i should multiply. I need different voltage. Here's the thing:

Imagine you have on your right hand a glove with metal contacts. When the measurment starts, you connect your glove(through a wire) to channel 1 with intervals of let's say 2-4s. On the chart there you can read when the hand was connected and when not - different voltages. Connected 5V, not connected highZ or 10V or something.
0 - 1s 5V     connected
1 - 3s 10V   not connected
3 - 4.2s 5V   connecetd... and so on.

BUT, if you had 4 channels(2 hands and 2 legs)...problem. The hand/channel that is not temporary connected shows the uncorrect 5V (that is because of the mux), where it should show 10V...

[TheWoost]

Hi manos,

It seems like you are running into problems because when the channel is not connected it is floating rather than biased.  Is your signal source floating or grounded?  If it is a floating source, I would recommend putting bias resistors on each of the inputs such that when there is no signal connected to them they will be pulled to ground rather than pulled to the level of the previously scanned channel.  If your signal source is grounded, it's not a good idea to do this because you could create a ground loop.  With a grounded signal source, an appropriately sized shunt resistor across the input terminals would be sufficient.  I don't know which terminal configuration you are using, but the following document provides information about the different types of signal sources as well as suggestions on biasing for each:
Field Wiring and Noise Considerations for Analog Signals

Just to provide more information regarding the characteristics of the multiplexed architecture in terms of ghosting, here is a KnowledgeBase article on the subject:
How Do I Eliminate Ghosting From My Measurements?

Regards,
Andrew W
National Instruments

[manos]

Thanks! That's it.

I have very little practice and experience in electronics so... The vi bytheway was correctly programmed all the time.

thanks again