How can I reduce channel time reconfiguration between AI acquisitions? thanks

[User_1979]

Hi everyone,

I need to acquire 40 sin signals (2000 kHz) at a sampling rate of 250kS/s per channel using a 6225 daq card with its AIs. In order to reach this sample rate I have to acquire all of the samples of the first channel (something like 2000 samples), then the same thing with channel 2,... up to channel 40.  My problem is that I am loosing a lot of time (roughly 170 ms) each time I move up one channel because Labview reconfigures the channels.

When I move up one channel to acquire the 2000 samples the only parameter that changes is the channel that's all.

I have tried a few things with the assistant DAQ and Daqmx examples but without any success. 

Is there a way to reduce this paramating time between 2 acquisitions ...?

Any help would be great.

User

[Dennis_Knutson]

That's the time required to stop the existing task and start a new one.

Is your post correct - you are sampling a 2Mhz sine wave at 250 kS/sec?

[User_1979]

Hi Dennis,

Thanks for taking some time to answer. I made a mistake, my Sin signal has a frequency of 2kHz.....you are right

Concerning my question, does this mean there is no solution to avoid this 170 ms reconfiguration using a 6225 PCI card at this frequency?

[Dennis_Knutson · Accepted solution]

If you insist on capturing the channels one by one, you cannot avoid the delay. You can of course, capture all of the channels at once. The effective sample rate would be 6.25 kS/sec if you did all 40 at once.

[LV_Pro]

Is there a reason why are you are sampling at 250 Ks/s? As Dennis points out, sampling at a slower rate would allow you to take a sample from each channel at 6.25 Ks/S. If the maximum signal frequency expected is 2Khz, then the minimum sample rate would be >4Ks/s per Nyquist sampling theorem, so 6.25 Ks/s would be ok.

[User_1979]

Hi Denis and LV,

Thanks for answering.

My purpose is to capture the 40 signals and get the RMS value of each of the channels (using one of the RMS.vis). 

I had already tried capturing the 40 channels at once (infact I only tried capturing one channel at 6.250 Ks/s) but the RMS value isn't correct enough (the value isn't stable). I have attached my VI and saved 10 of the RMS values in the array to show you.

Thanks,

User

[YannC]

Hi,

If you want a better stability, you have to increase the sampling rate.
Can you try to increase the sampling rate?

May be you can do the average of the 10 RMS value for better results.

Regards.

[User_1979]

Hi Yann,

Thanks for your help. 250KHz is fine to build up the signal and use the RMS.vi. As you said, the purpose of the 10 values is to make an average of 10 entire periods for each channel.

Concerning the time loss for reconfiguration the channels, my problems seems to be solved because I was testing my program on a USB 6015 Pad. Changing to a PCI card has brought the time loss from roughly 150 ms to a few ms.... I didin't know there was such a difference of time communication between PCI and USB.

Thanks to everyone for their help.

Regards,

User