Using external clock on USB-205

[Atanas]

Hello,

I've purchased USB-205 daq, which I plan to use for acquiring IQ data. I'd like to sample the signals synchronously, applying external ADC clock and trigger.

I tried to capture some data using DAQami (eval version).

It appeared, however, that the SW doesn't provide an option for using the clock provided on the pacer input terminal AICKI .

Could you advise, please, whether my observation is correct and whether options for circumventing that are available? (In the application, I plan to use a python script for capturing and processing the IQ data and hope, that that it won't be an issue.)

I have also a question concerning the synchronicity of sampling when using 2 analog inputs.

The attached screenshot shows one signal, sampled simultaneously by AI1 and AI0. It appears, that the signal on AI1 is delayed with 1 ADC clock cycle (SR=250kHz).  I wonder whether the delay is determined by device's HW design, or can be prevented by applying a proper configuration.

Thank you in advance.

Regards,

Atanas

[Jeffrey]

Hello @Atanas

You are correct, DAQami does not include the external clock option.

The device is capable of using AICKI for external clock but as stated, that feature is not included in DAQami.  it can be implemented using the Universal Library in Python (or any other supported IDE).  See example ULAI12, it features ExtClock mode (external clock).

The USB-205 has only one A/D converter, so when you set up a scan for 1 channel, you get 1 sample for each external clock pulse.

but when you scan more than 1 channel, each external clock pulse will acquire one sample from one channel, but with each clock pulse, the multiplexer will index through your channels.  in short, you need 2 clock pulses to sample from 2 channels.

This is how the USB-205 operates. 

Other devices include 1 A/D per channel such as the USB-1808, USB-1608FS-PLUS, or the USB-2020.  Many other devices have a scan option called BURSTMODE where for each external clock pulse provided, the device scans through ALL the channels you've selected in the AInScan function. This minimizes skew time.

But as stated,  the USB-205 does not include either of these options.

[Atanas]

Thanks for the prompt replay, Jeffrey!

I think, that the limitations posed by the USB-205 design are almost clear now.

I still hope, that there could be a way to use the device for my purpose. I could capture the I and the Q data in 2 consecutive scans (those waveforms are constant/unmodulated and synchronous with the trigger). A necessary condition would be, that the delay between the trigger and the first sample is constant and known. Could you advise on that, please?

Another question concerns the actually applied sample rate. I noticed, that sometimes it slightly differs from the entered value. Is there a way to prevent that in  DAQami? 

Thank you in advance.

Regards,

Atanas

  

[Jeffrey]

Scan set up time is between 20 mS and 80 mS.  the setup time is PC dependent, depending on how long it takes to setup an interrupt, etc.  I don't have better hard data than that.

you request a sample rate and the device driver tells you what it can really do.  This is due to the divider circuit built into the USB-205.  Most requested values it can achieve, but sometimes it can only get close.  DAQami is just reporting back what the driver tells it.

[Atanas]

Thanks Jeffrey.

I use external trigger on USB-250's TRIG input, and hope that the PC shouldn't be involved, since I first start the scan via   DAQami's control and only afterwards I enable the trigger via an external circuitry.  Thus, my question concerns the delay between the TRIG signal rising edge and the actual start of the scan.

WRT the sample rate, I understand that there are certain limitation with the generation of the requested rate. I expect that when external clock is provided to AICKI input, that problem won't appear. Correct?

Thanks.

Atanas

[Jeffrey]

Per the User's manual, https://www.mccdaq.com/pdfs/manuals/USB-205.pdf, page 16, Trigger latency spec states: 1 µs + 1 pacer clock cycle max

15 minutes ago, Atanas said:

I expect that when external clock is provided to AICKI input, that problem won't appear. Correct?

Correct.

[Atanas]

Thanks Jeffrey.