Author Topic: DAQ for EMG data acquisition  (Read 2015 times)

0 Members and 1 Guest are viewing this topic.

Offline dritechTopic starter

  • Contributor
  • Posts: 17
  • Country: it
DAQ for EMG data acquisition
« on: October 17, 2017, 12:56:51 pm »
Hi all,

I am looking for a DAQ for monitoring EMG signals. The required specifications are the following:

*At least 12CH ADC inputs
*At least 5kHz sampling rate per channel
*At least 14-bit ADC

I was looking at USB-1616FS, but at $1,015 it is a bit too expensive.
Can someone suggest any similar DAQ units which will work for this application.

Also, can someone please explain the following terms/sentences used in the USB-1616FS datasheet:
https://www.mccdaq.com/PDFs/specs/DS-USB-1616FS.PDF

*with sample rates up to 50 kS/s per channel, continuous throughputs of 150 kS/s, and 32 kilosample bursts up to 200 kS/s
*With hardware  paced  mode,  the  maximum throughput rate is 50 kS/s for one channel,  or  9.5  kS/s  per  channel  for all channels. With burst  scan  mode,  the  maximum throughput rate is 50 kS/s per channel for one, two, or three channels, and 16.5 kS/s per channel for all 16 channels.

Am I right in saying 50kS/s is equivalent to 50kHz? What is the difference between paced mode and burst scan mode?

Thanks in advance.
 

Offline JPortici

  • Super Contributor
  • ***
  • Posts: 3461
  • Country: it
Re: DAQ for EMG data acquisition
« Reply #1 on: October 17, 2017, 01:34:48 pm »
Am I right in saying 50kS/s is equivalent to 50kHz?

yes, saying 50 kS/s is equivalent to 50 KHz sample rate.

have you checked out USB DAQs from National Instruments?
http://www.ni.com/en-us/landing/low-cost-daq.html

if you hit "configure" you get options with more channels: http://www.ni.com/en-us/shop/select/multifunction-io-device?modelId=124894
« Last Edit: October 17, 2017, 01:37:08 pm by JPortici »
 

Offline dritechTopic starter

  • Contributor
  • Posts: 17
  • Country: it
Re: DAQ for EMG data acquisition
« Reply #2 on: October 17, 2017, 02:14:01 pm »
Thanks JPortici.

The datasheet of the USB-6211 does not state that the sampling rate is kS/s/ch.

Quote
Sample rate:
*Single channel maximum: 250 kS/s
*Multichannel maximum (aggregate): 250 kS/s

What would be the maximum sampling rate with this DAQ? will it be 250kS/s for all individual 16 channels?
Also, the USB-6211 is not a simultaneous sampling DAQ (unlike the USB-1616FS). Will this somehow make any difference when monitoring and gathering EMG signals?
 

Offline MasterT

  • Frequent Contributor
  • **
  • Posts: 785
  • Country: ca
Re: DAQ for EMG data acquisition
« Reply #3 on: October 18, 2017, 12:39:36 am »
I think, the data rate is limited by USB "bottle neck", regular PC may get 1-2 Mbits/sec over usb channel, dividing by 16-bits it boils down to 125 k sample per second. And after this limit, you have an option:
a) stream 2-3 channels only with max sampling rate;
b) store data in the memory buffer in burst mode from all 12-16 channels, than stop sampling and slowly transfer data to PC;
c) decrease sampling rate down to < 10 ksps per channel.

 Arduino DUE (20$ on aliexpress) is able to sample 16 channels, it has one ADC 1 Msps, though capable to function as DAQ with 64 ksps for all 16 inputs. But problem with usb would be similar. If you are familiar with DSP, algorithm-math I mean, I 'd advise to delegate some raw processing to arduino itself, doing some kind of compression, than transfer pure data to PC in lower volume.
 

Offline dritechTopic starter

  • Contributor
  • Posts: 17
  • Country: it
Re: DAQ for EMG data acquisition
« Reply #4 on: October 18, 2017, 07:49:31 am »
Thanks for your feedback.
The specifications of the USB-6211 does not mention what is the maximum sampling rate for all channels. Infect, it states that the maximum sampling rate is 250kS/s (for multichannel). Is there something misleading, or am I interpreting the datasheet incorrectly?

Datasheet: http://www.ni.com/pdf/manuals/375195c.pdf

Also, is the maximum speed of USB 2.0 1-2Mbits/sec? I was of the impression that it is much more than that.
« Last Edit: October 18, 2017, 07:56:10 am by dritech »
 

Offline rodpp

  • Frequent Contributor
  • **
  • Posts: 307
Re: DAQ for EMG data acquisition
« Reply #5 on: October 18, 2017, 03:02:07 pm »
I understand that the sum of the sample rates of all used channels must be limited to 250KS/s, probably because it uses one ADC to all channels.

As your requirement is 5KS/s per channel and 12 channels, it is ok.

Note that you will have 16 single ended channels or only 8 differential channels that provide a cleaner signal in many cases.

It is a good idea to contact a NI representative and ask for assistance, maybe there is another board that suits better your requirements, like having 32 AI channels to use 16 differential if necessary.
 

Offline Karel

  • Super Contributor
  • ***
  • Posts: 2217
  • Country: 00
Re: DAQ for EMG data acquisition
« Reply #6 on: October 18, 2017, 04:03:50 pm »
I am looking for a DAQ for monitoring EMG signals. The required specifications are the following:

*At least 12CH ADC inputs
*At least 5kHz sampling rate per channel
*At least 14-bit ADC

I don't see a specification for electrical isolation & patient safety. How are you going to deal with that?
 

Offline nctnico

  • Super Contributor
  • ***
  • Posts: 26906
  • Country: nl
    • NCT Developments
Re: DAQ for EMG data acquisition
« Reply #7 on: October 18, 2017, 04:16:45 pm »
I strongly recommend to get a DAQ which is designed for EMG use. This has the right safety precautions and standard connectors. One of the problems is that the signals need a lot more amplification than a standard DAQ will be able to do.
There are small lies, big lies and then there is what is on the screen of your oscilloscope.
 

Offline dritechTopic starter

  • Contributor
  • Posts: 17
  • Country: it
Re: DAQ for EMG data acquisition
« Reply #8 on: October 18, 2017, 06:14:55 pm »
Hi all,

Something that I did not mention is that some of the signal conditioning (isolation, amplification etc)will be performd prior monitoring with the DAQ.

@rodpp, the USB-1616FS one A/D converter per channel, but I am not sure about the NI USB-6211

EDIT: The NI USB-6211 has one A/D for all 16 inputs.
« Last Edit: October 18, 2017, 07:00:56 pm by dritech »
 


Share me

Digg  Facebook  SlashDot  Delicious  Technorati  Twitter  Google  Yahoo
Smf