Author Topic: Oscilloscope for high sample rate, short duration data recording  (Read 1909 times)

0 Members and 1 Guest are viewing this topic.

Offline csacelarisTopic starter

  • Newbie
  • Posts: 5
  • Country: us
Hello,

I'm going to be running an experiment where I need to record two voltage outputs at a 20MHz sample rate for a duration of under 1 second and I'm trying to learn what oscilloscopes might be appropriate for this.  I found these two:

OWON XDS3204E
SIGLENT SDS2204X

Both of which (I believe) have the memory capacity and triggering ability that would be required but I was wondering if anyone knew of any other lower cost oscilloscopes that might work in this application? I have looked into DAQs as well and found the MCC USB-2020 which looks like it would work but the versatility of an oscilloscope makes it much more attractive. Also if anyone has any general recommended learning resources for high speed low duration data sampling feel free to point me that way. I searched around but didn't find anything similar to this here, if there is and I missed it I apologize. Thanks for your time.

Cheers,
Chris
 

Online Fungus

  • Super Contributor
  • ***
  • Posts: 16562
  • Country: 00
Re: Oscilloscope for high sample rate, short duration data recording
« Reply #1 on: October 18, 2021, 06:23:25 pm »
What's the signal to be measured? How many volts?

The cheapest 'scope I can think of with enough memory to record two channels at 20Mhz for 1 second is the Rigol MSO5000 series but they're not famous for low signal levels.

If half a second is enough then the price will drop dramatically. Almost any entry level 'scope will do it.
 

Offline csacelarisTopic starter

  • Newbie
  • Posts: 5
  • Country: us
Re: Oscilloscope for high sample rate, short duration data recording
« Reply #2 on: October 18, 2021, 08:42:09 pm »
The sensors are going to be fed through a charge amplifier which has yet to be determined (have to see if we can find something we have that will work or if we'll need to purchase something) but it will probably output around +/-10V (guesstimate based off of the few amplifiers I've seen). A half second would be more than enough. Is there an entry level scope that can trigger off of an external input (likely have to interface this with an existing labview program)? If I could get something like a Siglent SDS1104X-E to work that would be amazing but didn't think they had triggering features like that.
 

Offline tautech

  • Super Contributor
  • ***
  • Posts: 28141
  • Country: nz
  • Taupaki Technologies Ltd. Siglent Distributor NZ.
    • Taupaki Technologies Ltd.
Re: Oscilloscope for high sample rate, short duration data recording
« Reply #3 on: October 18, 2021, 09:14:39 pm »
The sensors are going to be fed through a charge amplifier which has yet to be determined (have to see if we can find something we have that will work or if we'll need to purchase something) but it will probably output around +/-10V (guesstimate based off of the few amplifiers I've seen). A half second would be more than enough. Is there an entry level scope that can trigger off of an external input (likely have to interface this with an existing labview program)? If I could get something like a Siglent SDS1104X-E to work that would be amazing but didn't think they had triggering features like that.
Yes, typically with the entry level DSO's one just uses a spare channel as the external input and assigns the trigger to it and for a defined period of recording/logging uses a Single trigger and a timebase setting to have the intended record period fill the display.
Total memory size for the 4ch X-E's can be managed buy placing a channel on each ADC however using a 3rd for external triggering will halve the memory depth so it may be better to trigger from one of the channels of interest with an appropriate trigger type.
The new Logging feature in these scopes could certainly be useful for your needs so be sure to consult the new manual to check it out.
Quick mention here and a link to the App note:
https://www.eevblog.com/forum/testgear/siglent-sds1204x-e-released-for-domestic-markets-in-china/msg3689545/#msg3689545
Avid Rabid Hobbyist
Siglent Youtube channel: https://www.youtube.com/@SiglentVideo/videos
 

Offline rodpp

  • Frequent Contributor
  • **
  • Posts: 307
Re: Oscilloscope for high sample rate, short duration data recording
« Reply #4 on: October 18, 2021, 10:36:59 pm »
A dedicated DAQ device is the best tool for this job.

An scope shines in much other areas compared with a DAQ card, but it is not the best tool for data acquisition experiments.

A DAQ card will be much more flexible to program and integrate in a experiment system. It allows continuous data acquisition, and much better resolution. An scope, if I'm not wrong, is limited to 8 bits only. DAQ cards can be found with 12, 14, 18 bits.

Maybe an scope is sufficient for this experiment, but can limit future ones.

A high performance DAQ card can be much more expensive than an scope, but there are various options that may be good for your needs and not break the bank.

A established brand in this field is National Instruments. If you have no objection, maybe you can find good used options on ebay. 
 

Offline csacelarisTopic starter

  • Newbie
  • Posts: 5
  • Country: us
Re: Oscilloscope for high sample rate, short duration data recording
« Reply #5 on: October 18, 2021, 11:48:37 pm »
rodup

Certainly a DAQ (like the one I linked to) is more focused built for this sort of task. The main interest in going the oscilloscope route is the ability to use it for other things down the line beyond what a DAQ could do.

Resolution/bitrate does matter and the DAQ I had found can do 12 bits vs the 8 bits an entry level oscilloscope can do, so that is something to think about.

New NI stuff is expensive (NI-9775 is $3k just for the module) and buying used isn't really an option.

tautech

I hadn't thought of that but that makes perfect sense. I'll check out the link.



« Last Edit: October 18, 2021, 11:51:41 pm by csacelaris »
 

Offline bdunham7

  • Super Contributor
  • ***
  • Posts: 7729
  • Country: us
Re: Oscilloscope for high sample rate, short duration data recording
« Reply #6 on: October 18, 2021, 11:51:01 pm »
The sensors are going to be fed through a charge amplifier which has yet to be determined (have to see if we can find something we have that will work or if we'll need to purchase something) but it will probably output around +/-10V (guesstimate based off of the few amplifiers I've seen). A half second would be more than enough. Is there an entry level scope that can trigger off of an external input (likely have to interface this with an existing labview program)? If I could get something like a Siglent SDS1104X-E to work that would be amazing but didn't think they had triggering features like that.

External trigger is not found on many lower level scopes because it is quite a bit of extra circuitry that isn't needed all that often.  In your case, it looks like it would come in really handy.  If you really mean 20MSa/s and not a 20MHz bandwidth signal, then the SDS1104X-E just comes up short.  You'd have to use one of the channels as your trigger and with 3 channels active, you have 7 Mpts memory depth.  At 20ms/div, you have 280ms @ 25MSa/s or at 50ms/div you get 700ms @ 10MSa/s.  The SDS 2202X-E has the same total memory as the SDS1104X-E, but the external trigger will allow you to use it properly and get a full 700ms @ 20MSa/s on each channel.

However, going up one class of scope in whatever brand may get you a lot more memory, so unless this is a one-and-done project, maybe consider that.  The SDS2104X-E, for example, would allow you a full second at 10X your required rate, which means you could use a hi-res or ERES mode to improve on the 8 bits, if that is important.

 
« Last Edit: October 18, 2021, 11:53:39 pm by bdunham7 »
A 3.5 digit 4.5 digit 5 digit 5.5 digit 6.5 digit 7.5 digit DMM is good enough for most people.
 

Offline csacelarisTopic starter

  • Newbie
  • Posts: 5
  • Country: us
Re: Oscilloscope for high sample rate, short duration data recording
« Reply #7 on: October 19, 2021, 12:07:28 am »
bdunham7

Yes, I meant 20MSa/s and thanks for the explanation. This likely won't be used as a one off so that is definitely something I'll consider.
 

Offline Caliaxy

  • Frequent Contributor
  • **
  • Posts: 278
  • Country: us
Re: Oscilloscope for high sample rate, short duration data recording
« Reply #8 on: October 19, 2021, 01:17:06 am »
Have you considered Picoscope? They have a nice range of USB instruments with various sample rates, number of channels, amounts of memory and resolution. They come with dedicated oscilloscope software (for general use as PC oscilloscopes) and Labview drivers (vi’s) so you can integrate them easily in your own Labview data acquisition/processing vi. Some of them have external trigger input.
 
The following users thanked this post: egonotto

Offline rvalente

  • Frequent Contributor
  • **
  • Posts: 722
  • Country: br
Re: Oscilloscope for high sample rate, short duration data recording
« Reply #9 on: October 19, 2021, 02:35:26 am »
Did you check hantek usb6000 series? They may solve the issue.
 

Offline tautech

  • Super Contributor
  • ***
  • Posts: 28141
  • Country: nz
  • Taupaki Technologies Ltd. Siglent Distributor NZ.
    • Taupaki Technologies Ltd.
Re: Oscilloscope for high sample rate, short duration data recording
« Reply #10 on: October 19, 2021, 02:38:23 am »
The sensors are going to be fed through a charge amplifier which has yet to be determined (have to see if we can find something we have that will work or if we'll need to purchase something) but it will probably output around +/-10V (guesstimate based off of the few amplifiers I've seen). A half second would be more than enough. Is there an entry level scope that can trigger off of an external input (likely have to interface this with an existing labview program)? If I could get something like a Siglent SDS1104X-E to work that would be amazing but didn't think they had triggering features like that.

External trigger is not found on many lower level scopes because it is quite a bit of extra circuitry that isn't needed all that often.  In your case, it looks like it would come in really handy.  If you really mean 20MSa/s and not a 20MHz bandwidth signal, then the SDS1104X-E just comes up short.  You'd have to use one of the channels as your trigger and with 3 channels active, you have 7 Mpts memory depth.  At 20ms/div, you have 280ms @ 25MSa/s or at 50ms/div you get 700ms @ 10MSa/s.  The SDS 2202X-E has the same total memory as the SDS1104X-E, but the external trigger will allow you to use it properly and get a full 700ms @ 20MSa/s on each channel.
Correct, in this case an Ext trigger can sure be useful so to not halve the sampling and mem depth.
This is what it would look like but done with a SDS1104X-E instead:


Quote
However, going up one class of scope in whatever brand may get you a lot more memory, so unless this is a one-and-done project, maybe consider that.  The SDS2104X-E, for example, would allow you a full second at 10X your required rate, which means you could use a hi-res or ERES mode to improve on the 8 bits, if that is important.
No such model when instead I believe you meant SDS2104X Plus which indeed with its 200 Mpt mem depth also offering better sampling and a 10 bit mode (100 MHz limited).

I'll have one out later to PD and install options for a customer so will grab some quick screenshots as examples to show mem depth and sampling rates for timebases suitable for ~1s captures. These certainly have an Ext trigger therefore sampling rates and mem depth are not impacted like if needing to use a 3rd channel.


Second SDS1104X-E screenshot @ 5ms/div @ max mem depth for 70ms of capture record.
Avid Rabid Hobbyist
Siglent Youtube channel: https://www.youtube.com/@SiglentVideo/videos
 

Online Fungus

  • Super Contributor
  • ***
  • Posts: 16562
  • Country: 00
Re: Oscilloscope for high sample rate, short duration data recording
« Reply #11 on: October 19, 2021, 09:58:43 am »
If I could get something like a Siglent SDS1104X-E to work that would be amazing but didn't think they had triggering features like that.

You can use one of the channels for trigger but it takes away sample memory because you have to enable that channel.

The two-channel models of Rigol/Siglent usually have an external trigger input, as indicated by the 'E' in "Rigol DS1202Z-E" (for example).

 

Offline tautech

  • Super Contributor
  • ***
  • Posts: 28141
  • Country: nz
  • Taupaki Technologies Ltd. Siglent Distributor NZ.
    • Taupaki Technologies Ltd.
Re: Oscilloscope for high sample rate, short duration data recording
« Reply #12 on: October 19, 2021, 11:19:50 am »
Quick screenshot from SDS2104X Plus with a logging graph open and mem depth, sampling rate and timebase visible.
The graph can be reset or played and auto timescales as a record is allowed to build.
Open inputs and full sampling rate displays some noise that of course can be lessened with the 20 MHz BW limiter.
Avid Rabid Hobbyist
Siglent Youtube channel: https://www.youtube.com/@SiglentVideo/videos
 

Offline egonotto

  • Frequent Contributor
  • **
  • Posts: 650
Re: Oscilloscope for high sample rate, short duration data recording
« Reply #13 on: October 19, 2021, 04:50:36 pm »
  I found these two:

OWON XDS3204E
SIGLENT SDS2204X


Hello,

than perhaps a OWON XDS3062A with 12 Bits is better.

Best regards
egonotto
 

Offline csacelarisTopic starter

  • Newbie
  • Posts: 5
  • Country: us
Re: Oscilloscope for high sample rate, short duration data recording
« Reply #14 on: October 19, 2021, 07:30:24 pm »
Cailaxy

I believe I had looked at their DAQs but not their scopes. They seem highly capable for the pricepoint compared to some of the other options of course at the cost of requiring a computer. Not sure if it would be any cheaper to go that route but definitely something I'll look into.

rvalente

I had not, I'll look into that.

Thank you all for the info, got a good direction to go in now. I'll report back once the instrumentation is under way for anyone else interested.
 

Online Fungus

  • Super Contributor
  • ***
  • Posts: 16562
  • Country: 00
Re: Oscilloscope for high sample rate, short duration data recording
« Reply #15 on: October 19, 2021, 07:43:11 pm »
Quick screenshot from SDS2104X Plus with a logging graph open and mem depth, sampling rate and timebase visible.

Sure, but why buy a $1500 'scope when a $250 'scope will do the job?

Rigol DS1102Z-E has 24 Mpts of memory. That's 12 Mpts per channel when both channels are enabled so it can record 0.5 seconds at the required sampling rate.  It has a separate input for triggering.

https://www.tequipment.net/Rigol/DS1102Z-E/Digital-Oscilloscopes/

Given that the signal is 10 volts (I asked!), the next step up would be Rigol MSO5074 with 100 Mpts of memory. If you use two channels for the signal you get 50Mpoints per channel so it can record over two seconds of data at the requested sample rate. You can use one of the other channels as trigger input (it doesn't need to be enabled/displayed on screen to work as trigger source).

https://www.tequipment.net/Rigol/MSO5074/Digital-Oscilloscopes/

It can also be upgraded to double the memory so you could record over four seconds of data. You can buy the official memory upgrade option (still cheaper than the Siglent) or visit the hacking thread to find out how to unlock it for free.

nb. EEVBLOG members get a discount at TEquipment. I'm not allowed to post the code, you have to search for the thread.

« Last Edit: October 19, 2021, 07:55:53 pm by Fungus »
 


Share me

Digg  Facebook  SlashDot  Delicious  Technorati  Twitter  Google  Yahoo
Smf