5 in 1 oscilloscope from ebay, measuring accuracy

Thread Starter

ninjaman

Joined May 18, 2013
324
hello,

i bought one of these recently,
http://www.ebay.co.uk/itm/PC-USB-2CH-20M-Digital-Oscilloscope-Spectrum-Analyzer-DDS-Generator-Sweep-B-/111690731978?hash=item1a014951ca:g:nWgAAOSwYGFUs8Eh

i set the dds to 20Mhz on sine wave and tried to measure it on the oscilloscope and the oscilloscope couldnt measure it, it showed 20Hz instead. i dont know if this is how it would show it.

is there some way to test the accuracy of this device. i dont have any other test equipment and i am new to testing things.

any advice on this would be helpful. i cant afford much so getting an expensive £100 or more for an oscilloscope alone, not including dds and other stuff that comes with this.

thanks

simon
 

Thread Starter

ninjaman

Joined May 18, 2013
324
hello

i am sorry but i dont know how to do that. i have just read some information about nyquist and sampling. i think that to see the 20Mhz sine wave on the oscilloscope i would require an oscilloscope that could sample at least twice this rate. as i understand it the oscilloscope has a bandwidth of 20Mhz, i think that the bandwidth refers to the maximum bandwidth measurable but of a square wave. so the maximum bandwidth the oscilloscope can measure is 20Mhz but square wave. i think that there is a percentage or some level that says what maximum sine wave can be measured. i cant remember but i think 60-70%. so if the bandwidth says 20MHz then the maximum sine wave is around 60-70 percent of that(?)....is this correct at all?

thanks for your reply slk001

simon
 

dl324

Joined Mar 30, 2015
10,984
i think that the bandwidth refers to the maximum bandwidth measurable but of a square wave. so the maximum bandwidth the oscilloscope can measure is 20Mhz but square wave.
Bandwidth refers to sine wave response. For a square wave, BW needs to be 5-7X. Viewing a 20 MHz square wave with a 20 MHz BW scope will give you something more like a sine wave.

Try reducing the signal generator output a frequency several decades below the maximum (e.g. 20KHz) to see if you get the expected results.
 

bertus

Joined Apr 5, 2008
20,568
Hello,

The shown webpage shows me a sampling rate of max 48MS/s.
You will be lucky to see a good 6 Mhz sine on the screen.
On higher frequencies there will be not enough datapoints.

Bertus
 

crutschow

Joined Mar 14, 2008
25,122
As Bertus noted, a 48MS/s sample rate won't resolve high frequency waveforms.
The time between samples is about 21ns so any feature in the waveform that occurs more rapidly then that (such as the rise-time of a fast pulse) will not be resolved.

Also, at slow sweep speeds where the sample rate my be below its maximum, you can get an aliasing effect as SLK001 noted, which can make a high frequency sinewave appear as a low frequency one.
To avoid that is the reason good digital oscilloscopes sample at a much higher rate than twice the highest signal frequency. For example, my Tektronix 60MHz scope samples at 1 GHz/s.

To test for aliasing, increase the sweep speed to see if you can see a higher frequency signal.
 
Last edited:

WBahn

Joined Mar 31, 2012
25,918
To some degree it's because the marginal savings of using a slower sampling rate on the lower bandwidth scopes doesn't offset being able to simply amortize the cost of developing the 1 GS/s sampling circuitry over several families of scopes having different analog bandwidths.
 

Thread Starter

ninjaman

Joined May 18, 2013
324
To some degree it's because the marginal savings of using a slower sampling rate on the lower bandwidth scopes doesn't offset being able to simply amortize the cost of developing the 1 GS/s sampling circuitry over several families of scopes having different analog bandwidths.

funny, i was just going to say that!?!?!?!?!........though i would have no idea what i would have said
 
Top