PC based DSO doubts about repetitive mode

[pisoiu]

Hi,
Many commercially available PC based DSOs sample signals up to few hundreds MSPS in real time mode and up to few GSPS in repetitive mode. For real time mode, it should be no problem to aquire data if the banwidth of the analog section is at least up to Nyquist rate. But in repetitive mode, the analog section should respect the same criteria in order to be able to accurately represent a signal. For a few hundred $ scope is not a problem to have a 100MHz bandwidth if the real time sampling rate is 200MSPS, but if the sample rate in repetitive mode is 10GSPS, then a higher bandwidth is required, up to 5GHz to respect the Nyquist criteria, and this seem pretty hard to do at this price. Then what is the purpose of such high rate in repetitive mode? Just commercial bravada? Or just to see the same waveform with the same maximum frequency components acquired with more samples? No matter how high the sample rate is and what tricks are used in order to raise it, the maximum frequency will be bound to the bandwidth of the analog section.

/pisoiu

 

[flatulent]

Re: PC based DSO doubts

If the sampler is not preceded by any amplification and if the scope probe is decent there is no problem with this form of sampling. Several decades ago some scope manufacturers offered these samplers for their then low frequency scopes. As long as the bandwidth of the high speed signal is narrow there is no problem with the down conversion process.

 

[pisoiu]

PC based DSO doubts

Indeed, but I was reffering mostly to current available PC based DSO, like Pico, bitscope, etc. I am sure none of them have direct input into the sampler, they all have amplification. And one more thing, one such limitation can also come from the bandwidth of the sample and hold circuit, which is usually integrated into the ADC.

/pisoiu