| Products > Test Equipment |
| How much noise floor and other things matter in oscilloscope usability |
| << < (18/63) > >> |
| Fungus:
--- Quote from: G0HZU on December 25, 2021, 07:30:59 pm ---If it helps, I can switch the scope to 1Meg input and attach a 50R load and it looks pretty much the same. There might be a tiny bit more noise but any change is barely perceptible. --- End quote --- There's no point. Everybody here knows (and freely admits) that lower noise oscilloscopes exist. They make for pretty screenshots and youtube videos. The questions is: How much advantage does it give you in real life? For digital signals? None at all. For that you need bandwidth and high sample rates which is where the MSO5000 shines. For periodic signals in the mV range? I suspect the answer is "not much if you use averaging", hence me asking if anybody can post a picture of a Rigol MSO5000 showing power supply ripple with 1x probe, 20MHz limiter and waveform averaging. (Prove me wrong!) For non-periodic signals in the mV range? You'd have an advantage there but they're few and far between and you're probably better off looking for them in FFT mode than trying to trigger on them and view them as a trace. nb. For any signal in the mV range you can add a signal amplifier. They sell 30dB amplifiers on aliexpress for not much money. |
| Kleinstein:
The x1 probe is not always an option: it is slow (e.g. 5-10 MHz BW) and quite some load (e.g. 100 pF range). With a x1 proble the probe will limit the BW, but it still makes sense to enable the 20 MHz limit to reduce amplifier noise, as there is essentially no signal > 20 MHz anyway. With a non periodic signal the FFT is not an option. It would not give much (if any) useful information. Averaging only works well if one has a good signal to trigger from - so if there is only a small / noisy signal this will not help. Noise in the trigger signal can smoothen out the signal with averaging. With the rigol scope still at hand, one could measure the noise, to see if it is really much worse, or just looking higher noise with higher sampling rate. A point to compare would be with a short (input to GND), a time scale to get a comparable sampling rate (e.g. 1 Gs/s and thus BW limited by the sampling rate to 250 MHz) with only 1 channel active and than a high gain (e.g. 1 mV/div or 5 mV/div before the probe setting). Some digital signals like LVDS are not that large: 400 mV at the input would be 40 mV after a 10:1 probe and thus may want 5 mV/div sensitivity. |
| Kleinstein:
--- Quote from: Fungus on December 25, 2021, 06:11:48 pm --- --- Quote from: G0HZU on December 25, 2021, 05:43:59 pm ---Here's my old HP Infinium scope set to 1mV/div with the 30MHz bandwidth limit enabled. The Rigol scope was showing about 1mVpkpk on a 20MHz bandwidth setting which seems really noisy in comparison to my HP scope from the 1990s. --- End quote --- What's the bandwidth/sample rate of that? The Rigol has 350MHz pathways and is sampling at 8GHz which inherently produces noise, no way around it. Does your HP have waveform averaging mode? At least it looks like your HP can zoom out. --- End quote --- The picture show 10 Ms/s. So there is anditional BW limit (~ 5 MHz) there. To do a fair comparison one would have the switch the faster scope also a slower hirizontal rate to get the lower sampling rate. |
| Fungus:
--- Quote from: Kleinstein on December 25, 2021, 08:34:41 pm ---Averaging only works well if one has a good signal to trigger from - so if there is only a small / noisy signal this will not help. --- End quote --- If the ripple is too small to trigger from then the power supply is probably OK and there's nothing to worry about. |
| oz2cpu:
my siglent, looks like same signal as yours fungus clearly alot less noise, but this is only a 2GS scope :-) |
| Navigation |
| Message Index |
| Next page |
| Previous page |