This is for those who worry about the electrical performance of low bandwidth rated probes like the PP510.
I think I was able to demonstrate already that the PP510 is not different to the PP215 above 100MHz and both probes perform reasonably well up to at least 300MHz:
https://www.eevblog.com/forum/testgear/siglent-sds1104x-e-in-depth-review/msg1434665/#msg1434665So how is this possible?
It has to be mostly marketing, because if you build a passive probe properly, with the correct HF compensation for the scope input it's intended to be used with (or adjustable HF compensation like e.g. the TesTec TT-MF312), then you can expect a probe bandwidth of several hundred MHz and you would have to deliberately add internal filtering to limit the bandwidth to e.g. 100 MHz.
Because of the low actual input impedance, passive "high impedance" probes aren't terribly useful at frequencies that high anyway.
So in fact the probe bandwidth specification only means that the manufacturer doesn't guarantee anything at frequencies higher than that.
A long time ago I performed a test with several probes on the 300MHz Siglent SDS2304X, including the 100 MHz Tektronix P3010, which was the 2nd best after the original Siglent SP2030A! And this on a 300MHz scope which wasn't even the Tektronix for which these probes were originally designed! So a humble 100 MHz probe performed quite good and provided a system bandwidth of close to 400 MHz (450 MHz with the original SP2030A). More importantly, The 100 MHz Tektronix P3010 performed significantly better than a 500MHz Keysight N2843A (on this DSO!).
Now I have explicitly tested the cheap 100 MHz PP510 probe (in x10 mode of course) on a 1 GHz SDS5104X DSO.
Look at the attached images:
SDS5kX_PP510_1kHz_1ns
Signal = 1 kHz, 3 V square, 1ns rise/fall times
Ch.1 = direct 50 ohms coax connection
Ch.4 = PP510, X10 mode, 25 ohms source impedance
SDS5kX_PP510_50MHz_1ns
Signal = 50 MHz, 3 V square, 1ns rise/fall times
Ch.1 = direct 50 ohms coax connection
Ch.4 = PP510, X10 mode, 25 ohms source impedance
SDS5kX_PP510_FR1GHz
Frequency response graph for the PP510 from 1 to 999 MHz. Please ignore the glitch at the start of the frequency range – reference level for this measurement is +1 dBV and the grid is 1 dB/div.
Even the cheap PP510 only comes close to -3 dB at frequencies above 300 MHz and we need to go up to about 630 MHz to fall below that – but only briefly, as we are "back in game" at 750 MHz and reach +1.8 dB at 980 MHz. We also need to take the frequency response of the DSO into account, i.e. the probe-only response would be even higher.
It should be obvious that probes do not have a smooth, predictable low pass characteristic – at least not when connected to the scope. With the 50 MHz square wave we can see that the PP510 transition times are even a tad faster than with direct coax connection; together with the high overshoot, this can be attributed to the positive frequency response (above reference level) at very high frequencies.
What we also see is the fact that transition time measurements with the low 1 kHz frequency are still fairly close because of the deep measurement. I think this is quite handy, even though measurements aren't quite as fast as on some other instruments, which use heavily decimated data for all their measurements.