50 Ω Termination and probingI sometimes see discussions about 50 Ω termination at the scope inputs. Some seem to believe that scopes with switchable 50 Ω inputs are just a luxury, that can easily be replaced by an external through-terminator or a BNC-T with end terminator.
Even though I’ve often commented on this very topic already, I want to repeat it here with especially the SDS800X HD in mind.
The commonly agreed industry standard for the maximum acceptable input VSWR is 1.5:1, which is equivalent to 14 dB return loss. This tolerates 4 % power reflection with its associated amplitude errors. It goes without saying that oscilloscopes with properly designed switchable 50 Ω inputs, like the SDS2000X series and anything above from Siglent, provide a VSWR ≤1.5:1 up to and beyond their full bandwidth.
Now let’s have a look at all these emergency solutions, like external through-terminators and BNC-T with end termination, which all have a perfect VSWR of 1:1 up to 250 MHz when not connected to anything.
First let’s connect the ancient hp10100C through-terminator to a real scope input, like the 1 MΩ // 17 pF in case of the SDS824X HD:
hp10100C_SDS824X HD_VSWR_73M
The VSWR reaches 1.5:1 at 73 MHz. Beyond that, we’ll have to face substantial linear distortion.
Here’s another test with the through-terminator RS-456-150-50Ω, which is the Tyco part B35 X13 • 999 X99, connected to the SDS824X HD:
RS-456-150-50Ω_SDS824X HD_VSWR_78M
The VSWR reaches 1.5:1 at 78 MHz, hence is a tad better than the venerable hp10100C.
Finally, the BNC-T with 50 Ω end terminator connected to the SDS824X HD:
BNC-T50_SDS824X HD_VSWR_78M
The VSWR reaches 1.5:1 at 78 MHz, hence its effect is the same as the RS-456-150-50Ω.
Just for orientation, here’s the VSWR-plot up to 3 GHz for the hp10100C not connected to anything:
hp10100C_VSWR_1157M
The VSWR reaches 1.5:1 at 1157 MHz, hence would be useless for higher bandwidth oscilloscopes. By contrast, here’s the VSWR-plot up to 1.2 GHz for the SDS2504X HD and its 50 Ω input path:
SDS2504X HD_V200mV_VSWR_1130M
Here the VSWR reaches 1.5:1 at 1130 MHz, which is very similar to the hp through-terminator alone. Now, at the latest, it should become clear that 50 Ω input mode is more than just a resistor switched across the input connector. I do not have any SDS1000X HD here, but would not be surprised if its VSWR was similar to the SDS2000X HD.
An oscilloscope without 50 Ω inputs is fine as long as you stick to the supplied probes, but circuit loading might become a problem at relatively low frequencies like 20 MHz already. Passive x10 “high impedance” probes with e.g. 10 pF tip capacitance represent a 200 Ω load impedance at 80 MHz after all. On the other hand, it can only perform up to some 80 MHz without noticeable linear distortion when using direct coax connection.
Conclusion:Whenever you plan to do serious work at higher frequencies, then you should consider an oscilloscope with proper 50 Ω inputs. In order to avoid expensive active probes, you should better design appropriate 50 Ω test points into your prototype circuit.