Open Live Writer publishes tests

"I'm using a _100mhz oscilloscope, including a 100MHz passive probe, and I should be able to measure 90MHz sine waves correctly, right?" Is the oscilloscope or probe broken? ”
I used to be able to answer this question with a little common sense, but now I hear this kind of problem sometimes. Then it may not simply be attributed to common sense, especially in most oscilloscopes, where the oscilloscope's "system bandwidth" or effective bandwidth in conjunction with a particular probe is not described.
Both the oscilloscope and the probe have a bandwidth specification, which is the frequency value of the input signal amplitude attenuation of 3dB. Therefore, if your technical data indicates that the oscilloscope bandwidth is "100MHz", then you can ensure that at least about 70% of the signal amplitude is measured at its bandwidth frequency. The same is true for probes. However, the tricky point is that when you use both the oscilloscope and the probe, your oscilloscope + probe bandwidth, or "system bandwidth," may not be 100MHz. So, what is the system bandwidth in this case?
Before you know the system bandwidth, you need to know the response of the oscilloscope front-end filter. This information may or may not be available in the technical documentation, and if not, please call the oscilloscope Support hotline. If you do not want to contact the support hotline, I would like to introduce you a small technique, the filter response is deduced by the rise Time Technical index calculated in the technical data at the end of this article. However, experience tells us that if your oscilloscope bandwidth is less than 1GHz, then you can think of the filter as "Gaussian". If the oscilloscope has a bandwidth of 1GHz or higher, then its filter may be "maximum flat response (close to brick response)".
In the case of a Gaussian filter, the system bandwidth of the oscilloscope and probe can be calculated using the following formula as a traditional front-end filter type that has been used in analog and digital storage oscilloscopes for decades.

Chinese-English comparison

We use the above example to apply this formula. Since the bandwidth of both the oscilloscope and the probe is 100MHz, your system bandwidth will be 70.7MHz. In other words, your signal amplitude is attenuated by 3dB at 70.7MHz. Obviously, you won't see the full amplitude of the 90MHz sine wave!
In reality, the vast majority of oscilloscope manufacturers have added headroom to the oscilloscope and probe bandwidth specifications. So, if you see "100MHz" on the spec, it's very likely that it's a little more bandwidth, like 110 or 120MHz.
Now, suppose you are using an oscilloscope and probe with a "maximum flat response" filter response. On a 100MHz oscilloscope, a rectangular filter is extremely rare, in this case we only assume. In this case, the "Square and RMS" formula cannot be used. The system bandwidth Calculation formula is:

System bandwidth = minimum value {oscilloscope bandwidth, probe bandwidth}

If I apply this formula to the original example, your system bandwidth is now 100MHz, then you should be able to see almost the entire amplitude of the 90MHz sine wave.
I don't know why most oscilloscopes don't have this simple formula in their technical data. Perhaps most oscilloscopes today have enough bandwidth, and engineers do not have to follow their upper limits. Perhaps this content has been taught at school. However, this is a very useful technique, especially when you see unexpected measurement results.
By the way, it's a quick and easy way to judge whether your oscilloscope is using "Gaussian" or "Maximum flat response" filters. First, find the rise time information for the oscilloscope calculation. The following is an example of a keysightinfiniivision4000x oscilloscope.

Text in the table in English

Now divide "0.35" by the calculated rise time value. Taking the 200MHz oscilloscope (4022A) as an example, the result is: 0.35/1.75ns=200mhz
So, you confirm that the factor used to calculate the rise time is "0.35". 0.35 is the coefficient value of the Gaussian type response filter, so you know that the 200MHz oscilloscope has a Gaussian filter front end. Also, if the same formula is used in the 1GHz oscilloscope (4104A), 0.35/450ps=778mhz
The resulting value is 778MHz instead of 1GHz. Well, you now know that this oscilloscope uses a factor other than "0.35", but "0.45" (0.45/450ps=1ghz). If the coefficients are greater than 0.35, say 0.4, 0.45, or even 0.5, then the filter response of the front-end of the oscilloscope is approximate to the rectangular filter.

Open Live Writer publishes tests