SOLVED. Keysight 1000X acquisition bug?

[TK]

I am feeding CH1 (same error with CH2) with a 10MHz 0dBm sinusoidal signal (632mVpp).  When I set the acquisition mode to averaging (does not matter # of averages), when I change the timebase from 100ns to 500us, the signal disappears.  But if I select zoom, the zoomed signal is shown correctly but not the regular signal on top of the screen.  I think it is doing the same with high resolution acquisition mode.  It works fine if acquisition mode is set to normal.  Firmware is 1.10.  Input has 50ohm passthrough.








EDIT: I tested the same setup on my MSOX3024A and 54622D and all of them show the same behavior.  My Micsig TO1104 does not have this issue.

[maxwell3e10]

Probably its a memory depth issue. Keysight EDU-X have only 100kpts, so on 500usec/div x 10 div this corresponds to 50 nsec sampling rate and the 10 MHz signal is at the Nyquist frequency. I've also seen some statements here that the memory length can be even shorter depending on the operating mode and unlike other scopes, it cannot be set by the user.

[TK]

The EDUX1002G was hacked to DSOX1102G and it has 1Mpts.  The MSOX3204A has 4Mpts as well as the 54622D and both show the same issue at the same timebase.

[TK]

Probably its a memory depth issue. Keysight EDU-X have only 100kpts, so on 500usec/div x 10 div this corresponds to 50 nsec sampling rate and the 10 MHz signal is at the Nyquist frequency. I've also seen some statements here that the memory length can be even shorter depending on the operating mode and unlike other scopes, it cannot be set by the user.

You are probably correct, as the issue gets worse at higher frequencies.  What is strange is that at 50uS it shows the attenuated signal, but when zoomed, it shows the correct level (last picture)

[TK]

It looks like Keysight only uses 16Ksamples when doing averaging (unconfirmed): https://community.keysight.com/thread/19412

But still cannot explain why in zoom mode it shows the correct signal...

[MarkL]

But still cannot explain why in zoom mode it shows the correct signal...

The megazoom chipset has two memories.  One is the acquisition memory and the other is the measurement/display memory.

When you capture a signal it all goes into the acquisition memory at whatever rate the scope needs to fill more than 10 divisions, up to the maximum rate.  This rate is calculated automatically and is displayed on the right hand menu bar (at least on the 3000).

The acquisition memory is then processed with averaging and placed into the measurement memory memory which is further processed for the display.  The measurement memory, according to the manual, is at most 62500 points for the 3000, but I think you're right it's 16k by default.

So, each data point in the measurement memory is a result of averaging some number of acquisition points.  As you zoom out you're effectively averaging more points of the sine wave per measurement/display point and the amplitude will approach zero.

In split screen mode it's computing the display twice, once for the top zoom and once for the bottom zoom, but still using the same acquisition record for both.  When you zoom in on the bottom waveform, the scope is doing the same operation as in non-split screen mode and it computes the average using less acquisition points per display point and you get your amplitude and signal detail back.

You can also see this effect with a single-shot capture.

Hope this clears it up a bit.

I generally don't use averaging unless I have a few cycles of a waveform on the screen that I need to clean up.  High resolution mode has similar side effects, but doesn't need a repetitive waveform, so I usually choose that to reduce noise.

[TK]

Thank you MarkL for the excellent explanation.   

[Keysight DanielBogdanoff]

Here's what's happening.

High-res mode does point-to-point averaging. The bin-size for the averages (AKA how long is the averaging period?) is dependent on the time/div setting.

So, if you are at a large time/div setting and the bin>1 period, you'll get an entire cycle averaged together. So, you'll end up with just the DC component of the signal.

Averaging is essentially the same effect. If you're sampling is too small (not sample rate, but the 16k mentioned by TK), you are averaging together more than you intended.

[Keysight DanielBogdanoff]

Also, zoom mode does a re-plot of the original data, so it can look different.