@med6753: With the first version of that bloody manual out for review, I did return to that matter.
Here is what I found so far:
My last post was insofar not precise as the mention of 'single shot' was instead meant to refer to single aquisition, that is all modes which do not use the 'repetitive' function.
I'm sorry that I can not include the missing parts, that is the table from page 3 of the catalogue excerpt and the block diagram from the maintenance manual, but those are PDFs which I obtained either from K07BB or from the TekWiki and they are protected, so that I can't print or snapshot from them. One day hopefully someone will explain to me how this can be resolved.
Six places in the user manual refer to the phenomenon/function discussed:
My take on that is that there is a bandwidth 'bottleneck' in the scope and that the spec. of 150MHz is only fulfilled by a mode which works akin to the sampling verticals in the 7xxx series. That bottleneck seems to be the analogue CCD bucket brigade delay line.
But that mode would only work correctly if the part between input jack and sample gate (the equivalent of the sampler head in the abovementioned combination) does indeed have a BW approaching that, or the signal would be distorted. So the statement 'Full Bandwidth = 150 MHz analogue BW' applies not exactly to the 'analogue' part, but only to the 'unsampled' part, that is the vertical frontend. Because the signal going through the CCD is still analogue in amplitude but discontinuous in time. Specifically, it seems to be the part where the signal is entered into the CCD or the transport through it and not the readout from it.
But that did not yet make sense to me.
First, I would have not have expected that a BW limit in this part does manifest itself, as you did show, like a BW/risetime limit in a vertical amp. I would have expected that it distorts the signal more.
Second, you state:
It's not running in single shot mode. It's free running. Which makes it even more puzzling. As soon as you go beyond say 1MHz you need to turn repetition ON or else you start to get the roll off as originally shown. This did seem to low for me first. When you see that they equipped that scope optionally with a video function and also that they show video signals in the manual. As the video trigger function would select a specific line, the aquisition and display of its content with the necessary fidelity would mean that a single aquisition must be sufficient for a signal with around 5MHz.
But on a second thought it occured to me that we are talking sine signal when BW is specified. You stated that that signal is around 10MHz. And the signal which you did record looks to me like a signal for which the odd higher harmonics are limited to the 3rd 5th and 7th, with the latter already stunted. And yes, the signal with repetitive mode on looks like the components up to the 15th could be present - filling the corners and making the top reasonably flat.. That would have been 30MHz, 50MHz and a bit of 70MHz, so for a stated BW of 40 MHz it would be rather respectable and be in line with the conservative specifications that Tektronix is known for.
Somewhere in the doc. I did see data where they elaborated what you can expect in the time domain. I have to look at that again and think it through. Maybe we have become spoiled with all those Gs/s at our disposal nowadays. It correlates with the answers I got when I suspected my Analogic 2045 of being unwell. Someone pointed out that 800Ms/s WILL mean that only very simple 100MHz waveforms can be generated, and they will already show!
Also, I would like to fire my 2430a up and compare the result.
Further, you made me want to have get 2440 instead as backup/portable scope.
Thank you very much.
And next, I have to do a risk analysis. Yechh.