DS800C 4Ghz Oscilloscope

[wrieken]

I have searched this site and cannot find any references to DS800C. I am wondering if anyone has any opinion about this oscilloscope and if it is worth purchasing?

Thanks,

William

[elliott]

For what purpose?

[necessaryevil]

It's not really a 4 GHz scope. From the manual: DS800 will work only with repetitive signals, since it requires multiple signal repetitions to complete conversion.
That's cheating! But anway still impressive!

If you don't have an oscilloscope, I would recommend to buy a 'real one' and trade the bandwith for proven quality. No objective reasons though, I don't know everything and I haven't tested the DS800.

I hope someone does a review.

Oh, I assumed it's about this oscilloscope:
http://www.fastsampling.com/Products/DS800E/DS800Manual14.pdf


I was a bit wrong maybe!

But what I ment to say about quality is this:
You can better buy a good 100 MHz oscilloscope than a bad 4 Ghz one...

[T3sl4co1l]

"Cheating"?

Just how many signals will you ever look at that are one-shot at that sample rate, anyway?

It bothers me how many people drink the Kool-aid regarding sample rate versus bandwidth.

Tim

[tggzzz]

It's not really a 4 GHz scope. From the manual: DS800 will work only with repetitive signals, since it requires multiple signal repetitions to complete conversion.
That's cheating! But anway still impressive!

So I suppose you also think HP cheats! From fond memory, a 1GHz 40MS/s scope which was exceedingly useful. http://www.ko4bb.com/manuals/download.php?file=HP_Agilent/HP_54100A_1GHz_Digitizing_Oscilloscope/HP_54100A_Specifications.pdf

If you don't have an oscilloscope, I would recommend to buy a 'real one' and trade the bandwith for proven quality.

Good luck with that if you are interested in assessing at signal integrity - which is a primary use for high bandwidth scopes. You can't tradeoff analogue bandwidth, and sample/second are almost irrelevant.

For many users, the appropriate techniques are to:

• firstly, use an analogue scope to assess the signal integrity, i.e. that the receiver will be able to correctly interpret the analouge signal as the digital signal that was transmitted
• secondly, when signal integrity is assured, debug in the digital domain using an LA or embedded logic, not an analogue scope

[dom0]

It's the usual thing. People tend to not really understand what the specs are about so they focus solely on one or two "key specs" and will rate everything just by those two numbers... completely independent of the significance of that spec to their application.

As a side note it's also surprising how many people don't get the Shannon theorem right or think that digital signal processing is cheating and no real signal processing like analog systems or how many people think that sampling systems can't reproduce analog waveforms correctly and so on. This list is way to long.
I once overheard a mathematics tutor say "Irgendwie merkwürdig wie die alle so Grundlagen gar nicht verstehen" ("Strange how many don't [even] get the basics right"). I think this is true for many sciences, including EE and signals.

[wrieken]

Thank you all for your reply.

As for purpose? We have an experiment using 2.4GHz microwaves (magnetron). We want to look at the waveform shape as it travels through the experiments unique geometrical waveguide to see how it behaves. Our experiment uses pulsed and non-pulsed microwaves, but not at the same time (that is, it is either pulsed or CW). Also, we want to detect any plasma buildup, location and it's rate of buildup in the system as well, should that occur.

Our experiment includes operation in Air, Vacuum, Argon, and Nitrogen atmospheres.

Thanks,

William

[wrieken]

One more important note. We haven't built up this experiment yet. I am just trying to line up all my ducks and get on paper what we need to do the experiment and get the results. What we are looking at at this time is: 1. magnetron operation (stability), 2. beam formation in the wave guide, and 3. plasma characteristic and growth (should it appear), and influence on beam formation in the wave guide.

Thanks,

William

[nctnico]

If you can do many experiments with good repeatability you can use a sampling oscilloscope where each experiment contributes a few samples of the waveform. If not you'll need a realtime oscilloscope. Recently there was a thread about getting a >3GHz realtime oscilloscope and it seems a second hand one can be bought for around $5k.

https://www.eevblog.com/forum/testgear/recommendations-for-a-fast-%28ghz%29-scope/

[T3sl4co1l]

Thank you all for your reply.

As for purpose? We have an experiment using 2.4GHz microwaves (magnetron). We want to look at the waveform shape as it travels through the experiments unique geometrical waveguide to see how it behaves. Our experiment uses pulsed and non-pulsed microwaves, but not at the same time (that is, it is either pulsed or CW). Also, we want to detect any plasma buildup, location and it's rate of buildup in the system as well, should that occur.

Our experiment includes operation in Air, Vacuum, Argon, and Nitrogen atmospheres.

Thanks,

William

Ah, so the concern is entirely single frequency, AC steady state -- as far as the time-domain waveform is concerned.  That is, cycle to cycle, nothing much changes, so you would have no value in an oscilloscope with that capability -- all you'd ever see, on zooming in, is the same old sine wave!

You would be much better served with an RF voltmeter -- optionally a phase sensitive (vector voltmeter) type, and some field probes or power splitters or whatever is desirable for your test.

A much lower bandwidth oscilloscope, and a sufficiently fast amplitude detector, would serve nicely to detect the envelope -- which I'm guessing will be mostly limited by the drive into your magnetron, and the magnetron response itself, rather than group/phase delays in your transmission structures.

Tim