Thoughts about new o-scopes

My regular scope is a Rigol DS1052E, which is very nice and costs about $400. I can't see any reason to use an analog scope any more. All the traces are the same color!

We have an Intronix USB logic analyzer that works fine, something like $400. The only complaint is that bus pins must be connected in a certain order, and that forces the PCB layout for Mictor connectors into that order, which makes PCB layout a little more difficult.

There's no reason to spend tens of kilobucks on basic instruments any more.

John

Is that the one with the Dave Jones tuning option? :-)

You can also get a big old used one if you have the shelf space. I have a Dolch LAM-3250 which is probably around 30 years old now. They tend to have more sturdy connectors, clips and other miscellanea. The enclosure in baby-blue is more like late 60's but oh well, works nicely.

True. I am trying to talk a client out of spending kilobucks right now. Because we need that dough for something more urgent. All they really need is something like the 4ch GDS-2204 I have here and that's under $2k.

--
Regards, Joerg

http://www.analogconsultants.com/

Modern DSOs in the price range you mentioned have pretty much solved the whole "basic oscilloscope" problem, in terms of replacing analog scopes entirely. Not interested in starting flame wars on the subject, I've owned more analog scopes than most folks, and I probably still do. :) I picked up an Agilent MSO 6054A a couple of years ago when I started getting serious about FPGA work. The 6000s are great, and they're cheaper now that the new 3000X models are replacing them. You could probably get a new/nearly-new 1 GHz DSO/MSO6104A in the $6K- $10K price range you mentioned.

That said, I've only used the MSO feature a couple of times, and I could've lived without it. I can usually get enough information about what's going on just hooking up the four analog probes to various ports.

In any case Tek still sells good scopes but they no longer have any inherent advantages over anyone else. If you're looking for something brand new I'd check out the DSO/MSO 3000X models first, then the Teks, and go with whatever feels 'right' to you. My guess is that Agilent will come out on top, at the moment. Rohde & Schwarz is also pitching a new line of portable scopes that look really nice.

-- john, KE5FX

I'm guessing that you ribbon-cable it (forgive the verbization) instead of using the flying leads? Mictor... yup. Too fancy for me. ;-)

--
Rich Webb     Norfolk, VA

We have an adapter that lets up plug the USB LA pod directly onto a Mictor. We sometimes add a Mictor to a PCB layout where we expect an especially hairy situation, like for example a PCI Express bridge chip connected to an FPGA. We don't load the Mictor on production boards, once things are stable.

John

"John Larkin"

** My regular scope is a BWD821, dual trace 50MHz with a nice PDA, internal graticule tube.

Been using daily since it I bought it new since 1985 - and I have a spare CRT for it.

A Rigol DS1052E bought 18 months ago sits on a shelf.

Why?

Being a DSO means it is always likely to giving misleading info on the screen and it is VERY tedious to use for any job the BWD can do with ease.

.... Phil

There are some reviews of the Agilent and Rigol scopes here. There is a hack noted there for one of the Rigol models that doubles (?) the sampling freq, also. Dave

Ooops. Forgot the link:

Dave

thanks for the link! Last time I seriously used a scope Rigol was only a few years old. Seems like they took prior Tek functionality and glossed it up a bit and cut the price. I have a basic dislike for companies (Chinese especially) that copy something and cheapen it in the process. Linfan is a prime example - they took both Honda and B&S low HP engines, copied them exactly, and are dumping them for $100 each, compliments of Harbor Freight. Anyway, I digress.... It severly bothers me that Agilent rebadged and sold low end Rigol scopes. The new Agilent 2000 X series looks interesting but I am looking at little higher end scopes (300-500 MHZ BW) and not entry level units.

I

Thanks for your reply...I am not sure what the 'basic oscilloscope problem' is.....unless you are referring to the Nyquist Criteria for signal sampling and reconstruction, which I am very well aware of...While I am a bit biased towards analog scopes, and some of the inherent design issues with signal sampling, I am also well aware that analog front ends can only be so responsive. One can 'miss' signal artifacts with low BW scopes and also with high sampling rate DSOs. Which makes me ask the question that I have not seen answered: Exactly what criteria is used to specify BW of a DSO?

Thanks for the pointer to Agilent 3000x...they look very worthy of consideration.

-John

I worked with the Intronix logicport. It works but the memory is really small so it runs out of breath real quick. Also the probing is not suitable for high speed. I have a Tektronix TLA700 series logic analyzer which is very affordable from Ebay and it offers a lot more.

--
Failure does not prove something is impossible, failure simply
indicates you are not using the right tools...
nico@nctdevpuntnl (punt=.)
--------------------------------------------------------------

Yes, even in "time-stamp mode" (compression) it fills up pretty quickly if there's a fast clock on one of the sampled lines.

For my normal stuff, it's the bee's knees (always wanted to use that phrase) but it is pretty "old" in silicon years. It amazes me that they haven't re-spun it with modern components and an order of magnitude (or more) memory, etc.

--
Rich Webb     Norfolk, VA

I guess it isn't worth the effort. It will only make the product more expensive. The only real problem it has is the limited memory and occasional lockups. Decoding protocols like SPI and I2C is pretty cool.

I have used Agilent's 54600 series MSO but its just like a scope with a lot of channels. I always missed typical logic analyzer features like complex triggering. I think modern MSO's will also have protocol decoding feature and more logic analyzer like features.

--
Failure does not prove something is impossible, failure simply
indicates you are not using the right tools...
nico@nctdevpuntnl (punt=.)
--------------------------------------------------------------

The problem of really, truly replacing analog scopes. Contrary to what the scope vendors would have liked everyone to believe, this is just now becoming practical. Some of the best of the old-school analog oscilloscopes like the Tek 2467B models could trigger at rates of 100,000 acquisitions per second or more with minimal dead time between sweeps. The garden-variety Rigol DSOs are good for few couple thousand at most, as are the older-generation Tektronix scopes. The current Agilent models can approach a million per second, and both Agilent and Tek do a good enough job at display intensity grading that you can get the same visual information from their DSOs that you could see on a CRT.

Aliasing is the second half of that problem -- look at an AM signal on most DSOs, for instance. You'll find it impossible to distinguish the envelope and the carrier, and it may not even be obvious what timebase settings to use. Agilent in particular deals with aliasing by dithering the sampling clock, and their strategy works much better than what I've seen in the Tektronix offerings (disclaimer: the last Tek model I used was a 2000-era TDS 3034, so they're probably better now).

Same as with an analog scope, the 3 dB point (70.7% peak amplitude) unless the vendor specifies otherwise. As with most sampled systems, the front end of a DSO is band-limited with an LPF.

I played with one at the ESC show a few months ago and thought it looked pretty nice, but didn't spend enough time with it to make any sort of judgement call vis-a-vis my 6000-series model. It'd be worthwhile to look at both, if you can. The LCDs are very different

-- the 6000s actually give you more pixels, but a smaller screen. The new ones run Windows CE, the 6000s run something else (VxWorks?) Check out Dave Jones's EEVBlog 'teardown' videos, too.

-- john, KE5FX

I'll check it out..tyvm. Yes, they run VX works...a vastly superior RT kernel than Windoz CE. yes, I am very biased...developing hard RT embedded systems (hw and sw) for critical systems is my forte...and I would NEVER choose a MS product. But that is another story... Hard to imagin Agilent leapfrogging the competition with a super FPGA design, only to shoot itself in the foot by using CE..oh well... Best regards, John

A lot of people here have given you ideas for scopes how ever, I think which ever one you get, should have a cup holder! :)

Jamie