Oscilloscope recommendation question...

edded

1 channel)

a/s, 2

sa/s, 2

a/s, 2

a

ry

a

time

d 40mhz is

Never heard of these scopes. There are only two brands worth considering Tek or Agilent (HP). You can get a very nice example off ebay for about half what your considering.

Dude, Rigol is the Chinese company that makes the low end scopes for Agilent.

Hi,

Yes, I know. I haven't had much luck with used stuff, so I would prefer to buy a new one. But, I don't want to spend the money that the Tek's/HP's go for.

What is your take on the bandwidth vs the MSO. Given what I am looking to use it for, do you think the 40mhz is more beneficial than the MSO 16 channels LA ?

Thanks,

Alan

I have a TEK TAS 465 100mhz analog with cursers for measuring and digital controls with on screen menue $1,800 new.

Bob

I'd buy a used Tek. Do you have ham or electronics swap meets nearby. Craigslist? Everyone has ebay.

ve

ng

mbedded

, 1 channel)

Msa/s, 2

0Msa/s, 2

Msa/s, 2

s a

tery

tra

l time

hed

s 40mhz is

The low end Aglient scope ARE Rigol scopes (designed and built by them), they are just rebadged as Agilent.

Dave.

Hi Alan

As for bandwidth:

25MHz bandwidth is quite limiting when measuring high freqeency signals that have harmonics, like square waves. e.g. a 25MHz scope is really only good for accurately measuring the signal integrity of say a 5MHz square wave (general rule of thumb of 1/5th the bandwidth). 40MHz is better, but decent ones start at 60MHz these days. But you pay more of course. But as long as you are aware of this, 25 or 40MHz is perfectly fine. Probably not wise to spend extra for going from 25MHz to 40MHz, but 25MHz to 60MHz or 100MHz would be a different story.

Yes, the Velleman is a bit of a toy, spend the money and get yourself a decent scope.

Yes, battery powered scopes are isolated from mains earth, but they are dangerous when used with a normal scope probe. For *safe* high voltage power measurements you really need a professional high voltage isolated differential probe. New ones are expensive, but second hand ones come up on eBay at decent prices.

1MB of sample memory is *massive* for a low cost scope, great choice. Extremely valuable for short and medium term data logging and deep analysis of digital data packets etc.

Yes, ensure you get a sample rate at least 10 times the bandwidth.

Mixed signal scopes are *incredibly* useful, if it was me and I had to choose between the three, I'd spend $200 extra and pick the 25MHz mixed signal scope over the 40MHz non-MSO without question. When you have a mixed signal scope available you'd be surprised at the uses you can find for it. You are spending a decent amount of money here, so mixed signal is a very wise investment for future needs. If you really need extra analog bandwidth you can always get yourself a cheap used 100MHz Tektronix analog job on eBay.

Rigol are decent units for the price, they rebadge them as Agilent so you know the quality and performance is there.

Dave.

Just to clarify the above comment. In general use it's not often you want to measure the accurate signal integrity of the signal under test, so the 1/5th bandwidth rule of thumb doesn't really apply in that case. In most cases it's good enough just to see that a signal is there, so 25MHz is enough bandwidth for most general design and troubleshooting applications.

When you need extra bandwidth, then you usually need a *lot* of it. So going from 25MHz to 40MHz is not a big step.

So either save $200 and go the 25MHz non-MSO, or spend $200 more and get the MSO. Any decent PC based logic analyser is going cost you at least $200 anyway, and you don't get the advantages of having it so easily linked with your analog scope. Money very well spent.

Dave.

Hello Alan,

Other DSOs you may consider: GWinstek and OWON. I would drop the Velleman because of you want to look into transient phenomena. In case of data loging, check whether you can get the raw stored data on your PC (instead of a screen dump), for further processing.

Best regards,

Wim PA3DJS

remove abc from the mail address

Hi,

Thanks for the info Dave and Wim, I appreciate it!

Alan

Who needs a scope at all for embedded processor "stuff?"

Get a datalogger for this...

Yeah right, get a power monitor for this...

They are crap...scopes in general are useless, vast majority of people have no clue how to use them...

Umm, anybody who actually does it?

Oooh, crabbytrolling again.

An oscilloscope is a window into the world, senses in a domain where we have no inherent vision. Without one, all but the most trivial electronics is impossible. A modern digital scope, even a decent sub-$1000 type, is an astonishing quantitative measurement instrument.

If you don't love oscilloscopes, you have no business posting to s.e.d.

I have about 50 of them, the oldest being mid-1950's vintage. Two scopes reside permanently in my office, both Tek: an 11802 20 GHz sampler, and a TDS2012 100 MHz color digital.

John

Not #1. Low bandwidth, low sample rate toys are useless for digital stuff.

Given the budget, I'd go for #4.

The low-end Agilent color scope is made by Rigol. It's $1000 or so, 60 MHz. 1G sample rate. Tek has a similar scope; I prefer the Teks.

John

Have you considered using a PC (computah) based scope instead of a dedicated instrument? In particular, the data logging functions and image captures are best done with something that has a hard disk available.

I don't actually a USB DSO yet, but have borrowed a few to play with. So far, this seems good enough: $170 US One feature that caught my attention was that it decodes some common protocols including 1-wire (which is used in some of the wx stations I help maintain):

It also does SPI and I2C which may be handy for your embedded systems.

Also, for just audio work, I like to use a (free) sound card based oscilloscope, spectrum analyzer, and generator:

--
Jeff Liebermann     jeffl@cruzio.com
150 Felker St #D    http://www.LearnByDestroying.com
Santa Cruz CA 95060 http://802.11junk.com
Skype: JeffLiebermann     AE6KS    831-336-2558

Do you already have an analog scope? Sounds like not. Get one.

A digital scope works best in three situations.

1) where the rep-rate (of the desired event) is too low to keep the phosphor lit up on an analog scope. 2) where you want to view events leading up to the trigger event. 3) where you want to save the data to a log

In many other typical situations, a DSO is a drawback!

Incredibly, there exist DSO's without pretrigger memory. I bought an early Vellemen at a swapmeet. Got it home and discovered that you cannot view the trigger event, much less anything leading up to it. The scope is too light to be a doorstop, so it's useless.

If you expect to log data, make sure you can get the data out. On my Vellemen, you acquire the waveform, set up the serial port on the data logging computer, power cycle the Vellemen and it sends the data out the serial port. Not exactly logging friendly...

Consider your requirements carefully. It's easy to make a big list of measurements you'd like to make and end up with a very costly and versatile machine that is sub-optimal for 99% of the measurements you want to make.

With an analog scope, you stick the probe on the test point and twiddle the scope knobs until you see what you're looking for.

With the DSO, you stick the probe on the test point and wait for the memory to fill up so you get a display. Not what you want? No problem. Shift, menu, vertical, amplitude, up, up, up, up... each time waiting for the memory to fill up before you can see what happened. Hmmmm, looks like it's not triggered. Shift, menu, trigger, channel1, mode, dc...crap, wrong sweep speed...Shift, menu, horizontal, mode, up, up, up, up... You expect it to take a second for the display to refresh...but it doesn't. Should you wait another two seconds? Punch more buttons? Check to see if the probe is still attached? Two seconds doesn't sound like a long time, but it gets very annoying very fast. (yes, the more you pay, the more it can "feel" like an analog scope)

If you know what you're looking for an set the scope up correctly, a DSO is very handy. But if you don't, you often find that the display on the screen does not look anything like the signal. It's not uncommon to have an aliased display that looks sorta like what you're looking for, but the timing is ALL wrong. And depending on how you've set the options, you may be looking at the waveform from the previous test point. It can take a dozen button pushes to get to a mode where you can see what's there so you can set the other parameters so you can get back to what you want to see.

I've been using scopes for 50 years. I'm glad I have a DSO, but I drag it out only when I have to. An analog scope is MUCH more convenient for typical troubleshooting.

Go buy a TEK 465 or equivalent for your first scope. Get a ham radio buddy to put out a request. Most hams have at least two in the attic.

mike

--
Return address is VALID!

This sounds more like badly written software than a drawback on all DSOs. I've been using several different DSOs in the past and I've never experienced the problems you describe.

--
Programmeren in Almere?
E-mail naar nico@nctdevpuntnl (punt=.)

All our Tek digital scopes work and feel just like analog scopes, except that they have color displays, and you can freeze traces, measure stuff with cursors, save waveforms, do FFT's, average, stuff like that, if you want to. They have all the knobs that a typical analog scope has, except "focus." They have "autoset", which tells the scope to automatically set everything to show a waveform; that's handy.

They are also small and light.

The only time I ever use an analog scope is when we have something really strange going on, and we pull out the huge 1 GHz Tek 7104. That's rare.

A 15 MHz Tek scope, a 535 with a single-trace plugin, cost more than a Chevrolet when they were both new. A TDS1012 dual-trace 100 MHz color digital scope costs less than a tenth of the cheapest Chevy you can buy.

John

Regarding usefulness of scopes.

When you know how to use them, they are very useful. When you are aware of aliasing (there are sufficient documents on the net to get familiar with the DSO), DSOs are extremely useful.

I had a GWinstek 2202 (200 MHz, 1 Gs). It is real value for money. I changed to another one (LeCroy 334) because of the inability to zoom into the math trace (like FFT). But that one costs 4 times the GWinstek

When you are working with transient signals (like inrush characteristics of switched power supplies) and fast control loops, you cannot without one. Even for uC circuits they can be very useful.

For all-round time domain analysis, a DSO is a good investment.

Best regards,

Wim PA3DJS

don't forget remove abc from the mail address

I couldn't agree more with John Larkin. The sampling rates on modern DSOs are high enough and memories are deep enough that aliasing problems are relatively rare and easy to understand and overcome when they do happen. Here are a couple of easy ways to do a reality check in a new probing situation:

1) Vary the timebase over several ranges. Events near the trigger should scale appropriately. 2) Use Single Sweep and Dot Display. Unlike analog scopes, you can get a full-brightness sweep from just one trigger event. The 'dots' will indicate just where the scope took each sample. If the dots are far apart relative to waveform timing, aliasing is a possibility.

By the way, analog o'scope displays require judgement to interpret as well. It's not always easy to determine if the various parts of a waveform are from the same sweep. Of course, you can use Single Sweep. However, if you're pushing the upper limit of the scope's writing speed, the results can be so dim that it's really hard to see waveform features. That's why we used to work in dim light and use those scope hoods. I'm glad those days are over.

If I were still doing a lot of low-level analog work, I'd have an analog scope around. However, I find that I can get by with my DSO, even for most low level work, by using BW limiting and trace averaging.

Most of my clients no longer have ANY analog scopes around, and they're all making money, somehow. BTW, those old 465's ARE heavy enough to be a doorstop, but they're really a little too big for that purpose.

Paul Mathews