want to buy a handheld oscilloscope

Yes but typically those are current probes. That's appropriate (and necessary) for some applications but not all.

Need for what?

Repeat this three times every morning: "FIRST the requirements and THEN the design."

There are some small, low frequency 'scopes that are kind of graphical add-ons to multimeters. Something like

Then there are ones like this from Tek

and I'm sure there are others that reach up in both price and capability.

Does it NEED to be hand-held or is it OK if it's just portable and battery powered? How about small and portable but runs off mains power?

FIRST the requirements and THEN the design.

--
Rich Webb   Norfolk, VA

No, if it is measuring voltage, then it has to pierce the insulation to get a contact. If you 're looking at current, then the probe is a current transformer (rather expensive, too...).

If you are sure that you want one, and if the specs are good enough (since I do not know what you wish to do with it). You might want to try the Velleman-10 model at

Go to test equipment, and it should be there, if they still have them. Roughly

159.00 USD plus shipping.

It is digital, it is hand held, it is single channel, and 5MHz at the most bandwidth.

Since I don't know what you consider to be a decent price... Can't say.

I will say that handheld/portable tends to be more expensive than a bench top scope.

Harvey

Hi people,

Sorry for my silly question ... I thaught if i say too much it would make it even more obvious how little i actually know about this stuff :) But here goes, i'll try and explain the kind of thing i want todo ..

What i want to do is see a reading of signals from things like optical sensors and digital outputs from various things .. I could do this with a multimeter if the signal was there for any decent amount of time .. But as i *think*, an oscilloscope can help me here to 'slow down' time and see what is going on ?

So im interested in getting a little graphical graph displaying voltage over time, for use in low current digital circuits. 0v-12v max.

For example;

An optical sensor mounted on something and a wheel with a hole in it, spinning infront in such a way that on only ONE part of the wheel where the hole is the optical sensor would be triggered (or untriggered, whichever) .. how can i SEE that the sensor did infact work ...

The way i accomplish this kind of stuff at the moment is to incorporate the sensor into a circuit (amplification where needed) and program a PIC chip to wait for the 'high' or 'low' (whatever the case might be) and light an LED for say 100ms, so i can see a flash .. a funny way todo things :) very difficult to find problems when they occur ...

So really it does not need to sample anything faster than a PIC chip would operate, as that is pertty much all i work with. Of course PIC chips range in their speed, depending on how fast you run them, but i run them at 4 or 8 mhz tops.

I am looking at getting a "Velleman HPS10"

, this has a 10mhz processor built in, does this mean i can read samples upto 10 mhz ? so this would suite my needs of operating with PIC chips at 4-8 mhz ?

For this kind of usage, are the clamps useless ? If not, then can i use any brand of 'clamps' on this HPS10 unit ?

Thanks for your help! :)

Alex.

Or, rather, think of an o'scope as a really fast voltmeter.

While I *know* that others will disagree, I'd recommend that you bite the bullet and get a decent digital scope rather than spend $19.95 on a third-hand, 30-year old analog scope with dodgy electronics.

Recommendation: a Tek TDS1002 or TDS1012.

It uses "real time" digital sampling. That allows you to trigger on and capture a single event rather than relying on the repaint from a continuous series of waveforms. Very handy when working with digital electronics to be able to capture one particular wavetrain, e.g., one character on a serial data line or one transition on your optical sensor. It's nice to "know" that an event happened but it may also be useful to know exactly how wide that one pulse was and what the rising and falling edges look like.

You need to look at two numbers; the sample rate is one but the analog bandwidth is equally important. That defines the characteristics of the front-end electronics. At 2 MHz, this is at the limit of an 8 MHz PIC (four clocks per instruction cycle, so an output port can't be toggled faster than 2 MHz, IIRC).

A rule of thumb is that you'll want a bandwidth that's at least 5X the target signal frequency, so at least a 10 MHz bandwidth would be preferred. That is *not* the same as the 10 MHz sample rate (which is how fast the internal electronics sample the incoming signal). See also: Nyquist.

The Vellman is only a single channel. Sometimes that's OK but often you'll want to see what's happening HERE when something changes THERE and how long after THIS that THAT happened.

No, but clamps measure current whereas standard probes measure voltage. Each has uses, but a current probe won't tell you the state of a logic port as well as a the boring old standard probes will. Ignore the perceived coolness factor of clamps; virtually every o'scope is supplied with probes and not clamps because most of the time, probes are the appropriate sensor.

--
Rich Webb   Norfolk, VA

If digital, it can work to stop it completely. One shot mode.

The maximum data rate that a 10 MHz sampling rate scope can tolerate is 5 MHz, due to something called the Nyquist Theorem. I have one, and it's good to about 2-3 Mhz in terms of what you can really see. You must have at least one sample for a waveform's period to be able to reproduce it, that's one down and one up...

The frequency of the PIC is the clock frequency, which you might or might not be able to see... but the events that it generates, if at least a microsecond pulse, will be quite visible.

Any "clamp" that manages to match a 1 megohm 35pf (a standard) input impedance. The Velleman comes with a 10x probe. It will run from 5 nicad batteries, does not have a backlight, has 2 waveform memories.

It is also available for 159.00 + 6.00 shipping from

it's in the test equipment section, so let that determine your maximum price.

Harvey

Thanks guys! this cleared things up a lot ...

I will go with the cheap HPS10 for now, although the point about multiple samples at the same time (would be great for clocked serial data lines like SPI) is a good one .. i will consider that.

Atleast this unit will get me started, im sure i will soon want more features soon enough though .. but i think if i get an expensive feature packed one straight off that all these features will overcomplicate things perhaps ...

thanks again for all your help! :),

Alex.

Also bear in mind that, unless things have improved a lot recently, clamps are designed for high current work. So they're really for measuring currents of many amps, whereas digital logic is likely to be at microamps or lower. You're not going to be able to measure that without a lot of added noise messing up your signal.

The main reason for clamps is to prevent you having to interrupt the supply to insert an ammeter (you wouldn't want to rewire your electric cooker just to measure its current consumption, for example). But voltage measurement (usually) doesn't require breaking the circuit anyway, so this defeats the main advantage of the clamp.

Theo

--
Theo Markettos                 theo@markettos.org.uk
Liphook, Hampshire, UK         theom@chiark.greenend.org.uk
                               http://www.markettos.org.uk/

Theo Markettos wrote in news:ROu* snipped-for-privacy@news.chiark.greenend.org.uk:

TEK used to make a couple of clamp-on AC current probes for small signals and high BW,and a DC-50Mhz current probe that measured small currents,AM503/A6302,for the TM500 mainframe,now available,IIRC,from TEGAM.

--
Jim Yanik,NRA member
jyanik@kua.net

I have another question regarding the HPS10 now, i was reading some things about video signals, and generating them from PIC controllers;

im just wondering if this oscilliscope will be fast enough to capture the scan lines of a PAL video signal ? 64us long.

Alex.

640 samples at 10 megasamples/second.

I'm not too sure how it will look on the color signals, but it will capture the line.

Harvey

Hi Harvey,

It just arrived a couple of hours after posting my message! :). After unpacking it, i plugged it to the first signal i could find, a color video camera (PAL) and went through the booklet learning its functions.

At the same time i was reading a webpage describing what a PAL video signal looks like, so i could compare what readings i got, and what im supposed to get/look for.

I could get a clear picture of the scan lines, one after the other, all exactly 64us long(4us-sync,8us-color burst, 52us-signal) but i could not get the vertical sync's.

I played around with triggering a little bit, assuming i just couldnt see them because it was all too fast, still, i cant find them. What method should i be using to find such a portion of a signal ? is there a particular technique ?considering there is a very repetetive signal inbetween (ie: the scan lines).

Other than that, having never really used an Oscilloscope before, i can definantly say this one is easy to use and understand.

Thanks again,

Alex.