[sdiy] Simple Scope Question

[Magnus Danielson]

From: "Paul Perry" <pfperry at melbpc.org.au>
Subject: Re: [sdiy] Simple Scope Question
Date: Sat, 22 Jan 2005 08:48:05 +1100
Message-ID: <001401c50002$e432db80$0a01a8c0 at frostwave>

> 
> 
> > > [Would] a 15 mHz scope... be suitable for a while?
> >
> This is a thousand or so times better than no scope at all.
> Really. I wouldn't have a problem.
> The troubles people have with scopes is very rarely lack of bandwidth,
> rather it is lack of knowing what to look for, or how to interpret what they
> see.
> I include myself there, of course!

I agree, a 15 MHz scope is much better than no scope at all. For audio/synth,
you almost never would feel the bandwidth of being the limit. For digital
circuit debugging you need more with the modern stuff.

A good beginners scope should have these properties at least:
10-20 MHz bandwidth
2 channels
Alternate/add (many also have chop, but that is rarely very usefull IMHO).
X/Y-mode
Extra trigger input is a plus.
Inversion of signal is a plus.

There is a certain look-and-feel aspect which is hard to translate into text.
When getting used scopes, there is always tear involved, and depending on the
original quality, this may or may not be a major problem.

There are a few things which takes time to master, among those are:

* Trigging - setting the trigging correctly can be crutial. Setting the trigger
  level to get trigg may be the first blunder beginners (and temporary
  brainless expers) fail on. Then, when the trigger mode and trigger level
  cooperate to get trigg, moving the trigg-point to a suitable point at a
  suitable slope direction may take some thought. The trouble is that if you
  place the trigger at a point of the curve with a flat slope, then will noise
  on the signal make the trigger-point move in time, this jitter will make the
  waveform smooth out over the screen becoming fuzzier rather than the sharp
  curve you expect. Some scopes have trigger filters, these can be used to
  have DC or AC trigger and then you can also be fortunate to have a lowpass
  filter to remove any high-frequency noise. Use that, experiment with it, you
  can make much better results when you get it.

* Levels - being aware of the actual levels and impedances can also help. If
  you have a weak-signal, starting by taking it through a 1:10 probe isn't the
  best way to get noiseless signals. Also, think about the loading the probe
  may do. Most places you poke around is really no big problem, but then there
  are some places, where a probe is considered a HEAVY load. What you get on
  the scope is no signal or very little signal.

* Noise - besides the level of signals, be aware that your probe may introduce
  noise and unwanted signal. What amatuers really would kill for when they
  learn about their existence is the diffrential probes. The ground-lead may
  introduce noise, so some people cut the safetyground - bad idea, use a diff-
  probe instead since the common signal (ground-noise) is being compensated
  out. Also, you can watch the diffrential voltages. Very usefull tool. You can
  build simple diffrential probes fairly easy, three op-amps and a handfull of
  resistors.

* Time-base - when in trouble, zoom out and in from the time of your normal
  signal. Sometimes the trouble is deeper in at higher frequencies, but
  sometimes you need to zoom out to see the lower-frequency troubles. I have
  lost count how many times I have been asked to twiddle the knobs on scopes
  by colleagues in need. It can takes only a few minutes of fiddeling with
  trigger, level and time-base to pin-point a trouble. One trick is to trigger
  on the spurious and zoom in on it so you see the period of the unwanted
  signal.

* Two channels - use both channels (or even more if you are lucky to have them)
  to see what happends when, like the input and output of some part of the
  circuit. Sometimes you need to see them at the same time to have them make
  sense.

It's by no means a complete list, but could be a bit of inspiration for
beginners when trying to use scopes. You can do *alot* with 2 channels.
It is worth taking the time to experiment with different methods, ask the guys
here about advice for different problems, etc. Using an oscilloscope is a bit
of an art and you just continue to learn.

Cheers,
Magnus - who always have too little channels, too slow scopes, too much jitter, too little memory, too little resolution, too little good probes...