For what it's worth...
As Allison has said, with digital circuitry, rise time is a better way to
look at 'scope performance than bandwidth, though, as was pointed out, the
two are related.
From my recent experiences with the 11/45, you will
probably need a 'scope
with a minimum of 100MHz bandwidth - DEC specified the
453, which is a
50MHz unit, but you would have difficulty seeing the sharp "spikes" of noise
on the power rails with that, I could only just see them on the 100MHz
Philips scope I was using.
Probes are very important, and need to be matched both to the circuit under
test, and to the scope input. The scope input bit is easy, it will say
something like "1M in paralell with 10pF", which gives the resistance and
reactance combination (ie, the IMPEDANCE) at the scopes input terminal.
Matching to the circuit under test is more a matter of experience, but
generally, the impedance of the probe should be MUCH greater than the
impedance of the circuit under test, for example, measuring across a load of
1M with a 1M probe will result in a 50% error on the measured voltage. For
this reason, probes are supplied with various muliplication factors, for
example, x1 and x10, these multiply the input impedance of the scope by the
stated amount, but also mean that you have to multiply the reading on the
scope by the same factor to get the correct voltage (some modern scopes byt
the likes of Philips and Tektronix automatically sense when a x10 probe is
connected and alter the Volts/cm display accordingly). Tektronix (Tek), and
some other firms have probes with x1, x10, and x100 factors at small signal
levels, and even x1000 for high voltage work.
Tek scopes are excellent, and the older ones (upto about the 454), are
fairly easy to repair, however, they tend to command higher than average
prices for the spec. You may also want to look at HP, Philips (some of there
newer scopes are reasonably cheap secondhand, and excellent for computer
work - not so good in strong RF environments though), B&K, Telequipment
(became a subsiduary of Tek in later years), EMI / SE Labs (rare outside of
UK / Commonwealth countries), and even Marconi Instruments have made some
good instruments. If you want a rack mount unit, look out for Memcor
scopes - these seem to be a re-badged Tek 647, supplied under a military
contract for navigational beacons, and can be got for very little, as no-one
realises what they are (I got mine for ?30 including shipping in the UK, and
it is a 100MHz unit dual trace unit).
There are a number of good books on basic scope theory and use - mine are at
work, so I'll check the titles on Monday. For some useful info on probe
theory, check this out:
http://bama.edebris.com/manuals/tek/probes/
BAMA also has a lot of test equipment manuals and data sheets that are worth
looking through.
Fluke instruments have just released (at lest in the UK), a CD-ROM which
includes a CBT package on basic scope usage, it is free on application.
If I think of anything else, I'll write some more!
Jim.