Using a scope

[Tom Cloud]

>
>
>> Being a mechanical engineer, I have a natural fear of electrons or in
>> other words a little knowledge is dangerous.
>> 
>> If you hook up a scope to something like a injector circuit are you
>> suppposed to use isolator transformers or somekind of isolation circuit
>> to protect the scope?  Are these built into the scope?
>> 
>You don't have to in this case because the car is isolated from the 
>AC mains, however, if you were working on a TV, you would have to 
>use an isolation transformer.

be real careful here .... if you are connecting to a circuit
that needs isolation, you're probably wiser to use the
'scope in *differential* mode.  Using an isolation xfmr
can get you killed -- the 'scope will be at the potential
you're measuring.  I used to do this regularly, but built
a plexiglass enclosure for the 'scope and attached fiberglass
rods to the controls (was measuring a floating 10kV).

 - most 'scopes have 500 or 600 volts max input voltages
(with X10 probe, this *could* be increased by 10X, though
the probe probably is not rated for 5 kV -- check the
data sheet)

 - no isolation is needed if you hook the 'scope probe
ground to a *floating* circuit or to an earth ground.
It's only required if you're hooking the probe ground to
a point that has potential on it referenced to earth ground
(as the scope is connected to earth ground).

 - anytime you *isolate* something, you allow it to *float*.
Isolating a 'scope causes it's case/enclosure to go to
the potential that the probe ground lead is attached to.

Specifically, if you're measuring stuff in your auto -- and
it's not connected to anything in the *real* world (i.e. not
grounded to the AC mains, etc) then you can just hook the
'scope in anywhere you please (note that there will be
*loading* caused by the 'scope).  But, if you've, say, got
the circuit mocked up on your work bench and are using
a power supply that doesn't have floating outputs, then
you must be careful where you connect the probe ground,
as it can cause a short circuit and a burnt probe shield
and/or blown circuit components.  Most power supplies have
floating outputs -- unless you've connected one of the
outputs to earth ground (and I don't recommend that).

All of the above can be avoided if you hook your 'scope
up in differential mode.  Takes two channels.  Put 'scope
in "ADD" mode.  Will probably have to "INVert" one channel.
Then you don't use either of the probe grounds but rather
use each one of the two probes to connect to the two points
you'd ordinarily connect one probe and it's ground to.
The problem with this is it takes two channels and leaves
you with one less to use to measure something else at the
same time!

[written hurriedly .... probably more confusing than if
I hadn't said anything ... sorry  8^)

Tom Cloud <cloud at peaches.ph.utexas.edu>