Oscilloscope ground clip

I'm a bit confused about the ground clip on oscilloscope probes. I have
read that it can cause a ground loop when the measurement circuit is
grounded to the same ground as the oscilloscope itself. Is the ground clip
only to be used when there is an isolated ground then?

I have been taking measurements from 12V car batteries and DC power
supplies. Should the ground clip always be used in these cases?

If I am correct the ground clip should not be used to measure between two
voltages as one would do with a multimeter (referencing a different voltage
in the circuit instead of its ground). It confuses me why I would not be
able to use the ground clip to do that. An explanation/link would be
appreciated.

Thanks for any advice,

David.

it depends also on wheather your ground clip is connected with the
chassis and thus the mains ground of the scope or not.
Expensive scopes and those battery operated are floating, so one can
connect the clip to what ever needed, but cheap scopes have only one
common ground and this might cause the problems.
If the unit under test is not connected to any ground reference common
to the scope, then ground clip is must otherwise you have no reference
for your measurement.
So if you have simple scope and it is connected to mains and its
chassis is grounded via mains ground, then you have to connect the
clip on the car ground in order to measure something.

"David" <(E-Mail Removed)> wrote in message
news:T-(E-Mail Removed)...
> I'm a bit confused about the ground clip on oscilloscope probes. I have
> read that it can cause a ground loop when the measurement circuit is
> grounded to the same ground as the oscilloscope itself. Is the ground
> clip only to be used when there is an isolated ground then?
>
> I have been taking measurements from 12V car batteries and DC power
> supplies. Should the ground clip always be used in these cases?
>
> If I am correct the ground clip should not be used to measure between two
> voltages as one would do with a multimeter (referencing a different
> voltage in the circuit instead of its ground). It confuses me why I would
> not be able to use the ground clip to do that. An explanation/link would
> be appreciated.

Line operated equipment and bench oscilloscopes often lead to ground loops.
A big spark erupts when the ground clip touches the chassis! Has probably
happened to most workers at least once. Often, the circuit breaker trips
and that is the end of the matter (other than a burnt clip end and soiled
underwear). The reason is that the clip is grounded through the 3-wire
power cord on the scope. If the equipment is not line isolated, it is
possible for the chassis to be hot with respect to ground. The short
circuit that results is called a ground loop.

Solutions:

1/ Isolation transformer for the equipment (best) or the scope (shaky).
2/ Use the differential mode on a dual channel scope (be careful about
common mode voltage ratings).
3/ Use a separate differential amplifier before the scope.
4/ Use a battery-operated scope (again, one must observe the scope's voltage
ratings).
5/ Make a floating measurement (use an adaptor to defeat the scope ground
.... this is dangerous but is often done ... please AVOID this method).

Hope that helps.

"Charles" <(E-Mail Removed)> wrote in message
news(E-Mail Removed). ..
>
> "David" <(E-Mail Removed)> wrote in message
> news:T-(E-Mail Removed)...
>> I'm a bit confused about the ground clip on oscilloscope probes. I have
>> read that it can cause a ground loop when the measurement circuit is
>> grounded to the same ground as the oscilloscope itself. Is the ground
>> clip only to be used when there is an isolated ground then?
>>
>> I have been taking measurements from 12V car batteries and DC power
>> supplies. Should the ground clip always be used in these cases?
>>
>> If I am correct the ground clip should not be used to measure between two
>> voltages as one would do with a multimeter (referencing a different
>> voltage in the circuit instead of its ground). It confuses me why I
>> would not be able to use the ground clip to do that. An explanation/link
>> would be appreciated.
>
> Line operated equipment and bench oscilloscopes often lead to ground
> loops. A big spark erupts when the ground clip touches the chassis! Has
> probably happened to most workers at least once. Often, the circuit
> breaker trips and that is the end of the matter (other than a burnt clip
> end and soiled underwear). The reason is that the clip is grounded
> through the 3-wire power cord on the scope. If the equipment is not line
> isolated, it is possible for the chassis to be hot with respect to ground.
> The short circuit that results is called a ground loop.
>
> Solutions:
>
> 1/ Isolation transformer for the equipment (best) or the scope (shaky).
> 2/ Use the differential mode on a dual channel scope (be careful about
> common mode voltage ratings).
> 3/ Use a separate differential amplifier before the scope.
> 4/ Use a battery-operated scope (again, one must observe the scope's
> voltage ratings).
> 5/ Make a floating measurement (use an adaptor to defeat the scope ground
> ... this is dangerous but is often done ... please AVOID this method).
>
> Hope that helps.
>
>

Thanks Charles.

I'm guessing DC power supplies are usually isolated or is it typical for the
negative to be connected to the line ground? I will check this with a
meter.

If the DC supply IS isolated (or if I am using a 12V battery):

Can I connect the ground clip to any point of the circuit and measure the
voltage of another point with respect to the one the ground clip is
connected to (as I would with a multimeter) or is there some reason that the
ground clip should be connected only to a terminal?

I think I'll avoid the ground-defeating adaptors in hopes of having a better
understanding of ground loops some day (I have read about these in the past,
I'm not sure why I'm having such a difficult time getting the concept down).

I will investigate differential mode on a dual-channel scope. I thought
this would involve using two ground clips and two probes and show the
difference between the two probes but I may be way off on this.

Thanks again,

David

>Can I connect the ground clip to any point of the circuit and measure the
>voltage of another point with respect to the one the ground clip is
>connected to (as I would with a multimeter) or is there some reason that the
>ground clip should be connected only to a terminal?
>

Yes , correct


>I will investigate differential mode on a dual-channel scope. I thought
>this would involve using two ground clips and two probes and show the
>difference between the two probes but I may be way off on thi


negative,
differential measurement, should your scope support it, measures the
voltage between two probe tips and is therefore independent of the
ground clips.
However I made the experience, that only higher price class of scopes
do this properly. Also I fond that many scopes have then some kind of
restrictions on bandwith etc on such measurements.

Note: system using two ground clips and two probes is not differential
measurement, but this is adding the two signals together or substracts
them etc, simply ba inversion and so it is definitely not a
differential measurement.

> I'm guessing DC power supplies are usually isolated or is it typical for
> the negative to be connected to the line ground? I will check this with a
> meter.

There are both positive ground and negative ground power supplies ...
negative ground is more popular. Once, voltage doublers (transformerless
circuits) were fairly popular and they were nasty as far as ground loops are
concerned. Today, line transformers (60 Hz) have mostly been eliminated by
switch-mode designs. One must be very careful where one attaches the ground
clip in some of those. Again, an isolation transformer is a nice adjunct
for the troubleshooter's bench.

> If the DC supply IS isolated (or if I am using a 12V battery):
>
> Can I connect the ground clip to any point of the circuit and measure the
> voltage of another point with respect to the one the ground clip is
> connected to (as I would with a multimeter) or is there some reason that
> the ground clip should be connected only to a terminal?

The ground clip can cause loading effects and a ground loop of sorts even
when not connected to the chassis ground. This is an involved subject and I
cannot do it much justice here.

> I think I'll avoid the ground-defeating adaptors in hopes of having a
> better understanding of ground loops some day (I have read about these in
> the past, I'm not sure why I'm having such a difficult time getting the
> concept down).

Picture the power panel in your home or your lab. Typically, the pole pig
transformer outside has a center-tapped winding and the CT is earth
grounded. Inside your structure, the CT is connected to the metal panel.
Two hot circuits are available with respect to ground ... each of which is
120 volts. If you connected a light bulb to either hot side and to ground,
the bulb will light. But, there is also a neutral wire (the white one) used
to carry normal load current. Ideally, the grounds only carry current when
there is a fault. So, if you are standing on a wet floor and touch a hot
wire you can get a shock because the wire is hot with respect to ground.
This is a form of a ground loop and a very dangerous one at that.

> I will investigate differential mode on a dual-channel scope. I thought
> this would involve using two ground clips and two probes and show the
> difference between the two probes but I may be way off on this.

Differential mode uses the two probe tips only.