What is the proper way of setting up an oscilloscope?

[MRW]

Hello there!

I have an Instek GDS-2202 digital storage oscilloscope. I am trying to
figure out how to setup my scope to minimize the pick up of stray 60Hz
mains signal. If I change my scope's vertical sensitivity to 20mV/div,
then I can see the 60Hz signal. The signal's amplitude increases even
more when I turn on my DC power supply. I can pick this signal up
without even connecting the probes to anything.

Should I connect the oscilloscope to another outlet instead of putting
it in the same surge protector? How do I minimize the 60Hz pickup?

Also, I have a programmable signal generator that is pretty accurate. I
put out a 1 kHz sine wave test signal with a low amplitude (about 200
mV), and it seems like I have some display issues. On a Tektronix
scope, this signal looks pretty stable. But on my GDS-2202 the signal
drifts a lot, so instead of seeing one sine wave, I see a non-inverted
and inverted sine wave on top of each other.

I am new to this and would like to learn from you about the proper way
of setting up instrumentation equipment.


Thanks!

 

[Andrew Holme]

Hello there!

I have an Instek GDS-2202 digital storage oscilloscope. I am trying to
figure out how to setup my scope to minimize the pick up of stray 60Hz
mains signal. If I change my scope's vertical sensitivity to 20mV/div,
then I can see the 60Hz signal. The signal's amplitude increases even
more when I turn on my DC power supply. I can pick this signal up
without even connecting the probes to anything.

Should I connect the oscilloscope to another outlet instead of putting
it in the same surge protector? How do I minimize the 60Hz pickup?

Also, I have a programmable signal generator that is pretty accurate. I
put out a 1 kHz sine wave test signal with a low amplitude (about 200
mV), and it seems like I have some display issues. On a Tektronix
scope, this signal looks pretty stable. But on my GDS-2202 the signal
drifts a lot, so instead of seeing one sine wave, I see a non-inverted
and inverted sine wave on top of each other.

I am new to this and would like to learn from you about the proper way
of setting up instrumentation equipment.


Thanks!

It is normal to see mains 60 Hz pickup when the probes are not connected to
anything. It should go away if you touch the probe tip to the earth clip,
and also when the probe and the earth are connected to a signal source.

Use the triggering to get a stable display.

 

[Chris]

Hello there!

I have an Instek GDS-2202 digital storage oscilloscope. I am trying to
figure out how to setup my scope to minimize the pick up of stray 60Hz
mains signal. If I change my scope's vertical sensitivity to 20mV/div,
then I can see the 60Hz signal. The signal's amplitude increases even
more when I turn on my DC power supply. I can pick this signal up
without even connecting the probes to anything.

Should I connect the oscilloscope to another outlet instead of putting
it in the same surge protector? How do I minimize the 60Hz pickup?

Also, I have a programmable signal generator that is pretty accurate. I
put out a 1 kHz sine wave test signal with a low amplitude (about 200
mV), and it seems like I have some display issues. On a Tektronix
scope, this signal looks pretty stable. But on my GDS-2202 the signal
drifts a lot, so instead of seeing one sine wave, I see a non-inverted
and inverted sine wave on top of each other.

I am new to this and would like to learn from you about the proper way
of setting up instrumentation equipment.


Thanks!

Hi, Whatever. First off, it might have helped if you had mentioned how
you're making your scope connection to the board you're measuring. If
you're using a separate wire from the banana jack on the front or back
of the scope panel to the power supply (or Frith forbid, you're just
using a BNC-to-alligator clip thingie), you're almost guaranteeing this
is going to happen. Use the ground clip on the real scope probe. If
it's broken or missing, buy a new one. If you don't have one, get one.
Make sure it's rated for the frequency of the scope.

Try triggering off of the calibration signal on the scope (if it has
one). I'll bet you don't have this problem with that signal.
Actually, most gross ground noise problems of this type are due to the
scope probe, plain and simple. You might want to try borrowing a known
good probe, and trying it again. They don't last forever.

Good luck
Chris

 

[MRW]

Thanks all!

First off, it might have helped if you had mentioned how
you're making your scope connection to the board you're measuring.

Well, to start, I have the scope and the DC power supply plugged in to
the same surge protector. My circuit ground is the negative terminal of
one of the DC supply outputs. I connected the ground pin of the oscope
probe to this negative terminal.

When I noticed the 60Hz signal, the scope probe was not connected to
anything.

Thanks!

 

[default]

Hello there!

I have an Instek GDS-2202 digital storage oscilloscope. I am trying to
figure out how to setup my scope to minimize the pick up of stray 60Hz
mains signal. If I change my scope's vertical sensitivity to 20mV/div,
then I can see the 60Hz signal. The signal's amplitude increases even
more when I turn on my DC power supply. I can pick this signal up
without even connecting the probes to anything.

Should I connect the oscilloscope to another outlet instead of putting
it in the same surge protector? How do I minimize the 60Hz pickup?

Also, I have a programmable signal generator that is pretty accurate. I
put out a 1 kHz sine wave test signal with a low amplitude (about 200
mV), and it seems like I have some display issues. On a Tektronix
scope, this signal looks pretty stable. But on my GDS-2202 the signal
drifts a lot, so instead of seeing one sine wave, I see a non-inverted
and inverted sine wave on top of each other.

I am new to this and would like to learn from you about the proper way
of setting up instrumentation equipment.


Thanks!

GG poster

How is the scope grounded? Signal ground and Power grounds.

Unless the ripple really is present on the signal - you likely as not
forgot to ground the scope to your PS or have a ground loop - unlikely
with both on the same outlet.

 

[default]

Thanks all!


Well, to start, I have the scope and the DC power supply plugged in to
the same surge protector. My circuit ground is the negative terminal of
one of the DC supply outputs. I connected the ground pin of the oscope
probe to this negative terminal.

When I noticed the 60Hz signal, the scope probe was not connected to
anything.

Thanks!

At the 20 mv sensitivity you might see some ripple on all but the best
power supplies,

with the scope probe floating - it has a 10 megohm input impedance so
you should expect to see what is in the air - 60 hz is all around.
Put a 10 k resistor between the probe and the ground on the probe and
it should quiet it out to a flat line.

 

[MRW]

Hello!

How is the scope grounded? Signal ground and Power grounds.

Well, the outlet plug for both oscope and power supply are three
prongs, and both of them are plugged in the same surge protector.... so
I guess the power ground for both oscope and power supply are the same.

The signal ground is referenced to the power supply negative terminal,
so this is separate from the power ground. I hope I understood your
question correctly.

 

[Phil Allison]

** Groper alert !

I have an Instek GDS-2202 digital storage oscilloscope

** Too damn lazy to supply a link ??

http://exphil.com/pdf/Instek/GDS-2202.pdf

I am trying to
figure out how to setup my scope to minimize the pick up of stray 60Hz
mains signal. If I change my scope's vertical sensitivity to 20mV/div,
then I can see the 60Hz signal. The signal's amplitude increases even
more when I turn on my DC power supply.

** Got the scope sitting on top of the PSU ??

I can pick this signal up
without even connecting the probes to anything.

** The probe tip and its ground clip must be shorted to see nothing.

Should I connect the oscilloscope to another outlet instead of putting
it in the same surge protector?

** Sure that surge protector has an earth connection that works ??

How do I minimize the 60Hz pickup?

** Use standard earthing and shielding practice & avoid ground loops .

Also, I have a programmable signal generator that is pretty accurate. I
put out a 1 kHz sine wave test signal with a low amplitude (about 200
mV), and it seems like I have some display issues. On a Tektronix
scope, this signal looks pretty stable. But on my GDS-2202 the signal
drifts a lot, so instead of seeing one sine wave, I see a non-inverted
and inverted sine wave on top of each other.

** Noise in the triggering will do that.

Try using the external synch input on the scope linked to an output on the
generator that has a fixed level.

Use good co-axial BNC leads.

Maybe de-earth the generator - to do this safely, connect the AC terminals
of a 20 amp bridge rectifier ( with + and - linked) in series with its earth
wire.

This isolates small hum loop voltages while allowing any dangerous ones to
conduct to ground.



........ Phil

 

[Chris]

Thanks all!


Well, to start, I have the scope and the DC power supply plugged in to
the same surge protector. My circuit ground is the negative terminal of
one of the DC supply outputs. I connected the ground pin of the oscope
probe to this negative terminal.

When I noticed the 60Hz signal, the scope probe was not connected to
anything.

Thanks!

Hi, Whatever. That's the real source of your problem -- the external
wire GND connection to the power supply. Use the ground clip wire on
your scope probe to connect to the local GND at the point you're
measuring, and you'll be happy about how much the ground noise is
improved. Again, if your ground clip wire is missing or broken,
replace it. If you aren't using a real scope probe, get one which is
matched to your scope. It's the cost of getting accurate measurements.

To answer the question you're worried about, a working scope has the
ability to cancel out the millivolts of difference between local ground
for the scope and local ground for whatever's being measured.
Typically, you won't have to worry about the device under test and the
measurement device using the same power strip, although it certainly
can't hurt and is a good idea on principle.

By the way, for safety purposes it's recommended that the scope remain
grounded.

Good luck
Chris

 

[default]

The signal ground is referenced to the power supply negative terminal,
so this is separate from the power ground. I hope I understood your
question correctly.

Sounds like you're doing it right. The scope probe ground is usually
the better choice than using the ground on the scope chassis. (more a
concern when high frequencies are involved).

Bench (testing) supplies leave the power supply plus and minus
floating with respect to ground so a ground loop is unlikely on that
type of supply - switching supplies may ground the output, so a ground
loop is possible.

You mention seeing the ripple on a 20 mv scope setting. That is quite
sensitive, but on a good linear supply I'd expect the regulated output
to have 1% or less noise on it - and most of that might be due to
measuring technique.

The ripple you are concerned with? Is it sinusoidal or half/full wave
pulsing? - a supply falling out of regulation due to poor filtering or
a bad diode will show a half or full wave (16/8 millisecond humps) not
a sine wave.

Steady display with the sync set to "line"

 

[MRW]

** Too damn lazy to supply a link ??
Oops.


** Got the scope sitting on top of the PSU ??

Nope. The scope and the PSU are away from each other.

** Sure that surge protector has an earth connection that works ??

I don't know. I always thought that the surge protector had an MOV near
the mains cord, so that it shorts out any surge to ground and isolate
any of the devices connected to the surge protector. Right?

** Use standard earthing and shielding practice & avoid ground loops .

I'm not familiar with this. I have a bit of introduction to ground
loops after reading some PCB layout tips.

** Noise in the triggering will do that.

Will I be able to minimise this noise?

Thanks!

 

[MRW]

The ripple you are concerned with? Is it sinusoidal or half/full wave
pulsing? - a supply falling out of regulation due to poor filtering or
a bad diode will show a half or full wave (16/8 millisecond humps) not
a sine wave.

It's sinusoidal.

Thanks!

 

[Chris]

Nope. The scope and the PSU are away from each other.



I don't know. I always thought that the surge protector had an MOV near
the mains cord, so that it shorts out any surge to ground and isolate
any of the devices connected to the surge protector. Right?


I'm not familiar with this. I have a bit of introduction to ground
loops after reading some PCB layout tips.



Will I be able to minimise this noise?

Thanks!

Standard shielding practice is to minimize the length of unshielded
wire. By using the scope probe ground clip instead of using a separate
wire, you'll reduce this length to a couple of inches instead of a
meter or more. This is basic stuff.

New scope probes are fairly inexpensive, especially if you go with
another vendor besides the original scope manufacturer. You should
look at the scope front panel to determine input capacitance, and get a
replacement probe with that compensation range. Also, get a X1/X10
probe that's rated for a higher frequency than your scope -- equal
doesn't cut it.

One other thing -- this is one of the rare circumstances where buying
used is a mistake. Get a new probe and take care of it -- even
cheapies last a long time if you're careful with them.

Good luck
Chris

 

[MRW]

One other thing -- this is one of the rare circumstances where buying
used is a mistake. Get a new probe and take care of it -- even
cheapies last a long time if you're careful with them.

The scope is fairly new and came with the manufacturer's probes.

Thanks!

 

[default]

It's sinusoidal.

Thanks!

Another thing to check is that the scope shield is really connected.
Check from the chassis out to the ground on the probe itself and make
sure there's (zero ohms) continuity.

Most scopes will be zero ohms (unless you have the rare bird - true
differential inputs)

My own scope probes frequently have a poor connection where the ground
connects to the probe at the test end. (small spring clips) I can use
the chassis ground with no problems - but then I cut off the ground
pin that was on the power cord. Something of a safety hazard.

Of no use to you, but an interesting tidbit: The little ring close to
the tip on scope probes is a ground - and is used to minimize the
length of wire when the 4" from the normal ground is too long
(checking noise on switching supplies is one application).

Grounding practice for instruments isn't 100% the same as circuit
boards - but they can both be solved the same way - single point
grounding.

 

[redbelly]

Hello there!

I have an Instek GDS-2202 digital storage oscilloscope. I am trying to
figure out how to setup my scope to minimize the pick up of stray 60Hz
mains signal. If I change my scope's vertical sensitivity to 20mV/div,
then I can see the 60Hz signal. The signal's amplitude increases even
more when I turn on my DC power supply. I can pick this signal up
without even connecting the probes to anything.

That's like saying a radio antenna isn't connected to anything, so why
do we get a signal? There are 60 Hz waves travelling through the space
in your home/lab/wherever, and you are simply measuring them on the
scope.

Try disconnecting the probes at the oscilloscope input jack. The 60 Hz
should be gone, in which case it is not a problem with the 'scope
picking up stray 60 Hz or a bad grounding in the 'scope. But if the 60
Hz is still there, then yes there's a problem with the 'scope.

Next, reattach the probes to the oscilliscope, AND connect the probe
ends to anything that you know doesn't have 60 Hz. A common battery
would work. The 60 Hz should go away when you do this. But if it's
still there, you have a real problem.

Hope this helps.

Regards,

Mark