Maker Pro
Maker Pro

problem encountered when using function generator and oscilloscope

Chengjun Li

Oct 21, 2014
84
Joined
Oct 21, 2014
Messages
84
Hi,
I observed the following phenomenon when using Oscilloscope(Tekronix TDS 2024B) to measure the output voltage of function generator(Agilent 33250A).
upload_2015-2-17_10-24-6.png
1. connect both leads of function generator to the oscilloscope probe and ground clip, ideal sine wave 10 Hz is observed.
2. Only connect positive lead to the oscilloscope probe, a 60Hz signal is superimposed on the the 10Hz signal.
3. Only connect negative lead to the oscilloscope probe, a 60Hz signal is observed.

Things I know, 1. the 60Hz signal is electric supply. 2. The oscilloscope measures the potential difference between probe and ground clip which is the ground.

My questions are,
1. why the negative lead has a voltage of 60Hz, is it just due to the electric-magnetic field in the air?
2. In the first image, may I say that although the negative lead has a voltage over ground(from last image), when the it is connected to the ground clip, the voltage on it is forced to be zero?
 

davenn

Moderator
Sep 5, 2009
14,240
Joined
Sep 5, 2009
Messages
14,240
there is no problem ... the scope just needs to be used correctly and connected as in pic 1

with the GND lead disconnected as in pic 2, you now have an unbalanced situation, which is susceptible
to picking up random noise and that superimposed signal is probably a radiated one and the GND lead is acting as an antenna


2. In the first image, may I say that although the negative lead has a voltage over ground(from last image), when the it is connected to the ground clip, the voltage on it is forced to be zero?


not sure what you mean by that ?


Dave
 

Gryd3

Jun 25, 2014
4,098
Joined
Jun 25, 2014
Messages
4,098
There was a guy here not long ago that hooked up a microcontroller to light running along a bed frame. He only used 1 wire to take care of the signalling, and did not use any additional wire as a common of reference. He could not control the lighting because of what are you seeing. This is not a problem, it is a symptom.
You cannot do this unless you have some common reference point to measure from.

Otherwise, without a reference, you begin to pick up noise and other random emissions.
This could be picked up by either lead, and the results you are seeing may not reproduce 100% on other makes and models depending on their internal circuitry.

Keep in mind that you are only seeing this because the scope and function generator are isolated from each other... please also note the voltage that is being picked up at 60Hz. This value will not be consistent depending on nearby equipment and other factors.
 

Chengjun Li

Oct 21, 2014
84
Joined
Oct 21, 2014
Messages
84
there is no problem ... the scope just needs to be used correctly and connected as in pic 1

with the GND lead disconnected as in pic 2, you now have an unbalanced situation, which is susceptible
to picking up random noise and that superimposed signal is probably a radiated one and the GND lead is acting as an antenna





not sure what you mean by that ?


Dave
What I mean is that, since the ground clip is connected to the ground,I think we can tell the function generator's negative lead has certain potential difference with reference to ground from the last image(Vpp=1.4V). While in the first image, the negative lead is connected to ground directly, so there is no potential difference of negative lead with respect to ground, since ground has a potential of 0V, we can say the negative lead has a 0V potential. The absolute potential of negative lead is changed from Vpp=1.4V to V=0, is this analysis right?
 

Chengjun Li

Oct 21, 2014
84
Joined
Oct 21, 2014
Messages
84
There was a guy here not long ago that hooked up a microcontroller to light running along a bed frame. He only used 1 wire to take care of the signalling, and did not use any additional wire as a common of reference. He could not control the lighting because of what are you seeing. This is not a problem, it is a symptom.
You cannot do this unless you have some common reference point to measure from.

Otherwise, without a reference, you begin to pick up noise and other random emissions.
This could be picked up by either lead, and the results you are seeing may not reproduce 100% on other makes and models depending on their internal circuitry.

Keep in mind that you are only seeing this because the scope and function generator are isolated from each other... please also note the voltage that is being picked up at 60Hz. This value will not be consistent depending on nearby equipment and other factors.

"Otherwise, without a reference, you begin to pick up noise and other random emissions."
Are you saying connections in image 2 and 3 have no reference? But I think the since the ground clip is connected to ground directly, couldn't we say in image 2's situation the positive lead has a certain absolute potential difference with reference to ground?
 

Gryd3

Jun 25, 2014
4,098
Joined
Jun 25, 2014
Messages
4,098
"Otherwise, without a reference, you begin to pick up noise and other random emissions."
Are you saying connections in image 2 and 3 have no reference? But I think the since the ground clip is connected to ground directly, couldn't we say in image 2's situation the positive lead has a certain absolute potential difference with reference to ground?
No. Image two and three don't have any connections to 'ground'.
If one of those meters had a ground connection, and you hooked it up to a circuit that is powered from mains with a real ground, you would cause damage.
This is like leaving the black probe on a voltmeter disconnected and poking around in a car with the red lead. You may get results, but they will not be accurate, or predictable.
'Ground' is not a magic reference unless both devices are connected to it, and using it in part of the related circuit.
Remember that electricity flows in a circuit, it needs a return path which is why you need both wires. (Cheaper equipment may have the negative wire connected internally to ground... so you would get different results with it)
 

Chengjun Li

Oct 21, 2014
84
Joined
Oct 21, 2014
Messages
84
No. Image two and three don't have any connections to 'ground'.
If one of those meters had a ground connection, and you hooked it up to a circuit that is powered from mains with a real ground, you would cause damage.
This is like leaving the black probe on a voltmeter disconnected and poking around in a car with the red lead. You may get results, but they will not be accurate, or predictable.
'Ground' is not a magic reference unless both devices are connected to it, and using it in part of the related circuit.
Remember that electricity flows in a circuit, it needs a return path which is why you need both wires. (Cheaper equipment may have the negative wire connected internally to ground... so you would get different results with it)
"If one of those meters had a ground connection, and you hooked it up to a circuit that is powered from mains with a real ground, you would cause damage." That makes sense. But from the manual of the oscilloscope, it is said "The probe ground lead is at ground potential." I wonder if it is not connected to ground, how could it always keeps at ground potential? The other question is, if as you said, a closed loop path is necessary, otherwise what we got is something random and unpredictable, then why in image two, we can always see the there is a component of 10Hz signal(which is the one I want to get)?
 

Gryd3

Jun 25, 2014
4,098
Joined
Jun 25, 2014
Messages
4,098
"If one of those meters had a ground connection, and you hooked it up to a circuit that is powered from mains with a real ground, you would cause damage." That makes sense. But from the manual of the oscilloscope, it is said "The probe ground lead is at ground potential." I wonder if it is not connected to ground, how could it always keeps at ground potential? The other question is, if as you said, a closed loop path is necessary, otherwise what we got is something random and unpredictable, then why in image two, we can always see the there is a component of 10Hz signal(which is the one I want to get)?
The unpredictable behavior is referring to the use of other equipment. Just because it works on your gear does not mean it will work on others.
As far as the book for your scope stating that the ground probe is actually at ground potential is concerned. This is always a good thing to know.
If you have the book for your function generator, it would be best to dig through that to see if there is an internal ground reference.

It still boils down to improper use. Anything from here on out will be mostly specific to your equipment (well.. how it's internally wired).
 

Chengjun Li

Oct 21, 2014
84
Joined
Oct 21, 2014
Messages
84
The unpredictable behavior is referring to the use of other equipment. Just because it works on your gear does not mean it will work on others.
As far as the book for your scope stating that the ground probe is actually at ground potential is concerned. This is always a good thing to know.
If you have the book for your function generator, it would be best to dig through that to see if there is an internal ground reference.

It still boils down to improper use. Anything from here on out will be mostly specific to your equipment (well.. how it's internally wired).
I understand this is a special case, only for my situation.

I found the diagram of the function generator.
upload_2015-2-18_13-10-48.png
I think may be the return path is achieved by using earth as a conductor. Current flows like shown below.
Do you think this is reasonable?
upload_2015-2-18_13-11-52.png
 

Gryd3

Jun 25, 2014
4,098
Joined
Jun 25, 2014
Messages
4,098
I understand this is a special case, only for my situation.

I found the diagram of the function generator.
View attachment 18721
I think may be the return path is achieved by using earth as a conductor. Current flows like shown below.
Do you think this is reasonable?
View attachment 18722
Certainly reasonable. I would like to draw your attention to the diagram you posted from the function generator... Notice that the 'ground' actually has two symbols?
One is 'common' and the other is usually 'earth ground'... however, there is also a Vgnd... Vgnd commonly refers to Virtual Ground, which is not the same as earth.. this indicates a little additional circuitry on the ground side of the function generator.

If this were instead a direct connection to the same ground the scope had, you would most likely see nothing on the scope on image number 3 on your first post.
Additionally, you would not see the 60Hz mains freq on image 2 either.

I would attribute this to the ground(s) in the scope and function generator to be a high impedance. If you were to clip the negative lead of the function generator directly to ground, you may get different results.
 

Chengjun Li

Oct 21, 2014
84
Joined
Oct 21, 2014
Messages
84
Certainly reasonable. I would like to draw your attention to the diagram you posted from the function generator... Notice that the 'ground' actually has two symbols?
One is 'common' and the other is usually 'earth ground'... however, there is also a Vgnd... Vgnd commonly refers to Virtual Ground, which is not the same as earth.. this indicates a little additional circuitry on the ground side of the function generator.

If this were instead a direct connection to the same ground the scope had, you would most likely see nothing on the scope on image number 3 on your first post.
Additionally, you would not see the 60Hz mains freq on image 2 either.

I would attribute this to the ground(s) in the scope and function generator to be a high impedance. If you were to clip the negative lead of the function generator directly to ground, you may get different results.
I am still not clear about how was the 60Hz signal introduced into the close loop. Would you please give me more clue?
 
Top