LTS Analysis of RLC circuit and Testing w/o GND loop

Thread Starter

SamR

Joined Mar 19, 2019
1,293
1st the experiment for the following circuit is to measure Vpp across each component but there is a caveat in the instructions. "Avoid ground loops by moving resistor and inductor to keep the same ground point." I assume the resistor is in the correct place to measure and to measure the inductor they need to swap places in the circuit so when measured they will have the same relative ground point beside the capacitor?

2nd I would like to also do this in Spice. I have selected a "signal" and defined it for 10(volts) 6k(Hz) and AC sweeping to 6k(Hz) over an Octave? Or do a transient analysis?

RLC.GIF
 

WBahn

Joined Mar 31, 2012
24,850
If this is intended to be a physical circuit being measured with something like a typical oscilloscope, then the issue they are likely trying to deal with is that a typical scope cannot measure the voltage between two arbitrary points in a circuit -- the scope ground clip is electrically tied to the scope's safety ground wire. If the signal generator is ground referenced as shown in the circuit diagram, then what they are telling you to do is to rearrange the components in the circuit so that whatever component you are measuring you can connect the ground clip to the grounded node in the circuit and the probe to the other side of the component and be measuring the voltage across the desired component. In the current circuit you could measure the voltage across the capacitor. To measure the voltage across the resistor you would swap the resistor and the capacitor.

As for the second question, it depends on exactly what information you are trying to determine. Probably you want to do an AC analysis since it sounds like you are probably interested in the steady-state behavior of the components as a function of frequency.
 

crutschow

Joined Mar 14, 2008
23,487
Ground loops refer to the actual circuit, they have no effect on simulation unless you are deliberately trying to determine the effect of a ground loop.

AC analysis does a small-signal simulation over a frequency range assuming that all components are linear for any voltage level.
It first does a DC bias analysis to determine the operating points of the circuit components, and then uses a linear approximation for their operation at that bias point for the AC analysis.

Transient analysis with a single frequency sinewave, does the analysis at that single frequency and includes the non-linearities of the semiconductor devices as determined by the instantaneous voltages and currents in the circuit.

For an AC analysis, set the AC signal level to 1V.
The output plot will then be in dB value relative the input being 0dB, since the plot uses 1V as the 0dB reference (and dB is always a relative measurement).
 

Thread Starter

SamR

Joined Mar 19, 2019
1,293
Oh boy... Just tried to measure the Vpp across the inductor and:

It's oscillating...
SDS00001.gif

Here is what I was given:
IMG_0508.JPG

Here is what I have as I am measuring across the inductor which is oscillating:
IMG_0509.gif

I observed the apparent polarity of the signal generator and the scope and no shorts to GND apparently. When signal reduced to 100Hz I measure 73kHz across the inductor.
 

WBahn

Joined Mar 31, 2012
24,850
And you may well have a ground loop precisely because you didn't do what the NOTE says to do. Is your power supply floating? If not, then you definitely have created a ground loop.

Clip the ground ring on the scope probe to the negative terminal of the power supply and pretend it is welded there. Then move the components around in the series circuit so that the component you are measuring has one terminal tied to the reference point.
 

Thread Starter

SamR

Joined Mar 19, 2019
1,293
I assume the signal generator is floating since it only has a 2 prong plug. The outer part of the BNC is "assumed to be" GND and I moved the components around so the inductor is now connected to black SG lead as is the probe GND. OK success! Why wouldn't they just show that in the schematic!?!?

SDS00002.gif

Now on to the LTS. Thx guys!
 

Thread Starter

SamR

Joined Mar 19, 2019
1,293
Gotcha. They are listed by component name but hard to see. Ooops no they are not. K noted. Thx
 

WBahn

Joined Mar 31, 2012
24,850
I assume the signal generator is floating since it only has a 2 prong plug. The outer part of the BNC is "assumed to be" GND and I moved the components around so the inductor is now connected to black SG lead as is the probe GND. OK success! Why wouldn't they just show that in the schematic!?!?

View attachment 184668

Now on to the LTS. Thx guys!
The generator output may or may not be floating. The number of prongs doesn't guarantee anything either way. The fact that moving the components around changed the result significantly is pretty strong evidence that it isn't.

You can do a check by using a multimeter set for AC voltage and doing the following:

Have the generator and the oscilloscope completely disconnected from each other.
Turn on the generator with a fairly large signal being output.
Connect the scope probe tip to one terminal of the generator output while leaving the ground clip floating and note the signal.
Now move the scope probe tip to the other terminal and note the signal.
If there is a significant difference between the two, the generator output is not floating relative to the scope's ground reference.

Ideally, if it IS floating, you will see very little signal in either case -- just whatever noise is being picked up.
If it isn't floating, you will generally see a bit of noise when connected to the "ground" of the generator output and the full signal when connected to the other terminal.

As for why they didn't just show that in the schematic, they didn't want to draw three different schematics and felt the Note that they provided gave you the information you needed.
 

WBahn

Joined Mar 31, 2012
24,850
You don't want the display the voltages on the different nodes, you want to display the voltages across the three components. That's a very different thing. Configuring your simulator to display voltage differences is an extremely useful thing, so now's a good time to figure out how to do it.
 

Thread Starter

SamR

Joined Mar 19, 2019
1,293
Yes, that is what the 4 sine waves show, V across, PS, R, Xl, Xc. I had to stop for a while and need to do my calculations but already V/X are giving me different values for Xl and Xc and the V measured across R is almost Nil, very low mAs. I'll work on it more in the morning. And look at the SG outputs on the scope. Thx WB! Oops only 3 on that screenshot but I looked at all 4 to see the phasing.
 

Thread Starter

SamR

Joined Mar 19, 2019
1,293
f it IS floating, you will see very little signal in either case -- just whatever noise is being picked up.
If it isn't floating, you will generally see a bit of noise when connected to the "ground" of the generator output and the full signal when connected to the other terminal.
With no connection-
SDS00003.gif

Probe tip only to +
SDS00004.gif

Probe tip only to -
SDS00005.gif

Not sure what it is picking up but it's not 10V 10kHz that the SG is set for. V is the same but f changes from 106 to 60Hz? And they are connected minigrabber to minigrabber with the other connection open. So it's getting noise somewhere? And confirms that it is floating.
 
With no connection-
View attachment 184699

Probe tip only to +
View attachment 184700

Probe tip only to -
View attachment 184701

Not sure what it is picking up but it's not 10V 10kHz that the SG is set for. V is the same but f changes from 106 to 60Hz? And they are connected minigrabber to minigrabber with the other connection open. So it's getting noise somewhere? And confirms that it is floating.
Make these measurements again, but with the two prongs of your line cord plugged in reversed compared to the way they were for the measurements shown in post #13.
 

Thread Starter

SamR

Joined Mar 19, 2019
1,293
OK I did the math and didn't like the results. I did it over using the actual measured values for the components. They are kinda close but the percent error is a bit high even for experimental vs. Ideal (using measured component values). I only did it for 6kHz so far. For the measured value of L added the actual resistance across the inductor to Xl. If you can read my scrawl:

IMG_0510.JPG
 

WBahn

Joined Mar 31, 2012
24,850
Keep in mind that the inductor has resistance and that the tolerance on both the inductor and the capacitor is likely 10% to 20%.
 

Thread Starter

SamR

Joined Mar 19, 2019
1,293
That is why I used the measured values of the components and added the inductor resistance to Xl. I did expect something a bit closer though. I like to see less than 5% error. Everything except the resistor is less than 10% error? I did change out the cap as I was getting some strange resonance and having a hard time getting a value across the resistor and after doing so the scope sine wave settled down to a pretty steady value instead of fluctuating up and down a bit. I'll do the apparent and real power calcs later. I did order some better 5% .01uF and .001uF caps instead of the grab bag 20% ones I have for those values.
 
OK I did the math and didn't like the results. I did it over using the actual measured values for the components. They are kinda close but the percent error is a bit high even for experimental vs. Ideal (using measured component values). I only did it for 6kHz so far. For the measured value of L added the actual resistance across the inductor to Xl. If you can read my scrawl:

View attachment 184765
What instrument did you use to measure the inductor and the capacitor? What instrument did you use to measure the resistance of the inductor?
 

Thread Starter

SamR

Joined Mar 19, 2019
1,293
Capacitance measured with Vici VC8145 bench meter (also inductor R) and inductor with a GM328 Multi-use Transistor Tester for inductance. On my SZJB BM4070 LCR meter, I get 30.7mH & 76.1Ω for the inductor and 8.3nF on the cap. So the values are very close for both instruments. I can only assume from the calculation results that there might be some error for cap and inductor but the big error is for the resistor? Which measures at less than 1% error than the ideal 1kΩ marked. I could have swapped the resistor but swapping out the cap seemed to resovle measuring Vpp across it with the Oscope.
 
Top