EEVblog Electronics Community Forum
Products => Test Equipment => Topic started by: R_ZH on February 14, 2023, 09:33:49 am
-
Hello everyone,
I 'm using a impedance analyzer to test a RLC series circuit and want to extract the values. Finally I got a plot as shown below and the impedance analyzer gave me a set of parameters calculated by itself based on curve fitting. My question is in magnitude plot, is the magnitude axis linear or logarithmic? I need to find the self-resonant point of this RLC circuit like shown in the picture, if so, I have to set the magnitude to logarithmic on impedance analyzer. And if I set it to linear, it was not like this. I guess this results from impedance analyzer is a bode plot. I also use matlab to plot a bode plot, but it shows a linear axis of magnitude shown attached. I have been confused for long. Could anybody please explain this to me?
Best regards and thanks in advance!
Runze
-
The magnitude scale is in dB, which means that's a logarithmic scale. By definition, the Bel is the logarithm of a ratio:
https://en.wikipedia.org/wiki/Decibel
-
Do you mean that in the second picture, the magnitude is automatically set to logarithmic if the unit is dB, even though it looks like linear?
-
Not really.
To be pedantic, the first picture is wrong. The text on the Y axis reads "Magnitude (dB)", which means the numbers on the axis are decibels. However, the gradations on the Y axis are shown as a logarithmic ratio. The upper limit, for example, reads 108, so 100 000 000 (a hundred million). That hundred million is the ratio, not a hundred million dB! :scared:
First picture is wrong because the units of the magnitude axis is not dB, the unit measure there is adimensional (a ratio, has no units).
The second picture is correct. When the values are represented as decibels, the axis is usually linear. You can choose the axis of dB to be drawn in the chart in a logarithmic representation, but that would be rather unusual.
Either way, logarithmic axis with adimensional ratio, or linear axis in dB, the RLC values or the resonance of the circuit is not affected by the way the results are drawn in the chart.
-
Yes, the unit in first picture is not dB but probably Ohm. Since it is extracted from impedance analyzer and plot in matlab, there should be some difference from real bode plot. I can show you a screenshot directly from the impedance analyzer.
Yes, it doesn't make any difference to the results. But if I use linear axis for magnitude, it is shown in the picture below and I can't find the resonant point.
-
the unit in first picture is not dB but probably Ohm
It is not Ohm. A ratio has no dimensional units. A ratio only tells how many times something is bigger (or smaller) than something else. We don't say a tree is twice as big meters, or a pot is twice as big litters. We say it without dimension, because when we write a ratio, the measure units above and below goes away (when we simplify the ratio). 10W/100W = 0.1 (or -10dB), but NOT 0.1W
The resonant frequency is where the dip is in the magnitude plot, or where the phase changes in the phase plot. The logarithmic representation has the ability of "enhancing" or "magnifying" small values, so it is better to observe the resonance here.
-
OK, understand. But what ratio does it mean here? Is that the ratio of resistance at each frequency over the resistance at resonant frequency?
-
In this plot is no ratio:
(https://www.eevblog.com/forum/testgear/rlc-circuit-test-using-impedance-analyzer-agilent-4294a/?action=dlattach;attach=1716104;image)
The top of the Y axis is 10M\$\Omega\$, the bottom is 0.2\$\Omega\$, the |Z| means that the Y axis represents the modulus of an impedance (a complex number). So it's no ratio there, it's ohms represented on a logarithmic Y axis.
I don't know that instrument, the user manual might give more details about the meaning of that plot. Usually such instruments have markers that can be manually set, or they can autodetect where the highest/lowest peaks are.
Put a marker on the green line and move it to the lowest point, that is the resonant point. Looks to be at about 100MHz 30MHz, a marker should tell the exact frequency.
What do you need to measure, or to find out from the plots?
Is it a specific parameter you need to measure, or is it for didactic purpose?
-
OK, I will check the handbook.
Yes, I have already used marker to find out that point.
I want to find the resonant frequency point and use the slope below that point to calculate the capacitance value in this RLC circuit. The parameters displayed on the screen are given by the impedance analyzer. I want to calculate myself to check those values.
Yes, it is a specific parameter. It's actually a parasitic capcitance in a DBC layer.
-
If the first plot has the same data as the printscreen from the instrument, then the only thing that has to change is the units in the Y axis of the plot. I think the Y axis there should be "Impedance (\$\Omega\$)", not "Magnitude (dB)".
(https://www.eevblog.com/forum/testgear/rlc-circuit-test-using-impedance-analyzer-agilent-4294a/?action=dlattach;attach=1716077;image)
Resonant frequency seems to be at about 30MHz. This number corresponds with the values in the prinscreen (436.876pF and 65.265nH, which will give 29.805805MHz). \[ f = \frac{1}{2 \pi \sqrt{LC}} \]
-
Yes, you are right. It's reasonable to set Y units to Ohm and the resonant frequency is exactly the same as you calculated. Thanks for your answer!