LCR Impedance Viewer for Picoscope+Keysight+R&S Bode Plot Data (open source)

[Hans Polak]

One last thing, here is the same file 100R+10nF but now compensated for the measuring probe with a calibration file.
That makes a visible difference, but although Resr is now a bit noisy at LF, it is still considerably above 10Khz.
I have attached both the image and the Excel file.

Hans

[_Wim_]

And as said, it's obviously not the math that is wrong...

And as said, it's obviously not the math that is wrong   

but there must be something going wrong in your calculation giving L=246H and Resr=50R at 100Hz instead of 21.6H and 43.8K, and there is more.

I wonder if it has some has something to do with the gain limit I already told you in post 104?

[_Wim_]

You mentioned to limit the gain to one.
That will indeed increase the value of L considerably (and influence the capacity) , but also the value of Resr will increase to ca 47K at 100Hz and higher.

I just checked your excel file and did two things:
- copied all the formula's from the first row down until the end (formula's in the first row where correct but they did not go all the way down)
- set all of the positive gains to zero

=> Resr goes to -50 ohms  and stays there.

But, you are using a logarithmic axis in Excel, and that does not plot negative numbers

So after all the things I have done so far, it seems fair to say that now it's your turn to tell what exactly it is that you do on top of calculating the formulas , that may help to explain part of the differences.

I only limited the gain to negative numbers as explained immediatly when you reported the problem. With the changes above in Excel you get the same results

[_Wim_]

One last thing, here is the same file 100R+10nF but now compensated for the measuring probe with a calibration file.
That makes a visible difference, but although Resr is now a bit noisy at LF, it is still considerably above 10Khz.
I have attached both the image and the Excel file.

Hans

To prevent repeating myself:

I am not sure how I need to interpret these result. Is the just probe the result without the 75pF, the the calibrated the result with the 75pF cap & then calibrated? Can you post a plot where you see in one graph:

- DUT measured with no 75pf cap at scope input
- DUT measured with 75pf cap at scope input
- DUT measured with 75pf cap at scope input & with calibration correction implemented

1 & 3 should be identical in an ideal calibration situation, 2 should be different enough to see the calibration "working"

[Hans Polak]

Wim,

1) What an incriminating suggestion that the formulas in my Excel file were not pulled down and that only the first row was correct.

2) What a totally wrong decision to make all gain figures above 0dB zero, but leave the rest for what it is.
This results is a totally crippled plot giving that silly 50R for Resr in case of the file we were discussing.
There is nothing wrong with values above 0dB, it can be because of an LC resonance or a channel gain issue, but FRA Imp Viewer can never know  what the reason is.
In case of a channel gain difference, one thing the  user could do if it bothers is to subtract the same dB value from all samples, but a much better thing to do is to correct the file with the probe's calibration file.
Conclusion: FRA Imp Viewer is indeed willingly and totally superfluous corrupting files.

3) Keep repeating yourself when it's not obvious what I mean with file compensation with the Probe's calibration file. Everybody else will perfectly understand what this means.

It's obvious from the above that I'm done with you, the way you react is like stung by a bee and turning every discussion into an ordinary fight.

Hans

[_Wim_]

1) What an incriminating suggestion that the formulas in my Excel file were not pulled down and that only the first row was correct.
 

How do you explain than that I take your excel, add the CSV data with only negative gains, pull the formula’s down and get instantly the same numbers we have been arguing about for pages?

2) What a totally wrong decision to make all gain figures above 0dB zero, but leave the rest for what it is.
This results is a totally crippled plot giving that silly 50R for Resr in case of the file we were discussing.
There is nothing wrong with values above 0dB, it can be because of an LC resonance or a channel gain issue, but FRA Imp Viewer can never know  what the reason is.
In case of a channel gain difference, one thing the  user could do if it bothers is to subtract the same dB value from all samples, but a much better thing to do is to correct the file with the probe's calibration file.
Conclusion: FRA Imp Viewer is indeed willingly and totally superfluous corrupting files.
 

As explained this decision was made because it makes measuring single components like capacitors and inductors easier due to the gain approaching 1. I can agree that this way maybe not the best decision, but I told you immediately that this was done (and kept repeating it that this was the reason for the difference), yet you keep ignoring this and claiming my calculation was wrong.

3) Keep repeating yourself when it's not obvious what I mean with file compensation with the Probe's calibration file. Everybody else will perfectly understand what this means.
 

I somehow doubt that, but as explained for “me” to understand and know what I should add in my code and how it works, I needed the data in a certain way. For “me” it was not clear, and if you expect me to add your wishes to me application, I do feel I am entitles to ask for the data in a certain way…

It's obvious from the above that I'm done with you, the way you react is like stung by a bee and turning every discussion into an ordinary fight.
 

Hans, for several pages long you are claiming all sort of stuff is wrong with my app, and ignore my pointers completely to check your calculation. You make bold claims on how I do the mathematics wrong (which made me actually doubt what I have done and made spent quite a bit of time digging everything back up again, mathematics requires time & effort for me and is certainly not my preferred topic to spent my sparse free time on). When I can disprove one of your claims, you only keep finding new claims that “something” is wrong. (which would be still ok if you state that you see a difference, and are not sure who is correct, but you always state it as you are 100% sure it is me, and I have to find the proof it is not)

When I finally receive your Excel on which those claims are based, I only need to copy in the CVS data with gain set to zero, drag the formula’s down, and get exactly all the results I have with my app, while just before you claim to have done the same this with setting gain to zero, but still see differences:

You mentioned to limit the gain to one.
That will indeed increase the value of L considerably (and influence the capacity) , but also the value of Resr will increase to ca 47K at 100Hz and higher.
So after all the things I have done so far, it seems fair to say that now it's your turn to tell what exactly it is that you do on top of calculating the formulas , that may help to explain part of the differences.
 

For me this looked like you have all the data at your fingertips, yet it was up to me again to find proof. This was indeed very frustrating to me, hence me reaction.

[_Wim_]

Forgot to mentions one more thing, when I was checking your formula's you had an absolute value for both capacitance and inductance, which for me looked like you were fully aware a log plot in excel does not like negative numbers, yet made me look why ESR did not go negative in your plot...

[_Wim_]

1) What an incriminating suggestion that the formulas in my Excel file were not pulled down and that only the first row was correct.

Somehow you made me doubt again that this was what actually happened, so I figured to redo what I have done before...

See formula bar to row1 and row2 or Resr. And this after we have been discussing about ESR negative or not 

[_Wim_]

Due to a question in another thread (https://www.eevblog.com/forum/testgear/device-impedance-using-dso/) I wanted to run my app again, but it did not run any longer. I had updated recently to the latest version of "FRA for Picoscope", and it appears the FRA's API has changed, so a small update to my app was also required.

While I had the app open, I made 2 other small changes related to the discussion above:
- renamed the tabs to clearly indicate series capacitance & inductance are measured
- added the option to disable the positive gain limiting

As always, updated version added in the first post (version 1.014) To run this version, the latest version of "FRA for Picoscope" must be installed, which can be found here: https://bitbucket.org/hexamer/fra4picoscope/downloads/

For users that prefer to run with an older version of "FRA for Picoscope", version 1.013 can still be used (also attached in first post)

[Cykeltur]

Hi this was interesting
i hava a old RCL circuit that i like to see the impedance sweep for, but if i do some simulation on the RCL circuit  the resonance is about 20kHz, but the result if i use 119Ohm resistor as a refference the plot is not showing the  20kHz responande :-/ any ideas
the values are just all over the place :-/

[_Wim_]

Can you post some screenshots of the result you get and the setup you are using while measuring?

[Cykeltur]

sure! :-)

The ReferenceResistor: 121 

The RCL is the following components in series:
R: 1.2k  (1190  )
C: 183 K (so probably 18000 pF  ± 10%
L: 3300uH (3,3mH) (3300uH±10%)
should result in :Resonance at 20650Hz (Qfactor : 0,36)
Using a simple RCL calc, like this one:
https://www.omnicalculator.com/physics/rlc-circuit

 

[egonotto]

Hello,

I think you have connected the oscilloscope incorrectly (wrong.png). Connect the oscilloscope to your circuit according to the green text in better.png.

Best regards
egonotto

[Cykeltur]

hi thanks!, i get the exact same result (the second plot is overlaying the old one)

(the photo is the same board so that is not included in the photo..)

[_Wim_]

hi thanks!, i get the exact same result (the second plot is overlaying the old one)

(the photo is the same board so that is not included in the photo..)

The DUT is the RCL, not only the R. So your red probe must be between the ref-resistor and the L.

[_Wim_]

Looking at your pictures, I see that is was just your drawing stating the incorrect connection. But I also see in your picture you are using a scope probe to connect the signal generator to the reference resistor. This is not ok, as scope probes use resistive coax which can have quite a bit of resistance. Use a BNC cable with crocodile clips or something similar instead.

[_Wim_]

Looking at your pictures, I see that is was just your drawing stating the incorrect connection. But I also see in your picture you are using a scope probe to connect the signal generator to the reference resistor. This is not ok, as scope probes use resistive coax which can have quite a bit of resistance. Use a BNC cable with crocodile clips or something similar instead.

I would even recommend not to use scope probes for channel A & B also, but this should have less negative impact and the connection with the signal generator. However, I never used scope probes myself, I just used short BNC cables or a test jig...

[mawyatt]

Here's something that one might find useful using a SDS2000X DSO for Impedance Measurements.

https://www.eevblog.com/forum/testgear/capacitive-impedance-plots-with-sds2104x-plus-bode-function/msg4335745/#msg4335745

https://www.eevblog.com/forum/testgear/admittance-measurements-with-dso-awg-with-bode-function/msg4491952/#msg4491952

Best,