Stupid question about resonant frequency test.

[kayakerca]

Jim and Steve,

Glad you have worked this out. I don't know if you have started testing with typical volume and tone controls, and real or simulated cable capacitance into an amp input load, but the CGM-101 can do it.

Jim, if you aren't loading the pickup with more capacitance, the resonant frequency will be a lot higher. Check out Lemme's discussion on this in:

BuildYourGuitar.com :: The Secrets of Electric Guitar Pickups

-Charlie

Thanks Charlie. I'll have a look at that discussion.

[kayakerca]

Thanks Jim,

got it going ... thanks
[ATTACH=CONFIG]28196[/ATTACH]

Duncan 1980s 5R (13.37k DCR, 6.75H)

Excellent!

The phase chart on the bottom can be adjusted so the entire phase curve plot shows (not that I've learned what that represents yet), but without the split. Just right click in the chart area and select the appropriate range. I messed around a lot with the measurement ranges in he upper left panel until I landed in a place where the speed of plotting was acceptable (rate of ~ 1.07 I think) and the power was such that the scope image on the left wasn't clipping.

A cool little tool!

[Stratz]

For those using Circuit Gear PC scopes (I have the CGR101) Circuit Gear has a small update which works great for displaying the peak readings. No more screwing with the mouse to find the peaks in the cross hairs.
It's simply an updated netalyzer.tcl file to replace the one in your install directory and once installed you will have the new peak detector option which is activated by right-clicking on the magnitude plot window in network analyzer mode.

Contact Circuit Gear for the new netalyzer.tcl file.

[kayakerca]

For those using Circuit Gear PC scopes (I have the CGR101) Circuit Gear has a small update which works great for displaying the peak readings. No more screwing with the mouse to find the peaks in the cross hairs.
It's simply an updated netalyzer.tcl file to replace the one in your install directory and once installed you will have the new peak detector option which is activated by right-clicking on the magnitude plot window in network analyzer mode.

Contact Circuit Gear for the new netalyzer.tcl file.

I've sent them a note to see if the equivalent upgrade is available for the CGM-101 unit that I have. Not overly optimistic though, as the recent software and firmware upgrades did not contain that functionality.

[CheopisIV]

I've been in touch with the Syscomp folk a few times and just got an updated netalizer.tcl for the CGM-101 today. It allows the peak indicator and also resizing of the graph to fit your screen better.

I've been given permission to share it so here's the updated file;
http://www.sigilpickups.com/bench/netalizer.tcl

Just copy/paste it into the folder where the program is installed. It's going to be included in future software updates but why wait

[kayakerca]

I've been in touch with the Syscomp folk a few times and just got an updated netalizer.tcl for the CGM-101 today. It allows the peak indicator and also resizing of the graph to fit your screen better.

I've been given permission to share it so here's the updated file;
http://www.sigilpickups.com/bench/netalizer.tcl

Just copy/paste it into the folder where the program is installed. It's going to be included in future software updates but why wait

Thanks CheopisIV! I must not have a pretty enough smile as I sent a request to the support link two times without getting any response.

Works perfect BTW!

[CheopisIV]

Lol, I think it was how I fluttered my eyelashes.

Jim, have you been getting good results with your sweeps? I've found that my own measurements are (maybe) good for getting resonance peaks but all my plot lines look nearly identical up to the peak which tends to differ. I've tried somevery different pickups and the results are all coming back too close to be accurate. I've tried 3 different drivercoils with the same results and I know its all hooked up right.... did you have to adjust anythjng in the settings?

[the great waldo]

Hi Stratz
In case you missed my post syscomp have also updated the cgr-101 software to enlarge both scope and analyzer windows.

Cheers

Andrew

For those using Circuit Gear PC scopes (I have the CGR101) Circuit Gear has a small update which works great for displaying the peak readings. No more screwing with the mouse to find the peaks in the cross hairs.
It's simply an updated netalyzer.tcl file to replace the one in your install directory and once installed you will have the new peak detector option which is activated by right-clicking on the magnitude plot window in network analyzer mode.

Contact Circuit Gear for the new netalyzer.tcl file.

[kayakerca]

Lol, I think it was how I fluttered my eyelashes.

Jim, have you been getting good results with your sweeps? I've found that my own measurements are (maybe) good for getting resonance peaks but all my plot lines look nearly identical up to the peak which tends to differ. I've tried somevery different pickups and the results are all coming back too close to be accurate. I've tried 3 different drivercoils with the same results and I know its all hooked up right.... did you have to adjust anythjng in the settings?

Hi CheopisIV. Any snow out west this week?

Many of my winds have very similar plots, but I would expect that sort of thing as I have a wind pattern that I use most of the time that works for me. I'm looking for reproducibility within each "model" I wind, so it's a good thing in my view.

I followed charrich56's recommendations on the driver coil and wound a coil from AWG 32.5 to about the 85Ω area. I tried various ways of setting up how I place the driver coil wrt to the test coil. Placing it on top gave me terrible plots. I'm guessing it was a strength of the generated filed issue, but I can't be sure. I settled on placing the coils side by side (parallel configuration) with a spacer between them and set the waveform Amplitude to 50% on the scope. Below is a pic of the setup and spacing.



Here is a pic of one of my early prototype pickups (grey) and a pickup taken from a Affinity Series Squier I now use as one of my test guitars. You can also see the setting I use when I run plots. What I get from this comparison is that the secondary harmonics should be much more prominent in the prototype pickup as the output is a bit more than 4 db more powerful (and that is a lot) around the resonant peak area. You also see that the Squier is much flatter than the prototype which may also contribute to less dynamic secondary harmonics. Now I may be totally missing the boat in my layman analysis, but what I see here is what I hear when I put one of my test guitars beside a Squier, the test guitar pickups have a much more dynamic feel to their tone. Anyway, right or wrong, that's how I interpret plots from the CGM 101 scope. I like that little scope and it's inexpensive!

[CheopisIV]

Thanks for the pics. Are you running your driver coil direct without any resistor inline? I looked at the white paper again and realized they don't have one on the pickup circuit... I tried it both ways and get different results, but it's always a similar plot line even going from a humbucker to a single coil. With the resistor, I get much better looking results from the scope side when I check the different signals (square, sine, etc)

Here's how I had it set up;


and the resulting plot line;


This driver coil was approximately the same as the one you describe but with 38AWG. I just wound up another to 129k to see if there was any difference but it gives similar plots. I'll try it again soon and take some new pics. On your plot line you can see the similar nature of the lines in the lower frequency section with the same humps and valleys across both pickups. This is the part I'm unsure of.

Edit: I've also found that going 'Linear' is a nice way to clip the graph top the exact frequency range I want...helps to read the manual once in a while

[kayakerca]

Thanks for the pics. Are you running your driver coil direct without any resistor inline? I looked at the white paper again and realized they don't have one on the pickup circuit... I tried it both ways and get different results, but it's always a similar plot line even going from a humbucker to a single coil. With the resistor, I get much better looking results from the scope side when I check the different signals (square, sine, etc)

Here's how I had it set up;
[ATTACH=CONFIG]29283[/ATTACH]

and the resulting plot line;
[ATTACH=CONFIG]29284[/ATTACH]

This driver coil was approximately the same as the one you describe but with 38AWG. I just wound up another to 129k to see if there was any difference but it gives similar plots. I'll try it again soon and take some new pics. On your plot line you can see the similar nature of the lines in the lower frequency section with the same humps and valleys across both pickups. This is the part I'm unsure of.

Edit: I've also found that going 'Linear' is a nice way to clip the graph top the exact frequency range I want...helps to read the manual once in a while

I do not use a resister inline when I am running a plot with the driver coil in Network Analyzer Mode. I do have a resister inline when I am just checking the resonant peak of the pickup in scope mode. It looks like you have pole pieces in the driver coil. I am not an engineer, but I believe you should remove them so they do not in impact the field generated by the driver coil in any way. Making it a true air coil. The pole pieces may be why your Oscilloscope Display shows no clipping even though you have the Wavefore Amplitude set to 100%. I have to set mine to ~ 50% so there is no clipping showing up for the driver coil signal in the scope display for the driver coil.

Can you run it on a single coil to see what type of plot shape you get. I have no HB's lying around to test and compare compare to your plot.

Let me know if any of this makes sense.

[CheopisIV]

That was an older pic, I'd already pulled the slugs

I think I have it figured out now, my plots are looking a lot more dynamic with different pickups than previously especially in Linear mode where the graph will tighten up to the min and max cutoffs rather than the larger sweep. I played with the Amplitude settings and it didn't change much with my bigger driver coil but made quite a difference on my small one! I also have the Input A set on 1X and Input B on 10x. Here's a set of Tele pickups, neck VS bridge. The neck must be reverse wound which is why it has the spike on the bottom graph.

[kayakerca]

That was an older pic, I'd already pulled the slugs

I think I have it figured out now, my plots are looking a lot more dynamic with different pickups than previously especially in Linear mode where the graph will tighten up to the min and max cutoffs rather than the larger sweep. I played with the Amplitude settings and it didn't change much with my bigger driver coil but made quite a difference on my small one! I also have the Input A set on 1X and Input B on 10x. Here's a set of Tele pickups, neck VS bridge. The neck must be reverse wound which is why it has the spike on the bottom graph.

[ATTACH=CONFIG]29286[/ATTACH]

The spike on the lower graph is not really a spike, its measurement has just run off the bottom of the defined scale range that was set (starts again on top). For that particular pickup plot, right click on the phase chart and change the Phase Axis to (+90,-270) and rerun the plot. The entire plot will be shown as you would expect, withing the range defined.

In my opinion (my wife always makes me say that part), I think it is incorrect to run the input B at 10X for this application. From my reading of the Lemme book, you should be looking at peaks in the (I think I remember) the 4K to 7K area on the regular single coils, which is what I experience in my winds. If you rerun this at 1X, your peak will drop ~ 2K, which I think is probably a more expected result for a AWG 42 single coil.

Another thing is the Frequency Step. I find that it graphs nice and quick at 1.10 where you have it, but the steps are very large realtively which I believe makes it less accurate. I find running it at the 1.03 area gives me more consistent results that are very comparable to the shortest steps at the 1.01 setting. 1.01 is too slow to be practical so I run at the 1.03 setting.

Anyway, just a few of my thoughts and observations. Again, I don't know sh!t about this stuff as I'm not an engineer, just based on what I have been reading and I'm probably wrong about all of it.

[CheopisIV]

Thanks Jim, the tip on the bottom (phase?) scale is helpful and I think you're right about the 10x VS 1x on Input 2. Looking at the pics on the 'White Paper' shows their probes running at 10x on Channel A and 1x on Channel B, but I find switching from 1x to 10x on Channel A only moves the plot line up or down on the dB scale so is fairly irrelevant, but changing Input B between the two does give quite a different reading!

I'll have to find my copy of Lemme's book and re-read his discussion on it to see if there are any other insights I'm overlooking.

When the software designer passed on the new file, he suggested I offer any suggestions on improvements or issues with it... now, he may have just been being polite, but I'm going to take him up on it and compile a list! Here's what I have so far. Anyone else have additions to the requested features list?

1) The two plots are useful for direct comparison of 2 pickups but the first plot when saved loses all data. It would be really cool if there was an option to input data directly into the graph, or over top of the graph. Something along the lines of a multi line text box so I could enter the pickup model and resonance peak.
2) More plots would be really useful for comparing more than two pickups at a time. I currently take a screenshot and use Photoshop to pull the plots, change their color and overlap them on the same graph.
3) Export as JPG or other image format would save a few steps of taking screenshots and pasting/editing them in Photoshop.

[kayakerca]

Thanks Jim, the tip on the bottom (phase?) scale is helpful and I think you're right about the 10x VS 1x on Input 2. Looking at the pics on the 'White Paper' shows their probes running at 10x on Channel A and 1x on Channel B, but I find switching from 1x to 10x on Channel A only moves the plot line up or down on the dB scale so is fairly irrelevant.

I'll have to find my copy of Lemme's book and re-read his discussion on it to see if there are any other insights I'm overlooking.

When the software designer passed on the new file, he suggested I offer any suggestions on improvements or issues with it... now, he may have just been being polite, but I'm going to take him up on it and compile a list! Here's what I have so far. Anyone else have additions to the requested features list?

1) The two plots are useful for direct comparison of 2 pickups but the first plot when saved loses all data. It would be really cool if there was an option to input data directly into the graph, or over top of the graph. Something along the lines of a multi line text box so I could enter the pickup model and resonance peak.
2) More plots would be really useful for comparing more than two pickups at a time. I currently take a screenshot and use Photoshop to pull the plots, change their color and overlap them on the same graph.
3) Export as JPG or other image format would save a few steps of taking screenshots and pasting/editing them in Photoshop.

Being able to import and export plots would be very useful, but you would have to know exactly the setting for the Frequency Step and Waveform Amplitude to make them comparable as they couldn't dynamically adjust when you load a new plot generated under different app parameters.

I like that you can now "drag" the plot larger from the lower right corner, but it would be nice if you could us the ctrl key while dragging to keep the graph's perspective the same. Makes it more meaningful as when you look at it as a log graph stretched to a wide screen it looks almost like a linear graph if you know what I mean.

I use the Windows snippet tool to save a copy of a particular plot and lay multiple plots over each other in a product I use that is like Photoshop. So long as you take them in the same size (e.g. size when it the app is oscilloscope launched) they map perfectly on top of each other an you just use opacity settings on the layers to make them all visible. You can use the graph markings and setting characters to nudge them until they are all perfectly lined up. That said, this is a work around for having the ability to have more than 2 plots and and import function.

All in all, I really like the tool for the money and ease of use.