How to change the capacitance of a pickup?

[professorprogfrog]

Hey everyone! It's my first post here and I'm also a high school student with no experience in guitar work, so please do keep this in mind.




I am currently working on a physics research project that involves measuring the relationship between the capacitance of a pickup and the frequencies produced. As part of this project, I would need to have 5 different pickups with each having a different capacitance value. My question is, how do I change the value of the capacitance in a pickup?



I would appreciate all the help I can get, thank you

[antigua]

You'd probably want to put capacitors with value between 1nF and 100nF in parallel with the pickup to lower the frequency to an audible extent, as shown here  www.buildyourguitar.com/resources/lemme/  down in the section titled "Altering Pickup Characteristics". 

Probably the most straight forward way to demonstrate for a school project would be to get assorted caps on Amazon, a couple alligator clip leads, a guitar and an amp. Then somehow expose the positive side of the circuit somehow, such as unscrewing the end of the guitar cable so that you can access the solder joints, then connect the alligator clips to the positive and ground of the guitar cable. If the guitar cable doesn't unscrew, on some guitars such as a Strat or a Les Paul, you can take out the screws holding the input jack into the guitar, pull it out and connect the alligator clips to the input jack of the guitar. 

When you hook up cap values such as 2nF, 10nF, 33nF and 47nF (or anything in between), with the alligator clips, putting the cap in parallel with the guitar, you should hear a distinct difference in the tone when the higher cap values causes the peak frequency of the pickup to drop. You can plug the guitar into the amp, turn it on, strum a few chords, change the cap, strum a few chords and show how the sound changes. 

This would work best with a Strat or Tele, because the Fender single coils tend to start with a high peak frequency which you can then drop down with caps. Guitars with humbuckers tend to have a lower frequency to begin with, so the caps make a difference but it's not as pronounced. 

That's a quick and dirty presentation, if you wanted to get sophisticated you could change the cap value with a dial or switches. If you want to show the actual plot, you could try using the Rightmark audio software and use a computer soundcard and a small exciter coil to make frequency plots without spending much money, but that might still be a lot of work. 

[aquin43]

Jul 1, 2023 1:24:21 GMT -5 professorprogfrog said:

Hey everyone! It's my first post here and I'm also a high school student with no experience in guitar work, so please do keep this in mind.




I am currently working on a physics research project that involves measuring the relationship between the capacitance of a pickup and the frequencies produced. As part of this project, I would need to have 5 different pickups with each having a different capacitance value. My question is, how do I change the value of the capacitance in a pickup?



I would appreciate all the help I can get, thank you

What is the exact wording and purpose of the project? It is easy to alter the total capacitance in the circuit including the pickup as antigua has explained but it is difficult to measure the intrinsic capacitance of the pickup accurately and how would you alter it without altering all of the other parameters - with difficulty - perhaps immerse it in an oil with a high dielectric constant?

If the circuit behaviour is what is required then something interesting to explore would be the variation of the resonant frequency and Q of the pickup impedance with capacitive loading. Use a humbucker or a P90 and note how the Q variation with resonant frequency deviates from the ideal. Then provide an explanation.

[JohnH]

It could be a really good project.

The capacitance that is built into the actual pickups, due to the wire windings, is very small and not really heard on its own, and it tends to be dominated by other capacitances, such as the guitar cord and other wiring. eg, the pickup capacitance might be 100pF = 0.1nF, but added to that would be the cord from guitar to amp which might be about 500pF = 0.5nF. Then, when you turn the tone control down, a much bigger cap is added as well. Take a look on our 'tone control discussions' section.

So as Antigua suggests, you can get some action by adding capacitors as he suggests.

Maybe the school has a meter that can measure capacitance? worth asking about, then you could try to measure the cord and also some actual capacitors.

The other thing that you could get into about this would be the maths of it. Maybe you have covered some in class? like the resonant frequency of a coil with inductance and a capacitor. Even a fairly full analysis using complex numbers isn't that hard to convey and you can plot it on a chart. That may be a college topic though I remember it from maths at high school in the UK. Not essential though, just a nice demonstration of tones and a concept of what is happening will still be a great piece of work.

Please keep talking to us! lets keep you interested because this could lead to something for you.

[gpoint]

Hi everyone.
   Pickup's INNER capacitance C is not so easy to measure (it is a little bit tricky), but for more or less normal passive pickup it is somewhere in between 50 pF and 200 pF. It depends on construction, materials and distances inside - capacitance in between coil's wire and base plate, shielding, cover, close located grounded metal/conductive parts (magnets, poles, screws, springs, etc.) + very complex LCR capacitance in between windings.
   The OUTER capacitance is sum of pickup's cabling/wiring, passive electronics, output jack, + cable (including plugs), and the next inputs capacitance. Both INNER and OUTER capacitances are parallel and must be summed into TOTAL shunting capacitance. Normally it is somewhere in between 300 pF and 600 pF, cable provides some 50+% of it. But there are spiral cables with up to 3000 pF capacitance !
   There is a number of factors affecting pickup's tone, but shunting capacitance is one of the factors and a very special factor. Every pickup has a resonance peak in its graph, provided by INNER capacitance game with INNER inductance L. Normally this resonance peak is somewhere in between 2000 Hz and 8000 Hz. Resonance peak's frequency (and other parameters - height and width) are providing the most important part of the CHARACTER of the pickup.
   Shunting OUTER capacitance is changing this resonance FREQUENCY rather strong, as a result different cables "sound" different. You can easy model it using additional parallel connected capacitors.

[ms]

An LCR metter at 100 KHz can do a pretty good job.  For example, a Filter'tron I had sitting on the table measures 186 pf, kind of high for a pickup with two coils in series, but this does include the capacitance of the shielded lead made from very thin cable, probably high C. I prefer to measure the impedance over a wide range of closely spaced frequencies  and use some math to get a capacitance value at a frequency much lower than 100KHz.

[professorprogfrog]

Jul 1, 2023 3:08:55 GMT -5 antigua said:

You'd probably want to put capacitors with value between 1nF and 100nF in parallel with the pickup to lower the frequency to an audible extent, as shown here  www.buildyourguitar.com/resources/lemme/  down in the section titled "Altering Pickup Characteristics". 

Probably the most straight forward way to demonstrate for a school project would be to get assorted caps on Amazon, a couple alligator clip leads, a guitar and an amp. Then somehow expose the positive side of the circuit somehow, such as unscrewing the end of the guitar cable so that you can access the solder joints, then connect the alligator clips to the positive and ground of the guitar cable. If the guitar cable doesn't unscrew, on some guitars such as a Strat or a Les Paul, you can take out the screws holding the input jack into the guitar, pull it out and connect the alligator clips to the input jack of the guitar. 

When you hook up cap values such as 2nF, 10nF, 33nF and 47nF (or anything in between), with the alligator clips, putting the cap in parallel with the guitar, you should hear a distinct difference in the tone when the higher cap values causes the peak frequency of the pickup to drop. You can plug the guitar into the amp, turn it on, strum a few chords, change the cap, strum a few chords and show how the sound changes. 

This would work best with a Strat or Tele, because the Fender single coils tend to start with a high peak frequency which you can then drop down with caps. Guitars with humbuckers tend to have a lower frequency to begin with, so the caps make a difference but it's not as pronounced. 

That's a quick and dirty presentation, if you wanted to get sophisticated you could change the cap value with a dial or switches. If you want to show the actual plot, you could try using the Rightmark audio software and use a computer soundcard and a small exciter coil to make frequency plots without spending much money, but that might still be a lot of work. 

Hey, thanks for your reply, I really appreciate this. I used your reply to do my entire experiment, with a few minor changes.


Firstly, the input jack method did not work, so instead I took apart the pick guard and soldered a wire on the tone caps. I then managed to "peek" out these wires for each tone cap so I could manually attach alligator clips to them and just swap out caps. 


Now, I do have a few questions.


Firstly, I think a big mistake I did was I set the tone caps to 10 for all the experiments. Does this actually result in any perceived tone change or did I just waste my time? I can simply just swap out the cap values again because they're outside my guitar so it's no big deal, but I would prefer an explanation.

Next, how do I keep this experiment as variable free as possible? I'm referring to the strumming. I tried to keep it as similar as possible but it's hardly precise.

Lastly, how do I process the data and show mathematically that the produced sound is different? Does that involve frequencies or what. Any app a pro like you would recommend?

Thank you very much for your help, and I hope you have the time to respond.

[antigua]

Sept 22, 2023 15:31:20 GMT -5 professorprogfrog said:

Jul 1, 2023 3:08:55 GMT -5 antigua said:

Hey, thanks for your reply, I really appreciate this. I used your reply to do my entire experiment, with a few minor changes.


Firstly, the input jack method did not work, so instead I took apart the pick guard and soldered a wire on the tone caps. I then managed to "peek" out these wires for each tone cap so I could manually attach alligator clips to them and just swap out caps. 


Now, I do have a few questions.


Firstly, I think a big mistake I did was I set the tone caps to 10 for all the experiments. Does this actually result in any perceived tone change or did I just waste my time? I can simply just swap out the cap values again because they're outside my guitar so it's no big deal, but I would prefer an explanation.

Next, how do I keep this experiment as variable free as possible? I'm referring to the strumming. I tried to keep it as similar as possible but it's hardly precise.

Lastly, how do I process the data and show mathematically that the produced sound is different? Does that involve frequencies or what. Any app a pro like you would recommend?

Thank you very much for your help, and I hope you have the time to respond.

If the tone control is set to 10, then you won't hear a difference with the tone cap. You hear the biggest difference at 0, all the way "down". If the tone control is not at zero, then you have some variable amount of resistance in between the pickup and the capacitor.

Here are a couple plots that show the idea: 

this shows how the response changes with the tone knob:



this shows how the response changes with the caps in parallel with the pickup, meaning the tone control is at zero:




As you use higher value caps, you will hear the pronounced treble part of the signal move downwards in frequency, almost identical in sound to a wah wah pedal.

When it comes to strumming, if you are recording the sound and comparing waveforms, then strumming is a big issue because you can hardly get a constant sound with a plectrum. It's really difficult even with a mechanical device. But if you are just listening with your ears, just strumming consistently is enough. In the same way that the sound difference of a wah wah pedal is obvious no matter how you strum, so will be the difference with various caps. Just for show, you can put a piece of tape on the guitar, and only strum the strings where that piece of tape is located.

For the math, to show how the resonant peak changes, you take the inductance of the pickup (what kind of pickup is it?) ,  and then take your capacitance value and plug the values into here www.omnicalculator.com/physics/resonant-frequency-lc to get the resonant frequency.  You just have to make sure you get the units right, henries for the inductance and microfarads for the capacitors, so that the peak frequency calculates correct. This ignores the pickup's capacitance and cable capacitance, but for demonstration purposes, you can just make a notes that you've ommited that, and omit it.

[stratotarts]

Is this a learning assignment, i.e. school project? As a science experiment, it starts out on the wrong foot because it sets out with a dubious premise - that pickup capacitance greatly affects the tone. That isn't the case. You should do some preliminary research to find out what the actual major factors are, so that your experiment will be meaningful, and also easy to perform with simple equipment. That research will also give you adequate context to explain your findings. I suggest you do that, and re-frame your experiment.

[professorprogfrog]

Hi,

I've done my experiment and everything sounds right. Higher cap values have a "muddier" sound, which is good since it shows my experiment is working. Now it's time for data analysis.


So basically what I did is that I changed the cap value of the tone pot to other caps and recorded the sound via my Focusrite audio interface. I recorded 3 trials but also in two ways (so 6 total). In one I set the tone knob at 10 and recorded it, another I set it to 0.


Now then, what software do I use do actually show that the higher frequencies were "reduced"? Something like a graphical eq. It would be great if it's also a plug-in for ableton since it makes my life easier.

Additonally, it is my understanding that there are different types of frequencies, so what would be the difference between sound frequencies and, lets say, simple harmonic motion frequency?


Would love some clarification for this.

Thanks in advance

[mikecg]

Oct 24, 2023 9:25:51 GMT -5 professorprogfrog said:

Hi,

I've done my experiment and everything sounds right. Higher cap values have a "muddier" sound, which is good since it shows my experiment is working. Now it's time for data analysis.


So basically what I did is that I changed the cap value of the tone pot to other caps and recorded the sound via my Focusrite audio interface. I recorded 3 trials but also in two ways (so 6 total). In one I set the tone knob at 10 and recorded it, another I set it to 0.


Now then, what software do I use do actually show that the higher frequencies were "reduced"? Something like a graphical eq. It would be great if it's also a plug-in for ableton since it makes my life easier.

Additonally, it is my understanding that there are different types of frequencies, so what would be the difference between sound frequencies and, lets say, simple harmonic motion frequency?


Would love some clarification for this.

Thanks in advance

Hello prof,

The 'Ableton V11' audio processor appears to be a rather advanced piece of software with many bells and whistles - a bit of a 'sledgehammer to crack a nut', in my opinion. However, if you have access to the full 'Live 11 Suite' version you should be able to use the 'EQ Eight', and 'Spectrum' 'audio effects' (see section 24.15, and 24.41 in the user manual). Note that these 'audio effects' are not available in the 'Intro' and 'Lite' editions. These 'effects' allow you to view the frequency spectrum of the signal from your guitar, and you should be able to see and record the effect your added capacitance has on the guitar's frequency spectrum. By the way, if you are not familiar with the term 'frequency spectrum', you may be interested to know that it refers to a way of describing the content of an electromagnetic (EM) or pressure induced (sound) 'signal' in terms of frequency and corresponding amplitude. If you do not have access to these features ('effects') in your version of the 'Ableton' software, you can try using the audio spectrum analyser in the 'Audacity' free audio processor.

[antigua]

Ableton has a spectrum analyzer  .  You might have to try some different sound recordings to see the effect of the tone cap clearly, I think playing a single note in several octaves might work well, for example the high E and the low E at the same time, which should show a harmonic series in the output. 

> Additonally, it is my understanding that there are different types of frequencies, so what would be the difference between sound frequencies and, lets say, simple harmonic motion frequency?

Sound is changes in air pressure happening at rates between 20Hz and 20kHz, which is the range detectable to ears, but realistically most people have an upper limit closer to 15kHz. Harmonics are fractions of a fundamental wavelength, so when you play a low E, the lowest frequency 82.4Hz is the fundamental, and then when you perform a pinch harmonic at the 12th fret, it divides the length in half, so you hear it's 1/2 harmonic that has a frequency of about 165Hz, double of 82.4Hz. If you pinch at the 7th fret, you divide the length by three, the frequency triples to 249Hz, and so on.

The thing to remember is that all of those harmonics can be heard when you pluck the string, but they're not all equally loud. So when you pluck the low E with a pick, you hear 83Hz, and 165Hz and 249Hz, and up and up, all at the same time. When you do a pinch harmonic at the 12th fret, you mute the fundamental of 83Hz, so all you hear is 165Hz and upwards, it essentially becomes the new fundamental note, and all the harmonics are now harmonics of 165Hz. 

When you apply the tone cap to the circuit, what you're doing in effect is removing the harmonics from the output signal. If the cap rolls off all frequency beyond 1000Hz, then any harmonic above 1000Hz is removed from the sound output. The reason higher value tone caps make the guitar sound muddy is because the fundamental note all by itself, without any harmonic content mixed in, sounds "muddy", there's no treble in the sound, it's like the sound of the floor vibrating.

[timtam]

Making frequency measurements on a complex audio signal that is changing significantly over a short period of time is a complex problem in digital signal processing. There are multiple engineering textbooks written on the subject of different types of frequency measurements (FFTs, short-time FFTs, octave filters, spectrograms, wavelets, windowing etc), and what the limitations of each are. There is no single approach that works well for all applications. Each can introduce artefacts of the mathematical process required to calculate them, that can show things that weren't actually in the original audio signal.

On the other hand, digital audio workstation (DAW) software like Ableton is obviously intended more towards recording and displaying audio signals that are then judged mostly on what one hears. DAWs are not really designed for taking measurements. They are more concerned about creating useful displays for making subjective heard judgements than they are about precise accuracy.

For a high school project I suspect any frequency graph format you can get on the screen that simply shows the same sort of differences that you can obviously hear should be sufficient. But finding that may take some work. Exerting some control on that process would involve picking specific times after the sound is initiated at which to grab a screen shot of the frequency spectrum to include in your results. If you don't control for time in some way you will end up comparing parts of the signal where the frequency content is different just because it naturally changes over time, independent of the effect of the different capacitors. So if you don't control the measurement time you will find that the display doesn't match what you know you heard.

You will see something very different if you have recorded strums of all the strings, or just plucked single strings. A single string will vibrate at a fundamental frequency plus higher harmonic frequencies, at integer-multiples of the fundamental. So you may be able to see each of those discrete frequencies in the frequency spectrum chart of a single string, which will each decay in amplitude at different rates over time (hence the need to pick specific times; also higher string vibration frequencies decay more rapidly than lower ones). Those harmonics may clearly show the effect of different capacitors, when the tone control is on 0. If however you have recorded full strums, you will see a complex mixture of the fundamental frequencies of each string, plus all their harmonic frequencies. It will probably be harder to make sense of that.

Just like picking one or more particular points in time to focus on, you should pick one or more particular frequencies to focus on, within the range of frequencies that appear to be most affected by the capacitors. That may still be hard to get consistent results from, depending on how consistent your recorded signal is. These are all part of the difficulties in getting good measurements from audio signals.

[professorprogfrog]

Understood. I'm redoing my experiment due to some issues I had, and I'll try only playing one note. Maybe this will help show the dampened frequencies. Would you recommend I play the low e string? I want to play a note where the difference will be very pronounced, so please do suggest me on which string I play.

Additionally, how do I reply to somebody's comment? This is mainly a reply to timtam but of course I'm welcome to anybody's suggestion

[unreg]

Oct 25, 2023 12:20:58 GMT -5 professorprogfrog said:

Additionally, how do I reply to somebody's comment?

Hi professorprogfrog,

Simply:
1.) log in
2.) scroll up to, or just find, timtam’s post
3.) press the “Quote” button; it’s to the far right of timtam’s name on his post
4.) then delete, or cut away parts of his post you aren’t replying to; OR copy the text section you do want to reply to, then select all of his text inside the quote tags, and paste your copied selection you want to reply to.
5.) place your cursor below (or after) the /quote tag, then type your reply


Perhaps step 4 is easier on a cpu or Mac or Unix machine; I’m usually replying with my phone.