PDA

View Full Version : Technical: Adding SPICE to your build



ihasmario
27-04-2014, 08:48 PM
SPICE = Simulation Program with Integrated Circuit Emphasis.

Just today I started investigating the possibility of examining the electrical properties of the wiring I was planning. A fair warning, this is specific technical territory and may or may not be relevant to you as a musician. The goal of this thread is to draw circuits for each individual guitar component consecutively, help me (and maybe you) decide which resistance pots, cap values etc will give you neutral output and determine the electrical properties of your pickups.

In this case I will be using a Lace Blue single coil for my sample values (from http://www.lacemusic.com/sensors.php). I do not have the exact capacitance value, but have estimated it by using the "Peak Frequency" spec given by lace - I have emailed them for more specific values. You will learn how I did that with this post. If you are interested in exact values for your pickups, contact the manufacturer or use a multimeter.

First we will look at the main electrical properties of a single coil of your pickup, with an arbitrary 1V source (ordinarily this would be induced). The voltage is going to have no impact on the graphs because of how I will be plotting them, which will be focusing on tone rather than volume.

Please note that the following features ideal components under ideal conditions.

http://a.pomf.se/ndyqli.png

A single coil pickup has three main electrical properties,

Inductance
Impedance
Capacitance

Each of these are clearly labelled on the pictures. We will be running an AC octave sweep with 12 steps per octave from 20hz to 20khz. The strength of the inductor is measured in Henries (in this case 6.58H) and others you are probably familiar with, ohms and farads.

The following graph displays the frequency (solid) and phase (dashed) output of the pickup diagram, zero'd at 20hz.
http://a.pomf.se/rhqpld.png

You will notice that, in this case, the pickup has a very high frequency spike (approx +18dB)at around 3khz. This spike is characteristic of this pickup and is due to a function of the three electrical properties of the pickup. During mastering, the spectrum of 2khz-4khz is generally boosted relative to others to increase articulation, but it actually serves another purpose later.


Inductance
http://a.pomf.se/uhjrpt.png
http://a.pomf.se/pmhavi.png

Increasing the number of Henries has three main effects; it raises the volume of the peak relative to other frequencies, and pushes the peak, phase change and roll off start into lower frequencies.

Impedance
Increasing the resistance of the pickup will roll-off the high frequencies, and eventually cause the circuit to have a round roll-off, making the guitar very muddy, whereas too little will make the guitar too bright. It's also worth noting that resistance will also affect the rate of change of the signals phase, a steep change for low resistance and a slow change for high resistance. A secondary effect is a reduction in the peak value.

http://a.pomf.se/kpllgr.png
http://a.pomf.se/keeirl.png


Capacitance
Before beginning, it is worth noting that relative to a tone cap (0.022?F/0.047?F), the listed value of 436pF is approximately 100 times less than the 0.047?F(0.000436?F)

http://a.pomf.se/hgxpob.png
http://a.pomf.se/eodgel.png

You can see that adjusting the capacitance affects the overall output in a similar fashion to adjusting the inductance of the circuit, except it creates a smoother roll-on/off and lowers the height of the peak.

In the next post we will be looking at what this means for your pot and cap choices later in the circuit for tone (specifically when aiming for true-to-source).

ihasmario
27-04-2014, 08:48 PM
Similarly to the previous section, there are going to be three major electrical properties that affect the tonal balance of your guitar if you are running a single volume and a single tone with a capacitor. The locations of these are labelled.


http://a.pomf.se/xtypve.png

These factors are
Vol Pot Resistance
Tone Capacitance
Tone Pot Resistance


It should be noted that it each of the following graphs will be impacted by all of the components, and therefore only changes between the graphs for each component should be considered a change by that component relative to the original pickup graphs.

Volume Knob
One might not think it, but the volume pot you choose is going to affect the sound... slightly. Pictured below are two graphs featuring two common Pot arrangements for volume, 250k and 500k. When fully "closed" (Max volume) the 250k still remove a greater portion of the higher frequencies than the 500k. The resulting graphs are below, with 100% of the available resistance being in front of the ground.

http://a.pomf.se/qskxzk.png
http://a.pomf.se/tcnqwf.png

Effectively, the resistance of the tone pot adds a roll-off to the signal. When there is a lot of resistance preventing the signal from reaching the nearby ground, you receive closer to the original signal. The guitars quietest volume with also be affected by the resistance of the pot (but this is generally well below what matters).

Tone Knob
The tone knob is similar to the volume knob, but it only attenuates frequencies that are allowed through, which is determined by the tone caps capacitance.

Tone Cap Capacitance
The following can be compared to the 250k graph from above.

Tone caps have an overall smaller effect when the pot is resisting the tone path than resistance does.
http://a.pomf.se/rnzpxp.png

When the resistance of the tone pot is lower (allowing more cap influence), a higher capacitance will allow more frequencies (Sweeping from very high down to low) to be affected by the cap.
http://a.pomf.se/jceqxu.png
http://a.pomf.se/getvyi.png

Therefore, the tone knob has less influence the smaller the capacitance, and acts more like a volume knob, the larger the capacitance. For example, it is easy to flatten the signal by using a large capacitor.

http://a.pomf.se/qhryst.png

This signal represents a true to source (strings) reproduction of the sound that is somewhere in the middle of a tone pot, allowing play to either side of bright and dark. However, we will generally want to stick to the brighter side of this as a target because of unideal conditions, and Eddy Currents within the guitar increasing its total resistance (highly subject to pickup type).

Because of what we know we can understand that we can increase the resistance of the tone pot (up to 500k) to bring the treble back into the sweep in this case.

You can achieve a flat-enough signal with smaller tone caps as well, however the midrange will become scooped and the bass end will lean up verrry slightly.
http://a.pomf.se/ozjgzs.png

ihasmario
27-04-2014, 08:58 PM
***Warning: Highly Subjective Material Below***

The Useful Frequency Range of A Guitar
This is where subjectivity starts. Firstly, the first useful tone for our strings is 82.41Hz, and lower for the wood. But we are mostly concerned with high strings... For standard tuning as I see it the goal upper frequency to reproduce evenly (when shooting for a neutral tone) is either 1318.4Hz (The first harmonic of string 12, and the fundamental of a hypothetical fret 24) or 2636.8Hz, which covers the first harmonic of a hypothetical fret 24.

In my case, and I think probably most we are going to struggle to reproduce an even response from 82.41Hz to 2636.8Hz. However, 2636.8Hz is of very limited use to us. So 1318.4Hz is the bare minimum I would personally aim for when shooting for neutral output.

Even without tweaking, you can see that I have achieved a neutral output in ideal conditions that exceeds this. However, on with the tweaking...

Ideal Tone Control
An ideal volume control turns the guitar up and down. An ideal (solitary) tone control turns the influence of the pickup up and down. By influence of the pickup, I am referring to what the pickups signal looks like by itself (see the first graph in the thread).

But how do we find the sweet spot? In this case, we are going to set both the Tone Cap and Tone Resistance very low. As we learned previously, this means we will be sending only very high frequency information to the tone circuit, and most of that information will be grounded.

In my case the resulting frequency plot is like this;
http://a.pomf.se/innozg.png

The next step then is to slowly dial in the capacitor, slowly increasing the amount of the high frequency information that is being filtered. You want to get to a point where the majority, or all of the original peak has been attenuated (the peak will appear to move as you raise the capacitance).

http://a.pomf.se/cstpyq.png

Here the peak has begun to move, so we are within the correct region, however the entirety of the original peak has not yet been removed.

http://a.pomf.se/zcmcop.png
Here we have removed the original peak. From here you have the choice of shaping your tone control by adjusting the capacitance and the pot resistance. Higher pot resistance, as we discovered will allow more of the pickup to come through when open.

And so we have our range of tones for a now 500k pot (to emphasise the pickups resonance more)
http://a.pomf.se/pfbcsf.png

Interestingly when we get super low we get the mid-range quack from the cap in addition to tuning out the pickup resonance.

Notable tone drift;
Approximately 16dB of pickup resonance (ignoring 3k due to colouration) - This is approximately twice the loudness (18dB being twice as loud)
Approximately 4dB of midrange quack (just over twice the amplitude)

Here's the comparison of the same sweep for the recommended capacitance and resistance (0.022mfd and 250k pot) from lace.
http://a.pomf.se/mqkoqv.png

Thanks to rolling the tone pot resistance up and (somewhat) the cap, we have gained approximately 6dB of tuning the pickup.
Thanks to rolling the tone cap, we have gained approximately 3dB (about 2x the amplitude) of midrange quack.

The subjective part; The signals are well within 0.5dB of eachother up until 5-600Hz with the new cap value, whereas they are about 2dB apart by then with the stock modifications. In my opinion this is something that is good, as it allows me to dial in the guitars pickups (or the quack) without messing up my external EQ.

And, for good measure, a graph of the stock recommended cap with 500k tone pot.
http://a.pomf.se/quynus.png

And a comparison with a 500k pot and a 0.047mf cap (Note that this is to highlight the loss of quack, rather than the loss of neutrality, as the values would need to be modified for that).
http://a.pomf.se/mljgvh.png

It's probably worth mentioning that if I wish to pursue the option of the quack (at this point I do), I will need to ensure that the minimum resistance of my pot is at or below about 15k. We can probably push more out of it by using a 500k volume pot


8-)

ihasmario
28-04-2014, 05:48 AM
Post reserved for discussion about where to tweak to because of the effect of eddy currents, how pickup type affects eddy currents, etc.

And ultimately how to do this for your guitar using the program I have. And how to have a mathematically perfect RC filter for your tone control.

keloooe
29-04-2014, 03:25 AM
Wow my head hurts ihasmario...

Great stuff though!

wokkaboy
29-04-2014, 05:16 AM
exactly that Ihasmario, far too technical for me to understand but appreciate the time you spent posting all this info.

ihasmario
29-04-2014, 06:47 AM
When I get the chance I'll do some examples and data for the pitbull guitar pickups (I need to get a multimeter for that) I have and explain step-by-step how to do it (including some maths to save time with your capacitance). I was learning this as I was going, through experimentation and figured it would be best to record things so I don't forget them.

Some of the guys who know more about wiring than me are probably cringing at the tone cap being in front of the resistor (where an ordinary RC would have it after), but it wouldn't have affected the results in such a simple example.

If you're just a guy who plays or builds guitar and isn't really interested in the why of the electrical side of things (or getting audiophile about them) then obviously this thread is going to be of limited use.

Thanks for reading though ;)

ihasmario
06-07-2014, 08:23 AM
This post contains information how to determine capacitor values for your specific pickups, if you would like to undergo a similar design criteria. Once my DIs arrive, I will post a demo of the results for you to browse over, and determine if you like the tone controls that resulted from my experiment.

Technical Information You Will Need:
Pickup Inductance in Henries (see manufacturer spec)
Pickup Capacitance in pF (see manufacturer spec, or measure it)
Pickup Resistance (see manufacturer spec, or measure it)
Peak Frequency (Not essential, but can help you fill in the gaps for the above information)
Potentiometer Resistance


First you will need LTSpiceIV;
Windows - http://ltspice.linear-tech.com/software/LTspiceIV.exe
OSX - http://ltspice.linear-tech.com/LTspiceIV.dmg
GNU/Linux - Please google for an open source SPICE program.

Secondly, you will need a potentiometer subcircuit. This .sub file goes in %LTSpiceIV%/lib/sub/.
If it is not installed correctly, you will receive an error upon opening the circuit files.
Potentiometer Subcircuit - https://app.box.com/s/b35mo8psvg2fxsmh7yxo

Next step is to download and open the relevant circuit for you;
/</p>[/<\\/p>[]<\\/p>/]Single Vol/Tone Circuit</p>/
Single Coil Pickup - https://app.box.com/s/m3dczyos42lhfeb4l3if
Two Single Coil Pickups - https://app.box.com/s/l7nvd6x0qdu2cdy1behl
Three Single Coil Pickups - https://app.box.com/s/lmtuistkoj9neugojqgt


In order to edit the values, simply right click on the component and enter the relevant values. You can write them in full, or use abbreviations, for example pF for pico Faradays, and K for kilo (eg. Kilo ohms).

The volpot values are defined as;
RTot - The maximum possible resistance of the potentiometer.
Wiper - Functions as a percentage; 0 = Totally open (0% of Rtot) going to output, 1 = Totally closed (100% of Rtot) going to output.
Rtap and Tap = Values at certain points in the circuit for "log" (i.e. in this case two different linear) curves. If you take the midway resistance of your potentiometer, you can set RTap to that value and then set wiper to the value of tap, and that will give you the output frequency of your pickups when your pot is centered (handy if you have a center detent).

To test the frequency output of your circuit (relative to the natural sound of the strings), press the "Running Man" icon, and then move your cursor over output and click once (or twice to disable other points in the circuit).


If you want to do this and want help, or want another circuit don't hesitate to ask for help.

kells80aus
06-07-2014, 09:29 AM
Ah. I f---in love science. I'll go through this later on my PC so I can get a better perspective. As the tablet being a smaller screen tends to loose context.


PK from my T2

lawry
06-07-2014, 10:21 PM
While we're talking about Spice. I use an iPad app called Spicy Schematics. I originally bought it when I was doing some electronic engineering volunteer work in Guatemala. The beauty of Spicy Schematics is that you do all of your design work on the iPad and then it sends the file to a university in USA (can't member which one). A minute or so later it sends back the simulation data, graphs, numbers, etc. This way you don't have to have a whopping big program and library on your iPad but you get full a featured SPICE package. And you may also be surprised to know that just about everywhere in Guatemala you can get free wifi (cafés, public spaces, pubs, etc.). I kid you not.

leetbrown
08-07-2014, 08:46 AM
I had been looking at doing this as well. Found some information at http://buildyourguitar.com/resources/lemme/ that went into the analysis a bit and had a little bit more information about modeling the circuit in different ways.