Pickup on rails? :)

[Arx]

i totally understand the thing about 'sweet spots' not existing. but let me throw something out there.

what if the pickup was in such a position so that for a majority of fretted notes (or at least the most common fretted notes), even harmonics lie over the pickup. it would possible, however difficult, to map out the lower level harmonics for each fretted length of string. it wouldn't be too hard to avoid the 5th and 7th harmonics. then its just a matter of choosing the best fit. i really don't feel its that important, but just an idea.

of course, everyone has a different idea of what a 'sweet spot' is. for me, i would say an even mixture between closed and open tone. for others, something else.

The thing to remember is that the harmonics localized in specific locations. If you're 45 degrees off the point with the maximum amplitude, there's still significant content there. If you were to map it out, you would realize that as you analyze each successive harmonic the number of frets your selectivity is active on reduces, and quickly.

for the second harmonic, if you put the pickup at the peak (1/4 of the scale length) you'll have a null when you're playing on the 12th fret. The 50% point would be around the 5th fret.

That sounds like a reasonable range, but the thing to remember is that that spot will be a null on your fourth harmonic, so it's no good, you'll have to shift it to increase this one, but that decreased your second, and lets in more 3rd harmonic.

You can iterate it as far as you want. The further you go the more chaotic it'll seem (though it's actually not chaos, but a comb filtering effect).

Check this out: http://www.till.com/articles/PickupResponseDemo/index.html

Keep in mind it's logarithmic, The -10dB point is where it will sound approximately half as loud (ignoring psychoacoustic masking, which actually makes things much worse)

Now just add a second pickup, and see what happens

 

[Arx]

It's not possible to determine the most commonly fretted notes. I think their are only 2 solutions that would do what you want.

1) Have a 12 string bass, one note for each string and play all open strings.

2) Have an actuator that very quickly shifts the pickups based on what note you are fretting (watch the fingers or you'll have them sheared off LOL.) This would be ridiculously expensive to implement.

Even if this wasn't so impractical, it still wouldn't really accomplish much. If you were looking for the highest amplitude on one specific harmonic it would, but your other harmonics will be out of your control.

Incidentally, for #1, you may as well take it all the way. It's called a piano.

 

[WarriorJoe7]

I think with the 2nd solution, even with certain harmonics out of whack, you could still get consistency on all notes (i.e. make the pickup shift to the 3/4 mark for every note... that way however the harmonics lay for one note it will lay out the same way for all.) Not sure if this would make it sound any better and it would be expensive.

 

[uethanian]

i'd say the bass guitar would have a REALLY uninteresting tone if every note had the same harmonic content. thats basically a synth.

if you wanted to get really crazy, you could have one giant pickup that covers half the length of the string, and then digitally filter out the harmonics you don't want. durr that would look funny

 

[Arx]

i'd say the bass guitar would have a REALLY uninteresting tone if every note had the same harmonic content. thats basically a synth.

if you wanted to get really crazy, you could have one giant pickup that covers half the length of the string, and then digitally filter out the harmonics you don't want. durr that would look funny

Actually a big pickup like that would self cancel at a lot of frequencies. What you would have to do is have an extremely sensitive pickup extremely close to the bridge, and then EQ the crap out of it to get all the harmonics to equal levels.

This is actually my secret plan for my next bass . It'll probably be more work designing the filters to make it sound right than it will be to design the rest of the bass, but it should be fun.

Of course, even that won't be 100% consistant note to note, but I think it's the closest you can get.

It's definitely going to be an experiment though. It might jjust sound really wrong until I intentionally introduce some comb filtering, since that's a lot of what gives a bass its "tone"

-Nick

 

[uethanian]

Actually a big pickup like that would self cancel at a lot of frequencies. What you would have to do is have an extremely sensitive pickup extremely close to the bridge, and then EQ the crap out of it to get all the harmonics to equal levels.

This is actually my secret plan for my next bass . It'll probably be more work designing the filters to make it sound right than it will be to design the rest of the bass, but it should be fun.

Of course, even that won't be 100% consistant note to note, but I think it's the closest you can get.

It's definitely going to be an experiment though. It might jjust sound really wrong until I intentionally introduce some comb filtering, since that's a lot of what gives a bass its "tone"

-Nick

i really like your idea. but you would only be getting the higher octave harmonics. based on where the harmonics lie on a bass string, i'd say a bridge pickup by itself would read harmonics 9+ (just a guestimate).

do u mean a big pickup would put too much magnetic pull on the string and limit its harmonic content?

haha this is a good thread, but it has nothing to do with the post topic

 

[BobXboB]

i totally understand the thing about 'sweet spots' not existing. but let me throw something out there.

what if the pickup was in such a position so that for a majority of fretted notes (or at least the most common fretted notes), even harmonics lie over the pickup. it would possible, however difficult, to map out the lower level harmonics for each fretted length of string. it wouldn't be too hard to avoid the 5th and 7th harmonics. then its just a matter of choosing the best fit. i really don't feel its that important, but just an idea.

of course, everyone has a different idea of what a 'sweet spot' is. for me, i would say an even mixture between closed and open tone. for others, something else.

If you can listen to a note and tell me if it has more even or odd harmonics then this could make sense. If you can't then it's pointless. IMO

 

[BobXboB]

By the way, the old school method for controlling what harmonic content a note contains is moving your fingers. Try it. There is an amazing amount of control over tone by where you pluck.

Pluck exactly 12 frets above your fretted note and the first harmonic dominates.

 

[WarriorJoe7]

yeah i've noticed that it sounds like a ring modulator at the 12th fret of the open string and at the 15th fret when you fret the 3rd string etc.

By the way, the old school method for controlling what harmonic content a note contains is moving your fingers. Try it. There is an amazing amount of control over tone by where you pluck.

Pluck exactly 12 frets above your fretted note and the first harmonic dominates.

 

[Arx]

i really like your idea. but you would only be getting the higher octave harmonics. based on where the harmonics lie on a bass string, i'd say a bridge pickup by itself would read harmonics 9+ (just a guestimate).

do u mean a big pickup would put too much magnetic pull on the string and limit its harmonic content?

haha this is a good thread, but it has nothing to do with the post topic

No. The harmonics don't lie in a particular spots, they're everywhere to some degree, except for the infinitely small node points (nulls).

That this means is that your really huge pickup would pick up both positive and negative parts of the wave, and would cancel out.

near the bridge is the only place where all the harmonics line up in a simple and consistent fashion where I should be able to correct them with a relatively simple filter, as opposed to the crazy comb filtering which would be needed anywhere else.

The problem, is that the low harmonics have VERY low amplitude, and I may not be able to amplify them enough without ending up with a lot of noise, or even microphonics and potential feedback problems..

If you can listen to a note and tell me if it has more even or odd harmonics then this could make sense. If you can't then it's pointless.

A note which has primarily even or odd harmonics will sound quite different. Unfortunately there's not an easy way to get all the harmonics.. If you could, you could digitally filter them however you wanted (with some side effects of course).

I bet a piezo bridge is probably about the most harmonically full conventional pickup. It's coloured by the absorption of the wood a lot more than a magnetic pickup, but theoretically I think it should be quite good.

-Nick

 

[Jazz Ad]

The Westone Rail is by far the most advanced experiment ever made in this area.
I loved mine, really cool bass.

 

[Arx]

nifty... doesn't look like the most comfortable to play though.

 

[WarriorJoe7]

If Tillman's program is correct then it is working exactly the opposite of what some of us here on the thread think. Now this is according to trials on Tillman's Applet... so if his calculations are incorrect then so is this post.

To dispel one myth, getting "near the bridge" DOESN'T get you all the harmonics in the best pattern. As a matter of fact if you do this you only get a serious lack of fundamental, which you can't make up for. In the example below the small yellow line represents where the fundamental lies on the spectrum and the gray bar protruding from it covers 2 octaves up from it:

The other myth with using a huge bridge that goes from neck to bridge, gets you a massive sounding pickup. Extremely huge low end with each harmonic proportionately lesser. (more about why wider pickups sound more powerful later... this supports that though.)

It's funny but if you want to get the best harmonic response use as small (unwide) a pickup as possible and have the pickup at the 12th fret (keeping in mind that this is optimized for the open string.) Notice that the width of my pickup is only .1 inches in the applet. The closer you get to zero the more ideal. It is counterintuitive that the thinnest sounding and looking pickup picks up the most harmonics, but it's true. I suppose all the voluminous upper harmonic content is what makes it sound thin. (So with a low pass filter could you make a thin pickup sound thick?) The 2nd picture is more realistic unless you want to inlay a pickup into your fretboard. It is a .1 wide pickup as close to the neck as possible and it is the next best thing.

So in current designs it looks like the jazz bass bridge pickup is the closest that we have for the most harmonic content. It is the thinnest pickup and it is closest to the 12th fret. (no i am too lazy to go measure for exact measurements for placement and width so I winged it.)

Here is a humbucker width pickup in the same position as the jazz. Notice that the their is less upper harmonic content (making it sound more powerful.)

Now playing around with this applet I found something VERY interesting. The only thing that might be a drawback is that this applet only calculates in a perfect world so it might not work as perfectly as a discovery I made...

Lets say you put a 2 inch wide pickup in a certain spot. Now replace that with two 1 inch wide pickups right next to each other in the same spot. You get the same exact harmonic readout (in the perfect applet world.) The fact that you can put seperation between them means you have more versatility. If this is reasonably close in the real world then that means if you could make a pickup that was maybe 1/8 inch wide then you could add or subtract pickups (by using the volume) in order to make it respond like a jazz bass or a musicman pickup. And you could also move them apart for even more versatility.

According to the applet 2 pickups that are touching act like 1 pickup the size of the 2 combined. Though the fact that there will be more copper in the 2 seperate coils (and in a different configuration) makes me wonder how that would make it different in real life. I suppose the applet isn't taking EVERYTHING into consideration, nonetheless I am extremely curious about where this might truly lead.

I wish Tillman's applet had a series/parallel option rather than the somewhat useless (for bass anyways) phase option. Series/ Parallel would make us able to simulate a humbucker by having 2 small coils next to each other. I know how phase works where you just add the 2 signals together and it averages the signals out by either reinforcing or cancelling out frequencies. However I don't know how the combination of series pickups and parallel pickups are different from one another theoretically. Can anyone help us out here?

 

[WarriorJoe7]

Oh yeah I must say that piezos work completely different than Tillman's applet. They may indeed be the best capturer of all harmonics but we won't know from this.

 

[lemur821]

If Matt Till's program is correct then it is working exactly the opposite of what some of us here on the thread think. Now this is according to trials on Matt Till's Applet... so if his calculations are incorrect then so is this post.

To dispel one myth, getting "near the bridge" DOESN'T get you all the harmonics in the best pattern. As a matter of fact if you do this you only get a serious lack of fundamental, which you can't make up for. In the example below the small yellow line represents where the fundamental lies on the spectrum and the gray bar protruding from it covers 2 octaves up from it:

You're misreading that chart. Moving the pickup towards the bridge decreases the fundamental by just a little bit, and any amp's tone controls are enough to make up this "deficit".

Furthermore, the humps in the chart do not correspond to harmonics, which are not even shown. A J-bass bridge pickup is in position to catch every harmonic of the A string up to nearly 1kHz. In P-bass position, it doesn't even make it to 500Hz. Getting closer to the bridge can't give you all the harmonics, but it gives you a much more accurate picture of the lower ones.

EDIT: What I take away from this is that pickup position and width have little bearing on how deep, fat, or mid heavy your tone can be. Since even in P position the pickup gets a clear picture of the low harmonics (and even a neck pickup gets them up to 200Hz), it should be possible to accurately boost the lows of a bridge pickup to resemble those of a neck pickup. The question is how you want to filter the highs (and what tone you want before any EQ, of course).

EDIT AGAIN: Although I suppose different positions may have significant effects on lower harmonics as you play up the neck.

 

[WarriorJoe7]

I don't see how I am reading the chart wrong, but please clear it up if you can. (Your post does not help me so far.)

Here is what I see:

The input frequency is 110Hz which I belive is the fundamental for an open A string. (A is 440Hz... also 220, 110, 55, and also 880, 1760 etc.) I don't know which A is which off my head but if 440 is A4 then 110 is A2 (2 octaves below etc.)

The gray bar covers from 110 to approximately 440 and encompasses the last half of the fundamental and all of the 2nd harmonic, and the first half of the 3rd. The yellow line is centered on the fundamental harmonic which (correct me if I am wrong) since it doesn't sit in the readout at just 110Hz on a normal spectrum analyzer, it protrudes out to the left and right a bit.

Knowing this, the best measure of the fundamental is directly on the yellow line. And alot of the part under the gray bar is first and 2nd harmonic, where most of the power of bass (the low end) comes from.

If you look at some of the screen captures, the fundamental is strong when it is at zero dB (read it at the yellow line.) It is at 0dB in the 3rd picture. It comes close to 0dB within the gray bar on MOST of the other pictures. The one exception is the bridge placed pickup (first pic.). There is a HUGE difference in the fundamental with the bridge pickup. It is at -20dB at the yellow line and doesn't quite make it to -10dB under the gray bar. (Note: Notice that this pickup is placed almost as near the bridge point as possible. A regular jazz bridge pickup can be corrected fairly well with an eq but it is a few inches off the bridge. I don't think a jazz pickup position is what was proposed above by Arx. He wanted a pickup closer to the bridge but I am pointing out why it won't work well without major tweaking with an extremely good eq.)

Oh BTW -10dB is approximately 10 times weaker than 0dB and -20dB is a HUNDRED times weaker, nothing to sneeze at. Many Eqs only handle +/- 6 to 9 dB though some on a stereo go to +/-15. Also EQs are not perfect bands which is why extremes can get muddy easily. Sure you can boost way down there but since EQs are not in perfect bands it's not going to sound right (Sorry I can't say it better than this) if you just boost the lows in most cases. It will also boost some of the low mid frequencies (which are already fairly high) and make the thing sound so muddy. It would be much better to just move the pickup towards the middle for a much more natural bass boost.

I could be wrong but for the most part I think I hear what the graphs show. I will not hear minute differences but I think I can tell different pickup sizes and placements if I am given several recordings of the same exact bass with significantly different pickup sizes and placements. I think the graphs and the way I read them support what I hear.

But please tell me how to read them if this is incorrect.

 

[WarriorJoe7]

Also I must not that it would be easier to see my point if the harmonics were marked as you mentioned. The fundamental IS marked at the yellow line though and at the end of the gray bar is the 3rd harmonic. it's easy enough to see the first few with your mind and this is where most of the basses tone comes from anyways. You can tell by looking at the pic of the bridge pickup why it is so thin. The area of the first 3 harmonics is VERY weak. -20dB is kind of a big drop. The 5th harmonic or so is strong but you are clearly in upper guitar territory. There ain't no common eqs that can fix this, let alone one on an amp.

I think the only way to prove it would be to get a sound program with a spectrum analyzer and record a pickup as near the bridge as possible (can be done by mounting a pickup on the other side of the strings near the bridge. Yes above the bridge and in towards the body ever so slightly, then having it taped directly to a good short instrument cable into a DI.) Then use a normal EQ to try to correct the thin sound and post before and afters of your spectrum. Good luck eking a fundamental out of it without huge mud!

You're misreading that chart. Moving the pickup towards the bridge decreases the fundamental by just a little bit, and any amp's tone controls are enough to make up this "deficit".

Furthermore, the humps in the chart do not correspond to harmonics, which are not even shown. A J-bass bridge pickup is in position to catch every harmonic of the A string up to nearly 1kHz. In P-bass position, it doesn't even make it to 500Hz. Getting closer to the bridge can't give you all the harmonics, but it gives you a much more accurate picture of the lower ones.

EDIT: What I take away from this is that pickup position and width have little bearing on how deep, fat, or mid heavy your tone can be. Since even in P position the pickup gets a clear picture of the low harmonics (and even a neck pickup gets them up to 200Hz), it should be possible to accurately boost the lows of a bridge pickup to resemble those of a neck pickup. The question is how you want to filter the highs (and what tone you want before any EQ, of course).

EDIT AGAIN: Although I suppose different positions may have significant effects on lower harmonics as you play up the neck.

 

[WarriorJoe7]

There are a few shortcomings with Tillman's applet, however I think the shortcomings do not affect my analysis since I stay away from them or make up for them. The last one though I don't know how you'd account for.

1) no series parallel, just phase (useful for guitar but usually not bass.)
2) you can only analyze an open string (unless you have a list of scale lengths and frequencies to change the settings to.) It would be nice if you could fret the string by clicking.
3) the lower harmonics aren't marked very well and the upper harmonics are not marked at all.
4) It would be nice if you could select 2 pickups and transpose one over the other for comparision.
5) If the harmonics were on yellow lines it would be nice to be able to click on one and hear a wave that is very strong only at that freq (it is impossible to mute all the other harmonics completely though.)
6) It doesn't take into account the amount or configuration of the coil within the pickup.

Since Tillman has just updated his blog on that site on 5/25/08 he is still around. I am going to email him and see if he can chime in on the thread. I would love to know if I am not reading it correctly. He is an Engineer and obviously did a great job on this even if he didn't get it perfect for my needs. He obviously knows more about sound than i do also.

 

[uethanian]

geez this has gotten serious...

 

[DonTillman]

Yo.

I'm Don, the guy who wrote the Pickup Response Java Applet. I don't
normally do bulletin board things, but warriorjoe007 asked me to
coment.

There's a lot in this discussion that shows a serious misunderstanding
of the issues. The applet is the third in a series of articles about
some interesting aspects of guitar/bass pickups. I realize it's fun
to treat the applet as a video game by itself (hey, I had a blast
building it) but the nuances of what is being represented are
described well in articles 1 and 2. So please read those first.

htttp://www.till.com/articles
​

Joes's "shortcomings":

1) no series parallel, just phase (useful for guitar but usually not bass.)
​

Series/parallel is all about coil characteristics interacting, but
nothing about the coils are included in this simulation. The goal of
this project is not to accurately model all aspects of an instrument,
it's to help visualize this one very interesting aspect, where the
frequency response sort of mimics the body resonances of an acoustic
instrument, but scales that up and down with the string tunings. That
doesn't happen anywhere else, on any other instrument. It's part of
what makes guitars so cool. Likewise the applet applet does not body
resonances, coil characteristics, tone controls, speaker resonances,
or any of those other features.

2) you can only analyze an open string (unless you have a list of
scale lengths and frequencies to change the settings to.) It would
be nice if you could fret the string by clicking.
​

That's a misunderstanding, please read the articles. What's displayed
is an overall frequency response plot for all the notes on a given
string.

3) the lower harmonics aren't marked very well and the upper
harmonics are not marked at all.
​

Um, none of the harmonics are marked.

It would be a fine improvement to add harmonic ticks, sure, and the
next update I'll probably do that. Although, at this point I really
ought to recode the thing in Flash; Java Applets are no longer the
preferred technology for this sort of thing.

At the same time you need to understand that harmonics aren't all that
important in the grand scheme of things. I know, that's what they
teach at school, but let's be honest, if somebody sucked the 7th
harmonic out of a performance nobody would notice.

4) It would be nice if you could select 2 pickups and transpose one
over the other for comparision.
​

Maybe, but it would be a hard to read the display.

5) If the harmonics were on yellow lines it would be nice to be able
to click on one and hear a wave that is very strong only at that freq
(it is impossible to mute all the other harmonics completely though.)
​

Like I said, harmonics are not important.

6) It doesn't take into account the amount or configuration of the
coil within the pickup.
​

The coil characterstics are just an overall filter. That's not nearly
as intersting as what's going on here.

Let's see, there's more:

WarriorJoe7:
To dispel one myth, getting "near the bridge" DOESN'T get you all
the harmonics in the best pattern.
​

What could "all the harmonics in the best pattern" possibly mean? And
why is it a myth?

Getting near the bridge minimizes the nodes in the audio spectrum,
which is exactly why its used for guitar synthesizers; it keeps the
waveform consistant, and allows for better measurements on the
waveform.

WarriorJoe7:
It is counterintuitive that the thinnest sounding and looking pickup
picks up the most harmonics, but it's true.
​

Why is that counterintuitive? Wavelength is inversely proportional to
frequency. Tweeters are small, woofers are large. A wide pickup is
not going to be able to sense short wavelength action on a string.

In the screen capture examples you have there, the 4 inch wide pickup
starts losing high end around 300 Hz, the 1-inch wide pickup starts
losing high end around 2KHz, and the 0.1-inch wide pickup doesn't
start losing any highs until above the spectrum. So that's significant.

-- Don