Magnet length and changes in guitar pickup properties

[[email protected]]

Hi,

This is my first post and I don't know much about electronics, but I'm
in the process of making guitar (well, bass actually) pickups and
learning as much as I can. So, to my question...

How does the magnet length affect the pickup properties, where the only
difference is that all the magnets protrude below the bottom flatwork
(ie. the coil core is 6.2mm, leaving more than half the magnet exposed.
See examples below)? Keep in mind that I'm not referring to raising or
lowering magnets (eg. stagger in strat pickups).

For example, what is the difference in the pickup properties of example
1 and 2 below if they both have exactly the same dimensions except
those listed (both examples use alnico 5 magnets with a diameter of
9.5mm):

Ex 1

- Magnet length 28mm
- Wire covers 6.2mm from the top part of the magnet only. Magnet
exposed by say 1mm at the top flatwork and 20.8mm exposed beneath the
coil

Ex 2

- Magnet length 16mm
- Wire covers 6.2mm from the top part of the magnet only. Magnet
exposed by say 1mm at the top flatwork and 8.8mm exposed beneath the
coil

Regards,

Michael

 

[John Popelish]

Hi,

This is my first post and I don't know much about electronics, but I'm
in the process of making guitar (well, bass actually) pickups and
learning as much as I can. So, to my question...

How does the magnet length affect the pickup properties, where the only
difference is that all the magnets protrude below the bottom flatwork
(ie. the coil core is 6.2mm, leaving more than half the magnet exposed.
See examples below)? Keep in mind that I'm not referring to raising or
lowering magnets (eg. stagger in strat pickups).

For example, what is the difference in the pickup properties of example
1 and 2 below if they both have exactly the same dimensions except
those listed (both examples use alnico 5 magnets with a diameter of
9.5mm):

Ex 1

- Magnet length 28mm
- Wire covers 6.2mm from the top part of the magnet only. Magnet
exposed by say 1mm at the top flatwork and 20.8mm exposed beneath the
coil

Ex 2

- Magnet length 16mm
- Wire covers 6.2mm from the top part of the magnet only. Magnet
exposed by say 1mm at the top flatwork and 8.8mm exposed beneath the
coil

I am having trouble picturing where all the dimensions you talk about
are, physically. If you have some way to create a dimensioned sketch
of what you are talking about, and could either post it to some web
space (and tell us where to find it) or attach it to a post in the
alt.binaries.electronics.schematics newsgroup, I would be happy to
comment on your problem. I have put quite a bit of thought into
guitar pickup design.

 

[[email protected]]

Hi John,

That's great, I'll do that thanks.

Regards,

Michael

 

[BobG]

The pickups are a coil of wire around a magnet..... a solenoid..... it
has L and C.... L is proportional to number of turns and magnet
strength and permeability of magnet material. The tone of the pickup
has to do with the frequency of the resonant peak at 1/(2*pi*sqrt(L*C).
This is a very religious subject like moster speaker cables and angels
on the head of a pin, but one end of the spectrum of argument is that
the differences in pickups is basically this EQ setting built into the
coils.

 

[[email protected]]

Ok, I now have a diagram to illustrate the examples documented below.
Please disregard the original order of the examples and refer to those
below when referencing their image representation at this link:

http://users.bigpond.com/m.kat/html/tmp/mmpups1.jpg

So, what I would like to know is what is the difference in sound or any
other relevant property between the two pickups.

Ex 1

- Magnet length 16mm
- Wire covers 6.2mm from the top part of the magnet only. Magnet
exposed by say 1mm at the top flatwork and 8.8mm exposed beneath the
coil

Ex 2

- Magnet length 28mm
- Wire covers 6.2mm from the top part of the magnet only. Magnet
exposed by say 1mm at the top flatwork and 20.8mm exposed beneath the
coil

Regards,

Michael

 

[BobG]

My suggestion is buy an lcd multimeter that reads L and C. You can
measure the frequency response of the pickup by driving a loop of wire
with an audio oscillator (thru a resistor) and measure the resonant
peak of each pickup. My bet is on somewhere in the upper
midrange...5KHz
Tell us what they read! Is the measured peak in agreement with
f=1/2*pi*sqrt(L*C)??

 

[[email protected]]

Hi Bob,

Thanks for your comments. I haven't made the pickups yet and I don't
have those with similar specs that I mentioned, so I cannot measure
anything.

So, you're saying that the frequency at 5KHz will be impacted. Will ex1
or ex2 have more upper midrange? Will ex2 have less resistance and less
output than ex1?

Also, as I stated I don't know much about electronics and haven't
studied physics. Is L inductance? What's C?

Regards,

Michael

 

[John Popelish]

Ok, I now have a diagram to illustrate the examples documented below.
Please disregard the original order of the examples and refer to those
below when referencing their image representation at this link:

http://users.bigpond.com/m.kat/html/tmp/mmpups1.jpg

Thank you. I understand your description, much better, now. I assume
that these pictures are the view of the pickups, sighting along the
strings. So, if the strings were included in the drawings, they would
be seen in cross section, one above each magnet.

So, what I would like to know is what is the difference in sound or any
other relevant property between the two pickups.

Since permanent magnets have a low effective permeability, changes to
magnet shape and size have little effect on frequency response. The
dominant effect is on signal strength (or you might call it, sensitivity).

Ex 1

- Magnet length 16mm
- Wire covers 6.2mm from the top part of the magnet only. Magnet
exposed by say 1mm at the top flatwork and 8.8mm exposed beneath the
coil

Ex 2

- Magnet length 28mm
- Wire covers 6.2mm from the top part of the magnet only. Magnet
exposed by say 1mm at the top flatwork and 20.8mm exposed beneath the
coil

There may not be a significant difference, because the difference
between these two cases produce two effects that somewhat cancel each
other.

The signal is produced by a change in total flux passing through the
coil as the string vibrates. To visualize how this occurs, it may
help to draw a sketch shown from a vantage point from the right or
left of the one pictured, so the string runs across the top of the
pickup, and it is the coil that is in cross section.

The magnet sprays flux out one end and sucks it back up at the other
end. The metal string is a short cut for that flux as it makes its
way from one end of the magnet to the other, slightly increasing the
total flux the magnet can produce. As the string vibrates closer and
further (and to a lesser extent, from side to side) that total flux
varies, and that variation generates voltage in the coil in proportion
to the rate of change.

The second case generates more total flux, because it has a larger
magnet. But it also has a longer air path for the flux to pass
through to get from one end to the other, so the small short cut taken
through the string is a less significant part of the total path, so
the variations caused by vibration cause a smaller percentage
variation. The question to be answered by measurement or calculation
is whether a smaller percentage of a larger flux is a net gain or loss.

Often, pickups have some sort of iron or ferrite flux short cut around
the back side of the coil, to shorten the total amount of air the flux
has to be forced through, so that the small variations in this path
caused by the vibrating string constitute a more significant change in
the remaining smaller path.

Such high permeability materials, also increase the inductance of the
coil, altering its frequency response by reducing the high frequency
end. This may be more important for a lead pickup than a base pickup.

 

[BobG]

base? bass? There is extensive data about the mystical properties of
classic guitar pickup winds on the web. The freq resp graphs I have
seen all look similar but different... a 20dB low Q resonant peak in
the hi midrange... 6dB per octave rising characteristic going up in
freq to that peak. The freq and Q varies with coil wind, wire size, etc.

 

[Rich Grise]

Ok, I now have a diagram to illustrate the examples documented below.
Please disregard the original order of the examples and refer to those
below when referencing their image representation at this link:

http://users.bigpond.com/m.kat/html/tmp/mmpups1.jpg

So, what I would like to know is what is the difference in sound or any
other relevant property between the two pickups.

Ex 1

- Magnet length 16mm
- Wire covers 6.2mm from the top part of the magnet only. Magnet
exposed by say 1mm at the top flatwork and 8.8mm exposed beneath the
coil

Ex 2

- Magnet length 28mm
- Wire covers 6.2mm from the top part of the magnet only. Magnet
exposed by say 1mm at the top flatwork and 20.8mm exposed beneath the
coil

Do you have the magnets and coils? Can you pull a magnet out of a coil
and stick the other one in? Then why haven't you tried it yet? It's
called "empirical design", or "experimentation", depending on your
attitude. ;-) Try them one way, see how it sounds, then, leaving
all of the other settings untouched, swap out the magnets and see
how it sounds, and pick whichever one sounds better.

I've also seen high-end pickups where each magnet is threaded, with
a screw slot on top, so they're individually adjustable.

Then, of course, you can publish a big scientific paper and sell your
special "enhanced magnet" pickup at audiophool prices. ;-)

Cheers!
Rich

 

[BobG]

There is also the degree of freedom in how far from the bridge to mount
the pickup. The higher harmonics (trebley-er) are closer to the bridge.
I always wanted to put the pickup on parallel bars and slide it from
the bridge to the neck. You could position it right under the cool
sounding overtones.

 

[[email protected]]

Rich,

With all due respect. As I stated above already, I don't have the
pickups and have not started making them yet. This is an initial design
analysis decision. With almost anything, it's best to design before
building especially when the ground theory has already been researched
and documented. As Bob stated, what I'm dealing with is a solenoid,
magnets and wire: If wire is concentrated only on a small portion of
the magnet, what are the differences as to concentration over it's
length. Surely, this is not a topic so advanced that requires
"empirical design"...LOL.

What I'm asking is a theoretical question, which should be answered
quite easily I would think by people who have experience in this field
- it would be extremely worrying if this was not the case. John
provided useful information as did Bob. As a software engineer, I don't
bother with writing a single line of code till I have a solid design
for obvious reasons. Sure, there are times for experimentation, but to
experiment with things that are fundamental means there is a problem
with content retention or understanding of the topic. Obviously, I
don't have that knowledge, hence my post to this appropriate forum.

Regards,

Michael

 

[BobG]

I think you should dl a copy of switchercad from Linear Tech. The
examples directory shows how to get the freq resp of various filter
circuits. Borrow a guitar from 2 guys and measure the R L and C of the
pickups and 'model' it in switchercad as an L and C in parallel and the
R in series and compare the two freq responses. Then dial that freq
response into a stereo with a graphics equalizer and play some guitar
music thru it.

 

[Tony]

Hi. I just came across this thread, and I believe it has headed a
little off-track through a number of posts, so I'll try to sum up my
take on it as briefly as possible. The pickup does indeed form a
resonant RLC tank, which produces a classic underdamped 2 pole lowpass
filter - a peak then a rolloff (NO 6dB/octave rise). The height of the
peak can be anything from 0dB up to 20dB or more, depending on the
damping (pot loading, cable loading through a non-maxed volume pot,
plus iron and copper losses). The perceived tone is also a function of
the pickup-to-bridge distance, which determines the notch frequencies
(no "treble boost", just a comb filter), and of the magnetic circuit
geometry (which is quite significant, and AFAIK the subject that the
OP was asking about). Unfortunately, while the standard formulae
and/or modeling can easily determine the response of the lowpass and
comb filters (which can in any case easily be modified with Rs and Cs
if needed), they cannot as easily be used to model the pickup geometry
(and since the resonance can easily be adjusted externally, this is
the only insurmountable reason that pickups sound different).

The geometry can, however, be looked at fairly simply. The more
"magnetic coupling" is provided at the front and back of the pickup,
the more output will be delivered. But it is important that the top
pole is small or at least not uniform and parallel to the string, so
that the highest frequencies (above resonance) are not unnecessarily
suppressed. The back pole piece can be (and often is) extended to
provide an easy flux return from the string, more output, slightly
more inductance and a lower (less brittle, often better) resonant
frequency. Either way, it is still possible to use more turns to
achieve almost as much (a "hotter" pickup), and/or a shunt capacitor
directly across the pickup (lower resonance, no gain in output, but
less dependence on volume control position; perhaps a few hundred
pF?).

Now, to refer to the OP's specific question, protruding magnet
material can be viewed as the same as two or more magnets in series,
which is similar to a larger rear pole piece (can be a good thing),
but with more magnetic dipole sources. The extra dipole(s) on the back
don't couple through the coil, and so add virtually no output, but DO
couple to the strings and create string pull - in all, it slightly
degrades performance, compared to a short magnet with an iron pole
piece at the rear; more so when you consider the extra losses in an
Alnico magnet compared with, say, a thin sheet steel pole piece.

Years ago when I began to research this subject, Google yielded
nothing. I always meant to find the time to do some nonlinear finite
element analysis on it to try to quantify things, but never got around
to it. I'd be very interested if anyone has pursued this.

Tony

I think you should dl a copy of switchercad from Linear Tech. The
examples directory shows how to get the freq resp of various filter
circuits. Borrow a guitar from 2 guys and measure the R L and C of the
pickups and 'model' it in switchercad as an L and C in parallel and the
R in series and compare the two freq responses. Then dial that freq
response into a stereo with a graphics equalizer and play some guitar
music thru it.

Tony (remove the "_" to reply by email)

 

[YD]

Do you have the magnets and coils? Can you pull a magnet out of a coil
and stick the other one in? Then why haven't you tried it yet? It's
called "empirical design", or "experimentation", depending on your
attitude. ;-) Try them one way, see how it sounds, then, leaving
all of the other settings untouched, swap out the magnets and see
how it sounds, and pick whichever one sounds better.

I've also seen high-end pickups where each magnet is threaded, with
a screw slot on top, so they're individually adjustable.

Then, of course, you can publish a big scientific paper and sell your
special "enhanced magnet" pickup at audiophool prices. ;-)

Cheers!
Rich

Lots about it here:
http://users.chariot.net.au/~gmarts/pickups.htm

- YD.

 

[Tony]

Thanks - that's a good site covering a number of topics (albeit with
nothing on the OP's question).

Lots about it here:
http://users.chariot.net.au/~gmarts/pickups.htm

- YD.

Tony (remove the "_" to reply by email)

 

[[email protected]]

Hi Tony,

This is an excellent explanation (exactly the type of information I'm
looking for) and does actually provide a credible explanation of what
some people have claimed who have used this type of pickup. The picture
linked to above is of a single coil, but if it was a humbucker I assume
it would then further increase the string pull and further degrade
performance.

According to this information, the preferred option is to go with the
shorter pole pickup as it

- provides superior performance to the longer pole design
- is the cheapest magnet option
- is the cheapest build option
- will save on installation effort

If there is a preference to go with the longer pole pickup, the specs
are documented and further research can be conducted if required.

Regards,

Michael