YT YouTube2Text

YouTube2Text → Transcripts → T24 Pickup-Secrets neu

T24 Pickup-Secrets neu — Transcript

by Physik der Elektrogitarre · 10,122 words · 1,553 segments · language en · Watch on YouTube

Open in transcriber Summarize with AI Download SRT/VTT

Full transcript

0:08This time it's about the inner workings
0:11of the pickup, the secrets that define
0:14its sound. I began approaching these
0:19secrets back in the 60s. In the mid-60s
0:24, it started with what was then called
0:27radio tinkering. First a one-tube, then
0:32two-tube, then superheterodyne
0:35receivers, and on that occasion, I was
0:38given several spools of wire. The boy
0:42is tinkering with radios; we still have
0:44old wire in the basement, let's give
0:46him that. That was the foundation. Then
0:51, in the mid-60s, I wound my first
0:55pickup. I put a lot of wire directly
0:59onto the magnets, then shifted the
1:01magnets a bit, then the pickup was
1:04broken, and I soon lost interest and
1:07bought the stuff instead of winding it
1:10myself. The band came along shortly
1:15after, for which a whole series of
1:17amplifiers and speakers had to be built
1:20. Here I saw with interest the first
1:26oscilloscope from Radio Rim Munich, a
1:30kit called the Rock 7A. Later, that
1:36became an Amiga that served for many
1:39years and worked very well, along with
1:43a whole series of self-built measuring
1:47instruments to test tube and later
1:50transistor amplifiers. And since that
1:55time, the old spools of wire from the
1:5760s and 50s have been lying in the
1:59basement. I don't have the feeling that
2:03this is particularly good wire. Quite
2:05the opposite. I will no longer use
2:09these wires because you only risk the
2:11enamel insulation having become brittle
2:14, leading to shorted turns. It's better
2:19to buy new wire if you want to wind it
2:22yourself. Now, however, to the secrets
2:28of the pickup, which in principle—
2:31especially the single coil—is a very
2:34primitive component. There's the
2:37cylinder magnet, which creates a
2:39magnetic field in the air, and to this,
2:42the string is added. Arranged here
2:47above the magnet. Air is a poor
2:50conductor for the magnetic field. The
2:55string is a magnetically better
2:56conductor, so some of these magnetic
2:59field lines are pulled into the string.
3:02They reduce the magnetic resistance,
3:05and thereby the magnetic flux density
3:08increases in the magnet and also in
3:10this area of the air up here. Such
3:14magnetic field lines are actually
3:16closed; the fact that they end suddenly
3:18here is due to the drawing program,
3:20which didn't want to draw them all
3:21because it would otherwise get way too
3:23dense. So physics says field lines are
3:28closed. Yes, they are, even if they
3:31aren't in this illustration today. Now
3:35it doesn't help much if the string is
3:37stationary here. It vibrates, after all
3:38. And when it vibrates, it changes the
3:41magnetic field. And you often see in
3:44the literature that the entire field
3:46changes. Well, in a way yes, but it
3:51changes most strongly up here. And if
3:55we now attach a coil intended to
3:58convert this field change into
4:00electrical voltage, this coil is best
4:02placed up here and not down here. And
4:06we'll take a look at that right now in
4:07a measurement. Clamped in here is a
4:12small Alnico magnet. A measuring device
4:16measures the strength of this magnet,
4:19one might say, but since physics uses
4:21the term field strength, I have to be
4:24more precise here. What is being
4:27measured is the magnetic flux density,
4:30displayed in millitesla. And now let's
4:34slide the sensor onto the magnet here.
4:41Roughly 108 to 110 millitesla. And now
4:47I'll move a string closer to the magnet
4:49. And we notice that the magnetic flux,
4:56the flux density being displayed, is
4:59changing because this string has a low
5:02magnetic resistance to the magnetic
5:05field. And the lower the magnetic
5:08resistance, the greater the flux per
5:10unit area becomes. You shouldn't place
5:13the string completely on it, but yes,
5:18it works, and you can see a very strong
5:21flux change. Now the Hall sensor is
5:26placed on the underside of the magnet.
5:30And if I move the string closer now,
5:33practically nothing happens. I can even
5:38stick it directly onto the magnet.
5:41Minor changes. When the string vibrates
5:49, the magnetic field changes most
5:51strongly in the area up here between
5:53the string and the magnet. Wire
5:58windings located in the upper area will
6:01therefore detect more of this field
6:04change and generate a high electrical
6:07voltage. What is wound in down here is
6:12less efficient regarding voltage
6:15generation, but not entirely useless,
6:19because it still has significance for
6:22the inductance. We'll take a look later
6:26at what exactly the inductance depends
6:28on. So it's advantageous to have wire
6:33wound around the magnet up here, as
6:35close to the string as possible.
6:39However, these windings shouldn't go
6:42too far outward, because as you can see
6:45here, the alternating field is
6:48concentrated in this area. The field
6:51changes are drawn here. So for one
6:55half-wave, the magnetic flux goes in
6:56this direction, for the next half-wave,
6:59it then goes in the other direction.
7:02And this means that if a coil is very
7:05large, has a very large surface area,
7:08it captures both flux going up here and
7:12flux going down. And the balance
7:15becomes worse the greater the radius or
7:18the distance of the windings from the
7:21magnet. A pickup coil like the one Leo
7:27Fender developed for the Jazzmaster
7:29guitar is rather impractical from these
7:32points of view. It is best to
7:35concentrate the wire windings here in
7:38this area. These are the ingredients
7:43for a pickup. The black thing is the
7:47bobbin. Inside are six magnets,
7:52cylindrical, Alnico magnets. And wire
7:57is then wound onto this coil former.
8:01Not as thick as this one here, but
8:04typically 63 micrometers. Around that,
8:09a bit more or less, depending. We'll
8:12have to look at what influence the wire
8:15diameter has on the sound. Now, the
8:21question is what influence the position
8:23of the pickup winding has. You can see
8:27the position here, the winding is more
8:30on the left side of the coil former.
8:33There are only 20 turns, of course, not
8:35a complete winding. There is also only
8:40one magnet in the pickup, and a string
8:43will be attached here that vibrates,
8:47and then I'll measure the voltage
8:49induced by this. Then I will turn the
8:55coil around so that the winding is on
8:57the other side, and we'll see if the
8:59same voltage is generated. I've already
9:04introduced this test setup before. The
9:07shaker is at the bottom, the pickup is
9:10mounted hanging at the top. The string
9:14moves up and down. That is the shaker
9:21with the impedance head. And up here
9:29sits the pickup. In this case, the
9:33pickup coil only has 20 turns, so the
9:35generated voltage is very small. You
9:39have to measure selectively, e.g., with
9:41a DFT. I'm placing a marker there now
9:46so we have a comparison for the next
9:48measurement. The absolute voltage level
9:52consists of two parts. Full scale is
9:56-48 dB. And here in the display we have
10:02about -40, so -88 dB is the current
10:08voltage level. Now I've changed the
10:12setup. The coil is now at the other end
10:16of the bobbin, and we can see here that
10:19the level is lower, about 4 dB less.
10:23Yes, to be precise, at first it was
10:26-40.2. And now it's -44, so 4 dB less.
10:36Here we see the arrangement again in a
10:38schematic cross-section. A string over
10:41a magnet. The string is then moved in
10:45the axial direction. One time the coil
10:48winding is close to the string, another
10:51time it is further away. And in this
10:55case, with the small coil former, there
10:57is about a 4 dB difference. That is the
11:01better case, where more output is
11:02produced. If you take a typical
11:07Stratocaster pickup, the coil is longer
11:10there, and it makes an even bigger
11:13difference whether the winding is
11:16concentrated more on the part near the
11:19string or the part far from the string.
11:23Occasionally you see coils that have
11:26been wound a bit slanted. That can
11:30certainly be interpreted as purposeful;
11:33if there is enough space and you don't
11:36want to fill it to the brim, then it's
11:39useful to have the winding as close to
11:42the string as possible. This makes the
11:46pickup a bit louder. Now let's take a
11:51look at the individual parameters. You
11:53have to define the coil height here.
11:57There are two ways to do that. You
11:58could say this is the height, or that;
12:01I defined it this way. Winding up away
12:04from the magnets, or winding down there
12:07. That is the coil height. It's just a
12:12few millimeters, 3 or 4 mm, and a
12:15parameter that can be changed, which of
12:18course affects the sound, is the number
12:20of turns. A lot of wire is wound on,
12:27about a kilometer, between 5,000 and
12:3010,000 turns to cover a wide range. A
12:36typical Stratocaster pickup won't
12:38exactly be at the extremes. There are
12:42these magic numbers that the initiates
12:45know, regarding how many turns it was
12:47supposedly wound with. Let's see what
12:51influence the number of turns has. In
12:54the 50s, with early pickups, the number
12:58of turns was poorly controlled.
13:02Essentially, they just wound it until
13:04the pickup was full. The wire was also
13:08wound directly onto the magnets, which
13:12resulted in certain values. You read in
13:15the literature, for example with
13:17Duchossoir, it was between 8,000 and
13:208,700, though I wouldn't be surprised
13:24if pickups turned up that had a bit
13:27more or a bit less. It's simply unclear
13:30because they didn't pay close attention
13:32during production. Someone eventually
13:36took the average of those two limits
13:38and said it was 8,350 turns. And ever
13:44since, there are people who say that is
13:47the authentic pickup; it must have
13:508,350 turns, no more, no less. In the
13:5560s, counters were eventually
13:57introduced, though they initially had
14:00slippage and didn't measure the spindle
14:02revolutions all that accurately. I can
14:06easily imagine that there are
14:08significant tolerances there as well.
14:11If you wind wire with this number of
14:14turns onto a bobbin, you get an
14:17inductance that is proportional to the
14:20square of the number of turns as a good
14:24approximation. Double the number of
14:27turns, four times the inductance. We
14:30have the inductance in Henry for the
14:32air-core coil here. That means a coil
14:35without metal parts, without
14:37ferromagnetic materials. And depending
14:42on how high the winding was wound,
14:45there is a small difference here. I'll
14:48go back to that. The winding height.
14:52You can achieve a different height with
14:55a specific number of turns, for example
14:58by using wires of different thicknesses
15:01. We'll take a look at this influential
15:04parameter as well. So the number of
15:08turns has a major influence, and the
15:11winding height has a minor influence.
15:16Now, however, magnets need to go into
15:19the coil, and that brings us up here.
15:22Magnets, specifically Alnico 5 in this
15:25case, which the Stratocaster had.
15:27That's how you read about all Q5
15:29magnets in the literature. Such magnets
15:32conduct the magnetic field, the
15:34magnetic flux, better than air. That is
15:37why there is a higher inductance of a
15:41good two Henry. Again, the rule applies
15:45: as the number of turns increases,
15:47inductance increases. But we can
15:51already see that the difference in
15:52winding height is now minimal. And then
15:57you have to ask yourself, this is 0.7
16:00mm, which is quite significant. Is it
16:04still worth talking about this
16:06difference? You could say yes, but then
16:10you have to consider other small
16:13differences as well. In the 1980s,
16:18something else changed. That was when
16:22the wire was no longer wound directly
16:24onto the magnets. Instead, there was a
16:27bobbin into which the magnets were
16:30inserted, and the wire was wound onto
16:33this plastic bobbin. As a result, the
16:37entire winding became slightly larger
16:39with the same number of turns, and a
16:41larger area means higher inductance.
16:45For all these curves, I assume these
16:47are average values. There is variation,
16:52and one might ask about the 50s, 60s,
16:5480 s—what happened in the 70s? And
16:58you can find literature on that as well
17:00. I suspect someone just had a pickup
17:04from 1968, measured the inductance, and
17:08said that late 60s ones have so and so
17:11many Henry, but it all varies. I have
17:16Stratocaster pickups from the 70s, and
17:18they are at 2.2 Henry. So, you have to
17:23expect variations in this range, and
17:26for the early pickups wound in the
17:29first few years, the variations were
17:32probably huge because the turn count
17:35wasn't controlled. So, here we have the
17:40coil again. The somewhat narrower range
17:45for a Stratocaster pickup is between
17:48about 7400 and 8800. 7400 here, and
17:548800. And if someone says, "But I have
17:59one with 8900," or perhaps even less.
18:03Well, anything is possible. But let's
18:08take this range and see what changes if
18:11you wind a different amount of wire on
18:14a pickup. I have two charts here, this
18:19is the transfer function of the pickup.
18:22Now, the guitar electronics are added
18:25to the pickup. These are the two
18:29potentiometers, which typically have
18:31250 kilo-ohms. And the amplifier, which
18:35I’ve assumed to be one mega-ohm. The
18:39cable and amplifier input then form a
18:42capacitance for two cases. In one case,
18:45this capacitance is 400 picofarads; in
18:47the other, it is 900 picofarads. We see
18:50the transfer function here for
18:52different turn counts. Now, with an
18:56increasing number of turns, the pickup
18:58gets a little louder. We notice about
19:02one and a half dB here. And the
19:05resonance decreases because the
19:08inductance increases. The greater the
19:10inductance, the lower the frequency of
19:12the resonance becomes. And you can hear
19:16a difference like that in a direct
19:17comparison. It’s not an extreme
19:22difference, but anyone who listens
19:24closely will notice that one sounds a
19:26bit louder. Whether you notice that
19:29while playing is another question. But
19:32due to the lower resonance, it sounds a
19:33bit more mid-focused. Now, we can also
19:36see that I get a resonance shift from
19:39different cable capacitance. We are
19:42here, let's take the blue curve. At 4
19:45kHz and here the red curve at 2.89 kHz,
19:49and with longer cables it shifts even
19:53further down. Significantly lower
19:58capacitance becomes difficult. You can,
20:01of course, install a preamplifier
20:02directly into the guitar and provide no
20:04additional capacitance. That moves the
20:07resonance even further up. But those
20:10are rather exceptional cases; with 150
20:13picofarads at the amp input and then
20:16250 picofarads, you already have a very
20:19low-capacitance cable. 3 meters; yes,
20:23if you take 6 or 9 or even more meters
20:25and have cables with 150 picofarads per
20:28meter, which certainly exist. There are
20:31also some that have much more, then the
20:34resonance just drops significantly. But
20:38is it then the specific pickup that
20:40creates the sound? The small difference
20:44comes from the pickup. It is not
20:47insignificant; you have to take it into
20:50account. But cable capacitance also has
20:54a major influence. So, that was the
20:58number of turns. The more turns, the
21:01louder and the lower the frequency of
21:04the resonance becomes. Then it's about
21:06the enamel coating. Copper wire is used
21:10, almost always. The people who say it
21:14has to be silver wire haven't
21:15understood many things. I don't even
21:17want to go into this nonsense any
21:19further here. The copper wire must be
21:21insulated so there are no short
21:22circuits. What kind of copper it is
21:26doesn't matter at all. It is usually
21:29conductive copper, which consists of
21:3199.9 percent copper. Anyone who wants
21:35to find out about this,
21:36www.kupferinstitut.de, there is info on
21:39copper there. Dear guitar builders,
21:42don't just fantasize out of the blue.
21:45Take a look at the documentation on
21:47this. Then there is the enamel coating
21:49here. It can have a different thickness
21:53, and if such a copper wire is very
21:56thickly coated, the winding height
21:58increases for the same number of turns,
22:01and the inductance becomes greater.
22:05Therefore, the coating thickness has a
22:08small influence. Due to higher
22:12inductance, the resonance drops a
22:15little bit. But once again, and I will
22:18say it many more times, the cable
22:20capacitance has a much greater
22:22influence. Of course, you notice if you
22:26take a thinly coated and a thickly
22:29coated wire and compare them with
22:31otherwise identical parameters, you
22:33will notice a small difference. But if
22:36you change the capacity a little bit,
22:38you also notice a difference. That can
22:40become much larger than these effects.
22:44For a guitar pickup, simple thinly
22:46enameled copper wire is sufficient.
22:49However, manufacturers also offer wire
22:51with a thicker enamel coating. This is
22:55wire sometimes used in transformers
22:57where there are hundreds of volts, so
22:59the insulation breakdown strength must
23:01be correspondingly high. With a guitar,
23:05we might have maybe 2 volts at most.
23:08That is not a criterion at all; the
23:10thin, simply enameled wire is actually
23:12sufficient. Then another argument is
23:15brought up. Yes, but they were
23:17different enamels in the early days and
23:19later on, and then the enamel changed
23:22again later, and the first one was a
23:24bit thicker and later they became
23:26thinner and so on. Yes, but as I said,
23:30these are minimal effects that partly
23:32still concern the capacitance. We will
23:35look at those later. There are small
23:38differences there too. The material
23:40also has a small influence on the
23:42capacitance regarding its dielectric
23:45properties. That is coming up. Let's
23:49keep this in mind. The enamel makes up
23:52about 10 to maybe 20 percent of the
23:55diameter at most. Thicker enameled wire
24:00leads to a greater coil height, thus
24:03slightly increased inductance. This
24:07shifts the resonance a little bit.
24:10Capacitances, cable capacitances, can
24:12do that, but also let's go back to this
24:19data again. Of course, there are
24:22differences. I am not saying that these
24:24pickups all sound the same. We have
24:26significant differences in inductance
24:30here, but you can already see here the
24:34winding height, if we take 10%from 3.4
24:38to 3.7 or 3.8 it's a minimal difference
24:43. Here we already have about, I don't
24:44even want to calculate, 20%. So these
24:47are very small effects that occur here.
24:53Then it's about the type of winding.
24:56Layered or scatter winding. First of
25:01all, you might construct a picture that
25:03looks like this. All wires lie directly
25:06on top of each other in the bobbin. Of
25:09course, it wouldn't look like that. It
25:12is an ideal that the wires lie in the
25:16gaps. By hand, you can't manage that at
25:20all, though a winding machine, if very
25:22well adjusted, might achieve such a
25:24winding. And we already notice here,
25:28the distance, the air here is a bit
25:30larger than here and the winding
25:32becomes a bit larger with such an
25:34arrangement, but that can't be done.
25:38That would be an ideal case, the
25:40smallest possible volume. And it won't
25:45look like that, but if you wind by hand
25:48, for example, then these wires somehow
25:51lie here in the space on top of each
25:53other, crossed over. That's the scatter
25:56winding. Layered would be like this.
26:00Everything done by hand has larger gaps
26:03with the wire running in some way. Does
26:07it have an influence? Yes, partly due
26:10to hand-wound coils and a slightly
26:13higher winding height. I mentioned a
26:17small influence at the beginning. And
26:19the second is the capacitance. If I
26:22move the wires further apart, there is
26:25less capacitance, which is also
26:27important for the resonant frequency.
26:30There is a picture coming up. Let’s
26:33take a look at something from the 50s
26:34here. The book "Schröder, Electrical
26:37Communications Engineering" from 1959.
26:42He already showed pictures regarding
26:45winding capacitance, normal layer
26:49winding, one layer neatly wound over
26:52the other, or what he calls a random
26:56stepped winding. You start in one
26:59corner and then wind as far as possible
27:02over to the other corner. There, the
27:06beginning and end of the winding are as
27:09far apart as possible, which makes the
27:11winding capacitance a bit lower.
27:15Sectional windings are more common in
27:18transformers; that’s almost standard
27:20there, but rare for guitar pickups.
27:23There may be some that use this
27:26principle to reduce capacitance. It is
27:30not the standard. And finally, there is
27:37the bobbin, with or without. In the
27:42early years, Fender didn't have one;
27:44they wound the wire directly onto the
27:47magnets. If you then try to shift the
27:51magnets axially, the pickup might be
27:54destroyed because those very thin wires
27:57will break. They are less than a tenth
28:00of a millimeter, sometimes thinner than
28:02a hair. Nominal thicknesses are, for
28:05example, 0.063. mm, 63 micrometers
28:10copper diameter. It is very thin,
28:14breaks very easily, and if it rests on
28:17the magnet and you move a magnet, then
28:19it's broken. You should definitely not
28:22do that with old pickups. With or
28:27without a bobbin, we can see it in
28:28cross-section. Up here is the magnet
28:30side. With the old pickups, there are
28:34two flanges here, plastic plates that
28:36are drilled, pushed onto the magnet,
28:38and the winding simply goes in between,
28:41and it works. And for the others, a
28:45bobbin was pushed in between so that
28:48you could, for example, shift the
28:51magnets, and that in turn makes the
28:54winding, the winding height, and the
28:56area larger. With the same number of
29:00turns, the winding moves further away
29:03from the magnets, thus having a larger
29:05area and resulting in higher inductance
29:08. The influence is measurable, though
29:12not particularly large. Now for the
29:17winding capacitance, which is the
29:19second parameter for resonance. Both
29:23influence the resonance in the same way
29:25. If we double the capacitance, the
29:30resonant frequency drops to 0.7. And if
29:34we double the inductance, the same
29:36thing happens; the resonance also drops
29:39to 0.7. 1 divided by the square root of
29:412, to be precise. I have already posted
29:45several videos on this on the webpage.
29:48More theory and more practice on
29:51pickups. For those who want to know
29:54more, you can find additional
29:55information there. Now I have taken a
29:58pickup here, a Stratocaster one, that
30:01is not loaded. Transmission when the
30:04pickup is running idle. Idle means it
30:07is just terminated with a very high
30:09impedance. Its internal capacitance is
30:1290 picofarads in one case, 140
30:14picofarads in the other, and we can see
30:17that the increased capacitance pushes
30:19the resonance frequency down. I don't
30:23think many guitarists play with such a
30:26setup, terminating the pickup with a
30:28high impedance without potentiometers
30:30or cables, going into a direct
30:32impedance converter. A very special
30:36sound with that very high resonance.
30:39Typically, you attach a guitar cable,
30:41add two potentiometers, and then it
30:43looks like this, for example. These two
30:48thick curves show what happens when you
30:50connect a 400 picofarad capacitive load
30:52. Winding capacitance of 90 and 140
30:57picofarads, as before, 140 picofarads
31:00is the blue curve, 90 the red, plus 400
31:03picofarads of cable capacitance, and
31:06the amplifier is included too. And
31:09these 111 kilohms result from the two
31:11potentiometers in the guitar and the
31:13amplifier's input resistance. And now
31:16you can see that the difference in
31:19pickup capacitance is becoming very
31:21small. It is still there, but not much
31:25is happening anymore. These two thin
31:28curves here show what happens when you
31:31load the pickup only with the
31:33resistance—meaning the potentiometers
31:37are attached, but no cable yet—or an
31:40amplifier connected with a very, very
31:43short cable. That is what it's like
31:48when only a very high-impedance
31:50resistor is attached. And these two
31:54curves now show what happens when we
31:57only add the resistors. It has no
32:01significance in practice. This is
32:03getting more interesting now. You can
32:07see two different winding capacitances,
32:09and there is a clear difference. 90
32:12picofarads versus 140 picofarads. A lot
32:15has changed in the developmental setup,
32:18but the cable is clearly dominant. 900
32:23picofarad cable—that is the
32:25difference that remains. I would be
32:28interested to know who can hear that in
32:31a blind test. I fear that nobody can. I
32:34can easily imagine a guitar builder
32:37connecting the pickup directly to an
32:40amplifier without a long cable and
32:43saying, "Oh yes, I hear some
32:46differences." But in practical
32:48operation, it is completely meaningless
32:51. Here again are the insulation
32:54materials of the enameled copper wires
32:57used for the winding. Things changed
33:00first in '65 and then in '74. And you
33:05read that early insulations were
33:08thicker and later became a bit thinner,
33:11which has an influence on the winding
33:13capacitance. But they are low, very low
33:16, because the dominant capacitance is
33:19the cable itself, and many guitarists
33:22simply say: "I need this capacitance to
33:25bring the resonance into the typical
33:28Fender range." That, however, is a
33:30matter of taste. You shouldn't tell
33:32anyone what to do there. Perhaps
33:35someone wants the resonance at 6, 7, or
33:388 kHz to create a very treble-rich
33:40sound, while another might play with
33:43distortion and say it's too scratchy,
33:45so I'd rather shift the resonance down
33:48to 2 kHz. Yes, go right ahead; it's
33:51subjective, and you shouldn't interfere
33:54with that. I have compiled some of the
33:57parameters for pickup winding here. The
34:00wire is copper, made of standard
34:03electrical copper. The diameter is
34:06approximately 0.063 mm. There are some
34:10pickups where the wire is a bit thinner
34:13, but it is generally not thinner than,
34:16let's say, 0.05 mm. So the range
34:19between 0.05 and 0.063 is common. The
34:23insulation accounts for about 10 to 20%
34:27. Thicker insulation means the coil
34:30becomes a bit larger, resulting in
34:32slightly higher inductance; the number
34:34of turns has a major influence on the
34:37inductance. The winding can be in
34:41layers, arranged by machine, or
34:43hand-wound scatter-wound with or
34:46without a bobbin. And then there's
34:49another influencing factor. Many
34:51bobbins are vacuum-potted in wax so
34:55that the pickup is more compact, firmer
34:59, and reacts less to mechanical
35:02excitation. Because otherwise, it can
35:06happen that a sound wave hitting the
35:08guitar makes the guitar vibrate, and
35:10this vibration is transferred to the
35:13pickup, causing the wire windings to
35:15start vibrating. They shouldn't do that
35:17. That can cause unpleasant squealing
35:20noises, which we don't want. That is
35:22why many pickups are waxed. It changes
35:26the pickup's capacitance slightly.
35:29There are also good pickups that get by
35:32without wax. So, a small influence on
35:37capacitance and an influence on
35:39sensitivity to acoustic excitation. And
35:45then another influencing factor: the
35:47cover; if it's plastic, it's not
35:49electrically or magnetically active.
35:53Electrically almost not at all, and
35:55magnetically not at all. Metal covers,
35:58on the other hand, can lead to
36:00significant treble damping. There is a
36:02separate video on that. The parameters
36:07listed here have an influence on the
36:10inductance and capacitance of the
36:13pickup. An influence that is quite
36:16easily achieved. If you wind more on it
36:18, the inductance becomes greater. That
36:21is no secret. The costs for that are
36:24also very limited. Whether you put
36:286,000 or 8,000 turns on it, the winding
36:30machine just runs a moment longer. It's
36:34not a big deal. Both values influence
36:38the resonant frequency, but it is also,
36:41and even more strongly, influenced by
36:44the cable and the cable capacitance. or
36:48perhaps the capacities you switch into
36:51the circuit. Therefore, the influence
36:54of these parameters on inductance and
36:57capacitance is a fact. Yes, but it is
37:00also determined by the cable and by
37:03capacitance. You shouldn't
37:06overemphasize these parameters, and
37:09above all, you shouldn't use them to
37:12justify why a single-coil pickup costs
37:15145 euros just because it has a special
37:18number of turns. Right. New topic,
37:23magnetics. There is a separate video on
37:26that, too. For single-coil pickup
37:30magnets, various magnetic alloys are
37:35used. Alnico 3, 4, 5, 6, 8, and 9 are
37:39the most common ones. There are also
37:43pickups with ceramic magnets. That is a
37:46different area. I'll leave that out
37:47here. And if you don't know much about
37:51magnetics, you might think, oh yes,
37:52they are different, this one is way up
37:54high, so it must be especially good or
37:56loud. But these values here, which are
38:01the magnetic flux density measured in
38:04Tesla, mean nothing. In the middle of
38:10the magnet, we encounter these values,
38:13but at the end of the magnet, where it
38:16transitions into air, at the pole piece
38:19where the string is just a few
38:21millimeters away, the flux density has
38:24already dropped to these small values.
38:29And there we can see that the
38:31differences still exist, but they are
38:33no longer that large. So, one must ask:
38:37what happens if I swap the magnets in a
38:39single-coil pickup? Though it must be
38:44said that if it’s an old pickup with
38:46the wire wound directly onto the
38:48magnets, you can't just pull them out
38:51and replace them. If you tried that,
38:54you would destroy the pickup. If it has
38:57a bobbin—which Fender introduced
39:00around 1980—then you can pull the
39:03magnets out and replace them with
39:05others. Does that do much? Well, for
39:11one thing, you have to consider that
39:13the term "Alnico 5" is not unique. It
39:17is a group designation. There is a
39:20whole series of different Alnico 5
39:23magnets whose curves lie somewhere in
39:26here and which, as you can see, might
39:30even overlap with other magnets. There
39:35are overlaps in this area as well. You
39:39won't find the exact composition of
39:42Alnico 2, 3, 4, or 5 at any pickup
39:44manufacturer; if they mention it at all
39:47, they just say "Alnico 5 for that
39:49bluesy sound." But it isn't specified
39:53which Alnico 5 subgroup is being used.
39:57Often, the manufacturer probably
39:59doesn't even know for sure themselves.
40:03So, the magnets are different, and that
40:07can lead to a slightly different sound.
40:13However, the differences are much
40:15smaller than one would expect. I have
40:18plotted some transfer functions here.
40:22That leaves out Alnico 8 and Alnico 9.
40:26They are actually quite different. They
40:30have a significantly lower permeability
40:32, so the inductance of these pickups is
40:35lower, which causes the resonance
40:37frequency to shift slightly higher
40:39under otherwise identical conditions.
40:43You can hear that. But whether it makes
40:46sense to use Alnico 8 or Alnico 9 for
40:49that reason, in my assessment, I
40:52wouldn't do it. These magnets are also
40:56quite strong and therefore pull harder
40:58on the string. And with a Stratocaster,
41:03you have three pickups and three such
41:05strong magnets per string. I see no
41:10significant advantage; if you place
41:13great importance on the resonance
41:15frequency being a bit higher, which is
41:17certainly legitimate, then you just
41:20have to use a cable with slightly lower
41:22capacitance, and if it needs to go even
41:25higher, you would have to install
41:27active electronics in the guitar. That
41:30is no longer an issue these days. Then
41:33you can shift the resonance frequency
41:36anywhere in the range you want and
41:39don't have to resort to somewhat
41:42strange pickup magnets. For me, the
41:46best magnet material is Alnico 5, but
41:49Alnico 2, 3, and 4 are also suitable. I
41:53have marked the biggest difference here
41:56in blue and black. Alnico 2 and 3 are
42:00indistinguishable, and with Alnico 3
42:02and Alnico 5, there are small
42:04differences in resonance, i.e., in the
42:07treble, if you don't change anything
42:09else. And there is one more thing.
42:13Alnico 5 is a bit louder, maybe by 2 dB
42:16, but the volume of the pickup, or its
42:19transfer coefficient to be more precise
42:22, is normalized to 0 dB here. The
42:26Alnico 5 curve is now a bit higher, and
42:29therefore the pickup puts out a bit
42:32more voltage. But that's not a major
42:35criterion anymore today. Amplifiers all
42:37have sufficient gain reserves. I
42:42consider Alnico 5 the best compromise,
42:46but to change a pickup just because it
42:49only has, in quotation marks, Alnico 3
42:52installed—the differences are really
42:55very, very small. What you read in
43:00advertisements is generally not
43:02physically justifiable. "With Alnico 2,
43:06because of the weak magnet, the tone
43:08literally collapses." Yes, some editor
43:12in a test magazine once had a guitar. I
43:16can certainly imagine that he was of
43:18the opinion that the tone collapses;
43:20that does happen. And then the Alnico 2
43:23magnet was to blame. While another
43:27writes that the weaker Alnico 2 magnets
43:30promote sustain, which is a
43:31contradiction in itself. "The pickup
43:35does not lose highs thanks to its
43:38Alnico 2 magnet.""Pickups with Alnico 2
43:41magnets have fewer highs, are quieter,
43:43rounder, and somewhat less dynamic."
43:47These are statements from trade
43:49magazines where either guitars were
43:52tested or wisdom about magnets was
43:55being dispensed. Yes, that keeps
43:59contradicting itself. Stronger magnets
44:02produce fewer highs. Alnico 5, the
44:04stronger magnet, sounds more brilliant.
44:08That is certainly conceivable in
44:10individual cases. He is holding a
44:12guitar that sounds brilliant. Oh, what
44:14kind of pickups are those? Oh yes, they
44:16all have Alnico 5 magnets. What kind of
44:20cable is it? What about the cable from
44:23a year ago that gave another guitar
44:25with Alnico 5 slightly fewer highs?
44:29None of this is scientifically sound.
44:32Bluesy basic character with a
44:34pleasantly rounded note. Great, you
44:36can't argue with that. What is a bluesy
44:40basic character? Clean, slightly
44:42distorted, crunch, more distorted,
44:44heavily distorted, pleasantly rounded
44:47note. Yes, well. Clearer sound, wirey
44:51twang, punchier bass, fast response,
44:54and slightly less differentiated
44:56reproduction. All of that is marketing
45:00nonsense. It cannot be scientifically
45:02substantiated. Alnico 8, the pickup
45:06also has a high output with heavy
45:09picking, I can imagine, with low
45:12compression. That is nonsense again.
45:14The pickup doesn't compress; that's the
45:17amplifier. If I distort something in
45:21the amp, I get compression in the tone
45:24because the higher levels are clipped
45:27and compressed, but it’s not the
45:30pickup itself that compresses. Louder
45:36pickups have more sustain. Yes, I can
45:39imagine that. Louder pickups drive the
45:43amplifier more, and if the amplifier is
45:47pushed into clipping, the sustain
45:49increases. The more I overdrive, the
45:53longer the sustain becomes. But that
45:55primarily has to do with the amplifier.
46:00And one shouldn't write that the
46:02pickups possess sustain. At best, one
46:06should say the combination of pickup
46:09and overdriven amplifier, and if you
46:13turn down the gain on the amp, that
46:16sustain gain is gone again. Alnico 8,
46:21the greater magnetic force. Yes, that's
46:23true. Results in sustain loss. Yes,
46:26please. What is it now? And it's always
46:30nice to think back to the beginning.
46:33The developer of the Gibson humbucker
46:35once said in an interview, we bought
46:37whatever was available at the time.
46:39They were all good magnets. There
46:43wasn't much distinction made between
46:45Alnico 2, 4, or 5; whatever was
46:47available was bought and installed. And
46:51it all sounds good. To avoid getting
46:55too theoretical, I once built a guitar
46:59with four pickups installed that have
47:02identical coils—I even measured them
47:06to ensure the coils are truly the same
47:10—but equipped with different magnets,
47:13four different magnets. You can turn
47:18this disc here, this turntable, and
47:22move different magnets under the
47:24strings. And there's a magnetic switch
47:28on every pickup that only turns the
47:30pickup on when it's positioned exactly
47:33vertically under the strings, so that
47:35only one is active at a time. And with
47:40that, you can change the pickups and
47:42magnets very quickly to hear what it
47:45sounds like when you play Alnico 2,
47:47Alnico 3, or Alnico 5. And you realize
47:52that with Alnico 2 and Alnico 3,
47:54they're indistinguishable, and Alnico 5
47:57actually sounds just the same. It still
48:01always depends on how hard you strike
48:04the strings. You can't strike them with
48:07the exact same force down to a tenth of
48:09a dB. That's why I also did experiments
48:12here with a pendulum that always struck
48:15the string in exactly the same way to
48:18get some measurement results. The
48:21fourth pickup is a bit special. It is
48:26described in more detail in a
48:28publication, and I will also make a
48:30video here in which I would like to
48:33present the sound of these four pickups
48:36. So, now there is a bit of special
48:42advertising, as often in this series.
48:45Stay tuned. In pickup advertising, I
48:49have never seen meaningful frequency
48:51responses, impedance frequency
48:53responses, or transmission frequency
48:56responses—never, from a single
48:58manufacturer. At best, you find
49:02information on DC resistance, which
49:04says practically nothing, but no data
49:07on the winding or the frequency
49:09response. First guess: manufacturers
49:14think that guitarists can't do anything
49:17with that information anyway. Second,
49:22more likely guess: manufacturers are
49:24afraid that the tiny differences would
49:27be revealed, and they wouldn't be able
49:30to offer a range of different pickups
49:32that supposedly sound completely
49:35different. Third, even more likely
49:39guess: manufacturers and retailers
49:41can't even perform such measurements.
49:44They don't even know what they are
49:46doing. I once sent a broken vintage
49:49pickup to one of them. I asked them to
49:53rewind it and wrote: "Please also
49:55include the impedance frequency
49:57response." The pickup came back, was
50:00rewound, it worked, and there was a
50:02note included. "We don't measure
50:05impedance frequency responses; we
50:07aren't that technical." I fear that is
50:11the rule when someone manufactures
50:14pickups. The physical principles are,
50:17as I suspect, unknown to them. Instead,
50:22you read nice marketing copy about our
50:25'50s pickups sounding like Buddy Holly,
50:28Billy Gibbons, or Ron Wood, who, as we
50:31all know, sound identical because they
50:33all have that' 50s pickup sound. Well,
50:38with Buddy Holly, yes, we can agree he
50:41couldn't have played with a '60s sound;
50:44he didn't live to see the' 60s, unlike
50:47his other colleagues here. Quite
50:51interesting. There is a great video by
50:54a Billy Gibbons technician who explains
50:57how the great master has all his
50:59guitars routed out so they aren't too
51:02heavy, and how an equalizer curve is
51:05saved for each guitar so it sounds the
51:08way it's supposed to, and not the way
51:10it would sound if there were no
51:13equalizer in between. It is interesting
51:16what the heroes of this world do. How
51:20they modify their guitars, what they
51:23actually play with, and what
51:26advertising wants to whisper in our
51:29ears. What strings did the gentlemen
51:33play in the 50s? Don't you also need
51:38the information on which strings those
51:40were if you want this Buddy Holly sound
51:43? Were they pure nickel wound 12-gauge
51:47sets? Were they flatwounds? That
51:51influences the sound, and above all,
51:53what cable did the gentleman play with?
51:56What was the cable capacitance? That
51:59has a huge influence on the resonance
52:02frequency. But you learn nothing about
52:05that, because I had a loudspeaker. In
52:08one place I read that the recording was
52:10made through a 15-inch speaker. Great.
52:14And where was the microphone? Does the
52:17choice of microphone not have an
52:18influence on the sound? Of course it
52:20does. How was the recording processed
52:24afterwards? One can assume that in the
52:2750s there wasn't that much electronics
52:29available for processing. The band
52:33played, and they placed a microphone in
52:35the room. Everything was still in mono
52:37back then. For the younger ones, you
52:40only have one channel, and the same
52:42thing comes out of the left and right
52:43sides of the headphones. So, very early
52:46beginnings, but still, with our 50s
52:49pickups, you will sound like Buddy
52:51Holly, Billy Gibbons, Ron Wood, and
52:54whatever the other heroes are called.
52:58Put € 500 on the table, and you’ll
53:00get those 50s pickups too. Guitarists
53:05are encouraged to replace the pickups
53:09in their guitars for sometimes a lot of
53:12money because it supposedly sounds like
53:16one of the guitarists mentioned in the
53:19ad. Nonsense; do you actually get the
53:23fingers included too? If I want to
53:27sound like Billy Gibbons now and buy
53:29these amazing pickups from that one
53:32manufacturer for € 500, are the
53:35fingers important? Well, everyone has
53:40to decide that for themselves.
53:44Regarding microphone position, I have
53:46an example here, because the guitar is
53:49certainly an essential part of music
53:52production and, of course, as I already
53:55mentioned, the fingers—the guitarist
53:57also has something to do with it.
54:02Strings, cables, speakers. Today
54:06we’re making a big fuss about the
54:09microphone: which one, exactly where in
54:12front of the grille cloth, then a
54:14condenser mic a bit further back, and
54:17maybe a third or fourth mic to shape
54:20the sound. I’m presenting a very
54:24simple recording here: a Telecaster
54:27plugged straight into a Fender, with an
54:29SM57 in front of it that I’m moving
54:32back and forth. It sounds distinct when
54:35the microphone is directly in front of
54:38the speaker's center—pretty bright.
54:40If that's too much, just roll off some
54:42of the highs. I haven't processed this
54:44recording any further. I know full well
54:47that you wouldn't use this kind of raw
54:49track in a real production. I just
54:51wanted to demonstrate a few things.
54:54First, how does the sound change when
54:57the microphone is moved around? There
55:00are already a few pages on this in the
55:02physics of the electric guitar. It's
55:04nothing new. And what happens when you
55:07use a different speaker? Let’s have a
55:09listen to that. In this demonstration,
55:41only the speaker was swapped out. In
55:43both cases, it’s the same amplifier,
55:46same signal, same settings. We can see
55:50the corresponding spectral analyses
55:52here. The red curve shows the spectrum
55:57when the mic is on-axis directly in
55:59front of the center, and the blue curve
56:02is when the mic is angled off-axis.
56:06I’ve normalized them so the curves
56:09align here. We see in the first case.
56:13The volume changes, as I said, it’s
56:15balanced, but most importantly, the
56:17high frequencies change. For an
56:19acoustician, that’s no surprise. But
56:22there’s an interesting detail: the
56:26highs don't just drop off evenly; the
56:29spectral composition changes in this
56:32range. And up here is the typical
56:36Fender pickup resonance. The sound
56:40changes in the details, not just a
56:43blanket reduction of the highs.
56:47Consequently, it is important where the
56:50mic is placed, and the type of mic—
56:53whether it's omni or cardioid, how many
56:57mics you use—that’s all clear, but
57:00even with simple single-mic recordings,
57:03you can see the spectral shift. Here is
57:07the peak. If we turn to the side, the
57:09peak disappears. Quite different.
57:12Let’s go to the Celestion. There, we
57:15see a different behavior. You could say
57:19there is a more even high-frequency
57:21roll-off in this range. These top-end
57:26highs, which are still at 6 or 7 kHz
57:28here, are missing. It drops off much
57:31more smoothly and clearly here. I
57:34won’t comment too much on the rest of
57:36the range down here. The speaker
57:38cabinet plays a role there as well. And
57:41the fact that there's nothing here is
57:43simply due to the chord being played.
57:45It has no partials there. In the highs,
57:48you notice that the loudspeaker plays a
57:50role. Yeah, who's surprised by that?
57:52And the microphone position plays a
57:55role. And all that would also be
57:59important if one says, here is the 60s
58:03or 50s sound. Another example: 3 €
58:08and you sound like Clapton. Throw the
58:12pickups out of your Stratocaster and
58:15buy the 60s pickups for 3 x 150 € and
58:18you sound like Clapton. That Clapton,
58:23he's a very poor guy. He only has one
58:26single guitar with which he created his
58:29sound. That's why he always sounds the
58:32same. That gets almost a bit boring.
58:34But if you should like this boring,
58:38always identical sound, then buy these
58:41special pickups and yes, you sound like
58:45God in the 60s. Now let's take a look
58:49at what's actually inside such a single
58:52coil? You need six alnico magnets. If
58:56you buy them retail, you pay maybe 1.50
58:59each. That means you've already got 9
59:02€. The dealer buys it much cheaper,
59:05of course, but let's stick to the
59:07retail price. Then you need a coil
59:10bobbin, if you want one, if you don't
59:13wind the wire directly onto the magnets
59:15. If you wind directly onto them,
59:18that's gone and then you only need wire
59:20. About a kilometer, pricing is
59:24difficult there, as the price per
59:27kilometer goes down when buying in
59:29larger quantities. However, if you only
59:33want to wind one or three pickups, you
59:36might not want to buy 30 or 60 km of
59:39wire. And if you only buy in small
59:42quantities, you pay more, because of
59:44the extra costs too. You might have to
59:47search, if you want to do it yourself,
59:50to find a reasonable price. But if we
59:54add it all up, we find that yes, it
59:57ends up being those 150 € per pickup.
1:00:01Now dear manufacturers, who are already
1:00:04at 180, I also know that the material
1:00:07cost isn't everything. Yes, I once had
1:00:11a company for 15 years where we built
1:00:14measuring devices. I am quite familiar
1:00:17with costing. The manufacturing labor
1:00:20costs, storage costs, sales costs, the
1:00:23entire procurement must be organized,
1:00:26sales must be organized, and a few
1:00:28euros of profit should also remain. So
1:00:32let's calculate 10 € here, then the
1:00:34winding machine has to run for a few
1:00:37minutes, then it has to be bagged. Yes,
1:00:41and then you quickly arrive at 150 €
1:00:43because we have a market economy. No,
1:00:46this is not price gouging here. The
1:00:51price is based on demand, after all,
1:00:53and on what our dear fellow musicians
1:00:56are willing to pay. When it comes to a
1:01:00Leonardo, no one asks if the time and
1:01:03the 100 million euros are worth the
1:01:05materials. It’s about completely
1:01:06different things there. And it's the
1:01:09same here; for this, the master
1:01:12personally wound your pickup. Is that
1:01:15crazy? Yes, just as an aside, because
1:01:19I’ve been asked about this a few
1:01:20times. This pickup here, which I once
1:01:24showed in a video, does not have Alnico
1:01:27magnets. It has the standard steel pole
1:01:31pieces inside. There is a bar magnet
1:01:34under these two coils. But if you buy
1:01:39something like this and swap the pins
1:01:41out for six Alnico magnets, then you
1:01:44have a nice, good single coil. With
1:01:48lots of wire windings on it. Whether
1:01:51you like it or not, you have to try it
1:01:53out. It’s not a huge investment,
1:01:56after all. Yes, and that’s how it
1:02:00reads. There is a lot of nonsense out
1:02:02there. The farce, the 60s sound like
1:02:05Eric Clapton, who always sounds the
1:02:08same, but whose sound you can have. Buy
1:02:11these pickups along with our Tele
1:02:13pickups. People are happy to pay more
1:02:16than € 150 each for them, but then
1:02:19you sound like Jeff Beck. Keith
1:02:21Richards, Bruce Springsteen. Lots of
1:02:24heroes on the guitar. Richards plays
1:02:27guitar about as well as Jeff Beck.
1:02:29These are all the great idols, and if
1:02:32you want to achieve that, then buy
1:02:35these replacement pickups that produce
1:02:39the original 60s sound. If you say
1:02:43Springsteen now, well, we know that’s
1:02:46not really a Tele, as the second pickup
1:02:48was added later, but still, it’s the
1:02:51typical sound. Well, there is a very
1:02:56nice video. Take a look on YouTube for
1:02:59Springsteen Leipzig, the song "You
1:03:01Never Can Tell." Chuck Berry. It’s
1:03:05very nice to see how he first finds the
1:03:08tone on stage and briefly instructs his
1:03:11musicians, and then they play it. But
1:03:16for this topic, it’s much more
1:03:18interesting. You see his guitar in
1:03:22close-ups again and again, what kind of
1:03:26pickups are in it, and no one should
1:03:30tell me those are the original pickups
1:03:34from the 60s or even 50s. But see for
1:03:38yourself. Now, in conclusion, one
1:03:42little specialty. I said at the
1:03:45beginning that I hadn't seen a
1:03:47manufacturer yet who had presented
1:03:49reasonable curves. There is someone in
1:03:53Australia, far away, who some time ago
1:03:56—it’s been a while—sent an
1:04:01epistle to his fans. It's about the
1:04:05question of what happened to a pickup
1:04:08someone sent him; it had lost treble—
1:04:11not the person, the pickup—and Kinman
1:04:15analyzed it and found, "Aha, it's not
1:04:18the magnets, they haven't aged." Well,
1:04:23what a surprise, it’s the coil; it
1:04:25had shorted turns. He correctly
1:04:28recognized that this actually happens
1:04:30with old pickups, so he rewound it and
1:04:32the treble came back, which he
1:04:34documents with this image. First off,
1:04:37what’s with this incredibly stupid
1:04:40scaling? I mean, come on, we have such
1:04:44an awkward graduation here; he could
1:04:46have used 5k per tick if he wanted a
1:04:49linear representation, instead of
1:04:52something so insanely convoluted. If
1:04:55you're somewhere in between, how much
1:04:57is that supposed to be? Second, the
1:04:59abscissa: what frequency is indicated
1:05:02here? Did everyone spot the 1 kHz mark
1:05:05immediately? If you have experience,
1:05:08you know where it is. But if you don't
1:05:10have experience, he really should label
1:05:13it. And then there's this '64 original
1:05:17Strat pickup—that's the defective one
1:05:20that aged and lost treble. Sure, I
1:05:24believe that, it's possible. Shorted
1:05:27turns cause a loss of treble. Then he
1:05:29rewound it, and now the treble is right
1:05:32where it belongs. And he writes here on
1:05:35this curve 41.25 k. That’s the
1:05:42maximum value. The frequency here would
1:05:46be 1000 Hz; 1, 2, 3—a good 3 kHz
1:05:48would be quite typical for a Fender,
1:05:51with a cable, though he doesn't say
1:05:54which cable you should use. But how
1:05:57does he get such a low resistance? No
1:06:00idea; well, a slight hunch about what
1:06:03might have happened. In any case,
1:06:06that’s not what a Stratocaster pickup
1:06:08impedance frequency response looks like
1:06:11. That’s just measured incorrectly.
1:06:14And if he publishes something like this
1:06:16without realizing it, I can only assume
1:06:18he has no clue about measurement
1:06:19technology. The second thing, he
1:06:22doesn't comment on it specifically, but
1:06:24he shows these curves. And here, again,
1:06:27I can only assume he interprets that as
1:06:30a transmission frequency response. Why
1:06:34he starts at 20 Hz and shows this
1:06:36jitter there will remain his secret.
1:06:40It’s possible his measuring device
1:06:41has a built-in high-pass filter. That
1:06:44in itself would actually be quite
1:06:45useful. You don't really need to
1:06:48measure pickups at 20 Hz, and if you do
1:06:50, you need the right setup, not like
1:06:52this. And this jitter here—but let's
1:06:55look at this part instead. Low-pass
1:06:58behavior, yes, Minimum, then it goes
1:07:03back up. What on earth did he measure
1:07:06there? I have a few guesses. First, I
1:07:12tried to replicate this strange
1:07:14measurement—this is his original
1:07:16chart. So I determined his scale
1:07:21division is 4.4525 kilohms. I have 5k
1:07:25here. That is nonsense. Then I wrote
1:07:29down the frequency here so you know
1:07:30where you are. kHz 3 kHz, a good 3 kHz.
1:07:35So that's roughly what it looks like.
1:07:39And now I’ve tried to reproduce his
1:07:43nonsense measurement and realized that
1:07:46the pickup is simply being loaded way
1:07:49too heavily. You shouldn't perform a
1:07:53measurement like that. It is totally
1:07:56misleading. Then I thought to myself, I
1:08:01have to find out what these strange
1:08:03curves with the hole in them are all
1:08:05about. What did he actually do there?
1:08:09You get this curve when you measure it
1:08:11this way. You take a cable, connect the
1:08:15pickup with a potentiometer to it—
1:08:19basically the guitar—a generator, and
1:08:22a measuring device. That means he is
1:08:27simply sending the generator signal
1:08:29through the guitar into his measuring
1:08:32device, which also has a rather low
1:08:35input resistance. If you measure
1:08:38something like that, you get these
1:08:40curves, and they look very similar to
1:08:42his measurement curves. So, I assume
1:08:45this is his setup for measuring
1:08:48transmission frequency responses. And
1:08:52this measuring device likely has an
1:08:54input resistance of about 47k. And if
1:08:58you then use that to measure pickup
1:09:01impedance, this nonsense is what you
1:09:03get. A pickup manufacturer is
1:09:07publishing this. I am not saying his
1:09:09pickups are unusable, and that applies
1:09:12to the entire video. I don’t want to
1:09:15claim that any manufacturer's or
1:09:18dealer's pickups are worthless. I
1:09:21don’t even know them all. The ones I
1:09:24have had here were all usable. What
1:09:28bothers me, for one thing, is the
1:09:30pricing. I think you can ask for 50 €
1:09:34for a single coil. As I said, there are
1:09:37a whole range of costs involved besides
1:09:39the material costs. Anything above that
1:09:42would have to be justified by some
1:09:45special quality. But that often just
1:09:48lies in the fact that some manufacturer
1:09:52thinks they are something special, or
1:09:55perhaps that they only sell small
1:09:58quantities, so the revenue doesn't come
1:10:02from volume, but from the markups.
1:10:06There is this Chamber of Commerce
1:10:08guideline, after all. If a pickup
1:10:12manufacturer sells fewer than five a
1:10:15month, they are a boutique manufacturer
1:10:17and can charge whatever they want. The
1:10:20market will sort it out, and the market
1:10:22is sorting it out. So, dear guitarists,
1:10:25if you feel you have to pay 450 € for
1:10:28three single coils, well, don’t let
1:10:32me stop you. This is what the correct
1:10:36curve for a Stratocaster pickup would
1:10:39look like. At resonance, loaded with a
1:10:43cable, the impedance goes up to over
1:10:46300 kOhms. Mr. Kinman published this as
1:10:51the ideal curve. That is so far off
1:10:55base. That is nonsense. A few quotes
1:11:00from the homepage of a guitar builder,
1:11:02whom I won't name out of professional
1:11:05courtesy. Can be provided later if
1:11:09desired. Different copper alloys of the
1:11:13winding wire, used for the pickup,
1:11:15result in strong tonal differences. So,
1:11:18is that the case? Copper exists in
1:11:22different alloys. One could perhaps
1:11:26also consider work hardening and such
1:11:29things. The only technical physical
1:11:33quantity that is important here is the
1:11:35electrical resistance. And there you
1:11:38see that pure copper has a slightly
1:11:41lower resistance than conductive copper
1:11:43. The differences are about 3.5%. As I
1:11:48said, there are different types of
1:11:50copper. For those who want to know more
1:11:52, Copper Institute www.kupfer.de. Now
1:11:56there is also a manufacturer who once
1:11:58wound a pickup with silver wire. I
1:12:02examined that back then as well. Silver
1:12:05is indeed a bit lower in ohms than
1:12:08copper. Are these 3.5%or with silver 6
1:12:15to 7%significant? Here we see two
1:12:19pickup frequency responses, the
1:12:22copper-colored frequency response and
1:12:25the black one is the silver frequency
1:12:28response. Copper wire, silver wire. The
1:12:31differences are so minimal that you
1:12:33don't even see them in this range. When
1:12:36comparing different copper alloys, the
1:12:38differences are less than 0.04 dB. It
1:12:42is with absolute certainty inaudible to
1:12:45claim that the sound of a pickup would
1:12:48change if you use a different copper
1:12:51wire. Same diameter, same insulation,
1:12:55just a different copper alloy. Nah,
1:12:58there's nothing to it. Furthermore, one
1:13:02must consider that manufacturers of
1:13:04such enameled copper wire specify
1:13:06tolerances for resistance because they
1:13:09cannot guarantee the diameter with
1:13:11arbitrary precision. And due to
1:13:15manufacturing tolerances alone, one
1:13:17must expect an 8 to 9%tolerance. What I
1:13:21don't understand about this is that it
1:13:23would be very easy to conduct an
1:13:25experiment. Any guitar builder who
1:13:27believes their pickup needs a specific
1:13:30copper wire could do it. You connect
1:13:34200 ohms in series to the pickup.
1:13:37Granted, a lot can go wrong there if
1:13:38you have no idea about electrical
1:13:40engineering. You have to know how the
1:13:43switch is connected and about the
1:13:45supply lines and all that, but you
1:13:47could ask someone who has an
1:13:48understanding of electrical engineering
1:13:50. And then you switch in a blind test
1:13:54and find out that when you switch 12
1:13:56ohms in series to the pickup, it
1:13:58doesn't change anything at all. And
1:14:01that is why this statement is nonsense.
1:14:04Then we can still change the sound
1:14:06quite drastically with the choice of
1:14:08magnet type. Drastic words. If you let
1:14:12yourself be carried away into making
1:14:14such statements, you should at least
1:14:16have a Tesla meter. They used to be
1:14:18called Gauss meters. I have one and I
1:14:21measure my magnets with it. But if you
1:14:25conduct pickup experiments and swap
1:14:27magnets that you haven't measured, and
1:14:30whose properties you don't know, that
1:14:32does not show professional competence.
1:14:35I once made a video about this with a
1:14:38special guitar. www. Guitarphysics.de.
1:14:42Instead of DC resistance, specifying
1:14:45induction, or AC resistance, would be
1:14:48better. He says: "Well, there's some
1:14:52truth to that." DC resistance is
1:14:55actually uninteresting. But induction
1:14:59is not the same as AC resistance. What
1:15:03he means is inductance. Yes, a small
1:15:06difference, like between porn and
1:15:08cancellation. Induction is something
1:15:11completely different from inductance.
1:15:14You should know that if you're writing
1:15:17about magnetics. Likewise, no
1:15:21statements across different
1:15:22manufacturers regarding output volume
1:15:25are possible. Volume of the pickup. How
1:15:29can you reliably record values in
1:15:31millivolts when just half a millimeter
1:15:33more distance from the pickup to the
1:15:35strings drastically changes the volume?
1:15:37Again, we have drastic statements. In
1:15:41reality, it's one and a half dB.
1:15:44Millivolts can be measured. And when
1:15:46comparing pickups, you have to ensure
1:15:48that this distance is always the same
1:15:50today. Then you can compare. You have
1:15:53measured this and published it in the
1:15:55book and on the homepage. There is
1:15:58something else I would like to quote
1:16:00again here. There is a book E-Guitars
1:16:03with an author involved. The production
1:16:07of a magnet is relatively simple. No,
1:16:10it is very complicated. For example,
1:16:14the temperature, which is often around
1:16:161000 ° in the melt, must be maintained
1:16:18within a few degrees. Just a 10 °
1:16:21deviation results in extremely poor
1:16:24magnets. Who says that comes up later
1:16:28in the literature review. You need, for
1:16:32example, pure iron with at most 0.02%
1:16:35carbon, and the alloying components you
1:16:39add need 0.05%accuracy. Everything has
1:16:44to be very precise. Special temperature
1:16:46profiles must be maintained. The
1:16:49production of a magnet is not
1:16:50relatively simple. The base materials
1:16:54are merely exposed to a very high
1:16:56voltage, which aligns the magnetic
1:16:58forces in one direction. That with the
1:17:03direction yes, but no high voltage. You
1:17:06need a high magnetic field strength.
1:17:10With high electrical voltage, nothing
1:17:11happens at all. You need a magnetic
1:17:13field. So, wrong again. The field
1:17:16strength of the magnet is measured in
1:17:18the unit Gauss. And yet again, it is
1:17:20wrong. Gauss is the unit of magnetic
1:17:22flux density. Field strength was
1:17:25measured in Oersted before, and now in
1:17:28Amperes per meter. It's all the same
1:17:31author. Laminated steel plates are
1:17:34supposed to avoid magnetic eddy
1:17:36currents in the pickup, which lead to
1:17:38inharmonic distortions. There is some
1:17:42truth to that. If these steel plates
1:17:45are electrically insulated, they can
1:17:48indeed reduce eddy currents. However,
1:17:52eddy currents do not lead to inharmonic
1:17:54distortions. That is nonsense. Eddy
1:17:57currents dampen the highs. That is
1:17:59something completely different. The
1:18:02resonance frequency of a pickup, the
1:18:04so-called Q-factor, is significantly
1:18:06increased by using steel. And once
1:18:10again, it misses the mark. The Q-factor
1:18:12is not the resonance frequency. These
1:18:14are two completely different things.
1:18:17The Q-factor describes the resonance
1:18:19quality and not the frequency.
1:18:21Something completely different. And if
1:18:24you now insert steel plates into the
1:18:27pickup, rail pickups, then the Q-factor
1:18:30is not significantly increased, but
1:18:34rather lowered. So, wrong again. With
1:18:39ceramic magnets, the signal doesn't
1:18:41collapse even with a heavy strike,
1:18:43because the magnetic particles, well
1:18:45yes, the particles inside the pickup
1:18:47cannot be unsettled by the string
1:18:49vibration. Well, what can you say to
1:18:52that? I've read so much nonsense
1:18:55already that I don't let it unsettle me
1:18:58anymore. A small literature review that
1:19:01I used back then for the book. For a
1:19:05start, I recommend the bolded books,
1:19:08which are usually particularly
1:19:11interesting. The gentleman was at the
1:19:14Central Research Laboratory of the
1:19:17Permanent Magnet Association in
1:19:19Sheffield, England. There you can read
1:19:23about how magnets are made and how they
1:19:26work. You can also get data there on
1:19:30the old Alnico magnets, which aren't
1:19:33used as extensively today. Partly in
1:19:38speakers, yes, occasionally in
1:19:41measuring instruments, and in guitar
1:19:43pickups—nothing against these magnets
1:19:47, but if you write about them in a book
1:19:50or on a homepage, you should have some
1:19:53basic knowledge. Yes, now to the end.
1:19:59Lastly, if someone were of the opinion
1:20:01that they'd like to try winding it
1:20:04themselves and doesn't want to wind the
1:20:06individual turns by hand or with a
1:20:08drill. There are winding machines.
1:20:12Mains voltage, motor, yes, and wire
1:20:15gauge—that is an important parameter.
1:20:19Guitar pickups typically have 0.06 mm,
1:20:23roughly a bit more or less. That could
1:20:26be the winding width and so on. What
1:20:30interested me in particular was the
1:20:32maximum rotation speed of the "
1:20:33policeman." 6,000 circles per minute.
1:20:38Everything else is okay again. Five
1:20:40more in stock. I don't know if it's
1:20:43good or bad. I found it on the internet
1:20:46and thought to myself, if someone ever
1:20:49wanted to make a policeman rotate, this
1:20:51is how you do it. For those who didn't
1:20:56get it: "Cop" has several meanings.
1:20:59Have a nice weekend.

About this transcript

This page contains the full transcript of T24 Pickup-Secrets neu by Physik der Elektrogitarre, generated from the public captions YouTube serves with the video. The transcript has 10,122 words across 1,553 segments, with the original timestamps preserved so you can click any line to jump to that moment in the embedded player.

What you can do with it

Use the transcript to take notes, quote the speaker, build a study guide, generate a summary with ChatGPT or Claude via the YouTube Summary tool, or export it as a timed subtitle file with YouTube to SRT. You can also re-open it in the transcriber to translate the transcript into 100+ languages.

Free YouTube transcript tool

YouTube2Text is a free YouTube transcript generator — no signup, no daily limit. Paste any YouTube link and get the full transcript instantly, with timestamps, click-to-jump, translation to 100+ languages, AI prompts for ChatGPT, Claude, and Gemini, and exports to TXT, SRT, VTT, or Markdown.