| ampage Tube Amps / Music Electronics |
For current discussions, please visit Music Electronics Forum. |
ampage archive
Vintage threads from the first ten years
Re: Q Ratio... What does it mean to you!!!
:
 9/2/2002 6:57 AM |
|
Andy |
Re: Q Ratio... What does it mean to you!!! I'm with Wolfe. Attempting to justify a good (or bad) pickups sound with Q is a bit like "The Meaning of Life being 42". Valueless. You can't create a pickup with Q Keep that Art and Artisans! Andy |
|---|---|
| 9/2/2002 9:59 AM |
|
| Tony@Mastertone |
I must agree... I dont really use Q on a daily basis, where I have used it is comparing a custom pickup to a known good off the shelf type just for my own interest.... again it all comes back to the ears at the other end of the pickup, whoever that may be. Just for info I once got told by a research engineer that the perfect Q for a coil is 1...from an efficiency point of view, please don't come back at me with anything you lot as he didn't explain what type of coil that was and used in what capacity... but some of my coils I use are in the 0.98 bracket... which got me thinking. Tone |
|---|---|
| 9/2/2002 3:12 PM |
|
| Dr Strangelove |
Q means... Back in the early daze of Ham Radio, Q was short for "Quality factor." Used with coils, it told you how much of a coil's overall impedance was inductance vs resistance. It gave an answer to the question, "how good is this coil as an inductor?" Guitar pickups are transducers as well as inductors and need a bunch of other tests if you want to characterize them usefully. Just _once_ I'd like to see frequency plots of different model pickups within the same style. -drh -- |
|---|---|
| 9/7/2002 11:21 PM |
|
| R.G. | Re: Q Ratio... What does it mean to you!!! Pickups are not by any means my forte, but I do understand a bit of the techie stuff under them. And wherever there's an electron, an EE can kibitz, right? 
Q was, as Strangelove intimated, named that for the "quality factor" back in the Golden Age. However, there's more to it than that. Q is technically defined as the ratio of the energy stored per cycle divided by the energy dissipated per cycle in a resonant circuit. Strictly speaking, Q only applies at resonance, but there's a lot of fudging on the definition to be able to apply the Q concept to other points. Getting out my 1976 ARRL handbook, at resonance, the Q factor is indeed the reactance of the inductor divided by the circuit resistance. It's also the capacitance's reactance divided by the resistance, as the capacitive and inductive reactance are the same at resonance. Off resonance, some of the extensions to the definition let you say that the Q of the coil or capacitor is just 2*pi*F*L/R or 1/2*pi*F*C*R. High Q coils and capacitors alike have very low Equivalent Series Resistance. Specifically, in coils, "high Q" almost always means "low resistance for the inductance you get", and usually comes with a test frequency in the specs. Q is actually a measure of lack of damping. High Q resonances have very sharp resonant peaks. Low Q circuits have only gradual frequency response changes. You can, in fact, relate the Q of a resonant circuit directly to the bandwidth of a resonant peak. A Q of 1 turns out to be about an octave wide. In fact "high Q" is just about the polar opposite of "well damped". So what does this mean to pickups? That's really, really hard to ferret out, and I believe that Mother Nature's obscurity here is the reason you're all fairly divided between the measurers and the "I don't know how to measure but I know what I like" guys. (and no, an LCR meter is not a good way to measure L, C or Q for a pickup; there's too much going on at different frequencies) A pickup with the huge number of fine turns of wire in what is essentially an air cored coil that is an odd shape for a coil (electronically speaking). What that does is make it into a coil with many internal distributed capactances. Things like this are modelled poorly by simple lumped element models. (In real human speech that means that you can't really say it's just an inductor and a capacitor and have that be accurate.) The internal capacitances have different linkages to different parts of the coil, especially if you scatter wind instead of layer wind. Scatter winding does distribute the intra-coil capacitances to different layers, and causes several interacting resonances, not one sharp one like a simple model suggests. The exact pattern of winding controls what happens, and scatter winding is, as the name suggests, not very predictable. I bet if someone *did* do some frequency response testing, they'd find that pickups have multiple resonances, all related to winding style. A better model for doing pickups with low capacitance and predictable capacitances might be to wind a pickup like radio coils are often wound (and for the same reason; they solved this one three quarters of a century ago) - sectionalizing. The overall coil would be sliced into sections across the axis of the coil, and each section wound independently; in radio coils, each section is wound in a distributed way so that each turn covers the whole width of the section, and successive turns are "phased" slightly away from one another so the coil section can be self supporting. This kind of thing makes for very predictable and low self capacitance. I wonder if it's possible to section a pickup coil a couple or three times and see if this turns out a cleaner top end. It would be interesting. R.G. |
|---|---|
| 9/9/2002 4:32 AM |
||||
| Dr. Strangelove |
RG wrote:
The guitar pickup has a ferromagnetic core. The core's determining quality is its ability to conduct magnetism and is called permeance (absolute value) or relative permeability (compared to air). The core permeance is a major factor in defining coil inductance. Permeability calculations I've done give values around 8-10 depending upon the magnet size and shape. This means we get pickup inductances ranging from 1.5 - 9.0 Henries instead of fractional Henries. What continues to bother me is that by treating a guitar pickup is a passive RLC network and measuring it as such, we only specify its internal losses, sort of like characterizing a battery by its internal resistance instead of current output. The measurement is significant but not salient. A guitar pickup, or variable reluctance tranducer, transforms magnetic energy into electrical energy. It's a generator, not an energy soak. We need a different I/O model to characterize pickups usefully.
If we put a pickup under an AC magnetic field and perturbed it by frequency sweeps, we'd see some resonance points, awritey. The real fun would come from doing pulse tests and MLS waveforms, then comparing the results between pickup types. Differently wound layers would have different group delays on the rising/falling edge of the pulse. It would probably correlate very well with the differences we hear between single coil, stacked humbucker, and side-by-side humbucker pickups, differences that aren't addressed by discussions of magnetic aperture.
Sectionalized pickups in the form of stacked humbuckers are on the market. If you section-wind a coil, you can't cross windings at as extreme a pitch; you control distributed capacitance at the expense of higher self-inductance. I'm unable to estimate the degree of this effect, though. Another way of minimizing distributed capacitance in transformers is to add an occasional layer of insulating tape. -drh -- | |||
|---|---|---|---|---|
| 9/9/2002 6:48 PM |
|||||
| R.G. | Strangelove wrote: [QUOTE]The guitar pickup has a ferromagnetic core. The core's determining quality is its ability to conduct magnetism and is called permeance (absolute value) or relative permeability (compared to air). The core permeance is a major factor in defining coil inductance. Permeability calculations I've done give values around 8-10 depending upon the magnet size and shape. This means we get pickup inductances ranging from 1.5 - 9.0 Henries instead of fractional Henries.[/QUOTE] Yep, I know. I picked that kind of thing up back when I designed transformers for a living. The permeable material in the magnets does replace some of the flux path with higher permeability stuff. The fact remains that the fringing field outside the center of the core is vastly more important in the flux distribution than the stuff in the center. As you say, the inductance only comes up to units-of-henries, a gain of from three to ten times, instead of the several thousand times that would happen if you used a high permeability material in the full flux path (ignoring that a full-iron path wouldn't make a functioning pickup).
Well, it does turn out that those are pretty much the same. The way you figure out the value of the internal resistance of a battery is to load it at two different levels and compute the difference of the output voltage. The delta-v over the delta-I is the internal resistance. A better example is the secondary of a transformer. To compute the effective internal resistance you compute the real secondary resistance, the reflected primary resistance, the effective leakage inductance reflected to the secondary, and from that you can pretty much characterize the response, at low frequencies at least. The point is that the internal losses are the same whether you're driving the coil or it's sourcing signal.
I had in mind suspending a part of an iron string over the pickup and mechanically driving that with sinusoidal motion to more closely simulate the way that a string drags the fringing field around instead of feeding the coil fields directly. Something like a dual coil speaker with the cone removed and a couple of kevlar or carbon fiber arms holding a quarter of an inch of B string. You should be able to drive the speaker with a sine wave, sense the position with the second voice coil, and then relate that to what the pickup was putting out. True, the testing would all have to be sine wave because of the mechanical moving mass problems, but it would still turn you out some very accurate data.
I wasn't clear, I guess. Stacked humbuckers are aimed at hum cancellation. The sections I'm aiming at are intended to run the distributed capacitance down, self inductance up, as you say. The point is to reduce and make stable and predictable the self capacitance, something that stacked humbuckers are not aimed at, to the best of my knowledge.
Yeah, that works. The lower capactance reflects the distance between layer-sheets of conductive winding layers; this reduces the coil window available for winding and as a result you either get lower output or have to live with smaller wire diameter to get the turns back up, but also higher resistive losses. R.G. | ||||
|---|---|---|---|---|---|
| 9/9/2002 9:41 PM |
|||
| Dr. Strangelove |
Strangelove wrote: >> The guitar pickup has a ferromagnetic core. > > Yep, I know. I picked that kind of thing up back > when I designed transformers for a living. Perhaps I mistook your intent when you posted that the pickup "is essentially an air cored coil,"
If you could elaborate a bit more on fringing, you would find an attentive audience. In the simplest case of a bar or rod magnet, how does fringing differ from the textbook donut-shaped flux path? Is it the weird little flux loops on sharp edges that don't follow the simplest case? >> by treating a guitar pickup is a passive RLC >> network and measuring it as such, we only >> specify its internal losses > > Well, it does turn out that those are pretty > much the same. It does not tell us the device's efficiency or accuracy in converting one form of energy to another. Neither of us would consider building novel test rigs if the RLC model were sufficient by itself. [re: pickup testing] [QUOTE]I had in mind suspending a part of an iron string over the pickup and mechanically driving that with sinusoidal motion to more closely simulate the way that a string drags the fringing field around instead of feeding the coil fields directly. Something like a dual coil speaker with the cone removed and a couple of kevlar or carbon fiber arms holding a quarter of an inch of B string. You should be able to drive the speaker with a sine wave, sense the position with the second voice coil, and then relate that to what the pickup was putting out. True, the testing would all have to be sine wave because of the mechanical moving mass problems, but it would still turn you out some very accurate data.[/QUOTE] Following the Principal Of Maximum Laziness, I have in mind removing the the voice coil entirely from the speaker and setting it on top of the pickup, the rationale being to generate complex test signals and see what the pickup does. While this test method is hardly a real world condition, it allows high reproducibility and promises data which can't be had by poking the pickup in a guitar and whacking at it. We have a plethora of guitar whackers but the magnetic induction testers seem absent or very shy about speaking up. IMO, there's nothing so powerful as a bad idea whose time has come. <
You were plenty clear, and sectionalized pickups do exist, though admittedly not for the purpose of controlling distributed capacitance. Heh. Commercial CNC coil winders mention a programmed "orthocyclic" winding pattern. Since you've designed and built transformers, could you explain how it differs from perfect lay winding? Refer to a good link? I've read the Patent Office definition but it's kind of dense...or maybe I'm the dense one. -drh -- | ||
|---|---|---|---|
| <<First Page | <Prev | Page 3 of 5 | Next> | Last Page>> |