|
Post by stratotarts on Mar 23, 2024 15:24:53 GMT -5
There are screen shot images in that section of the manual, that show considerably less range than the +60 to -60dB range in your plot. So there must be some way.
|
|
|
Post by stratotarts on Mar 23, 2024 10:58:56 GMT -5
You could have a switching network to engage the proper number of bottom coils. It's an interesting idea, needs some further thought...
|
|
|
Post by stratotarts on Mar 23, 2024 9:47:19 GMT -5
That would only work with some certain number of top coils selected. You said one, but what if two are selected?
|
|
|
Post by stratotarts on Mar 23, 2024 9:35:20 GMT -5
Have you tried turning off the phase plot? Edit - oh, you did. NM. What happens when you Change the Amp/Freq settings under the Param Set menu as outlined on page 19-3 of the manual?
|
|
|
Post by stratotarts on Mar 23, 2024 9:32:45 GMT -5
Next step might be to enlarge the radius of the bottom coil.
|
|
|
Post by stratotarts on Mar 20, 2024 12:37:17 GMT -5
From a quick look, I can say that the amplitude plots you posted look okay. However, the resolution and division settings you chose for the scope display, are not very fine. So if you want more certainty about the measurements, you should increase the resolution and adjust the plot division/scale to show more of the plot, and show it in more detail.
To re-iterate, I know that the integrator has big phase shifts in the subsonic range. It's not practical to build a simple op amp based integrator without a low end cutoff because the DC offsets in the feedback loop are subject to an extremely high loop gain in that case, which produces unacceptable DC offsets at the op amp input. The current design therefore uses capacitors to separate the AC from the DC gain, which is made equal to one for the absolute minimum input offset. Both the series input RC and the parallel feedback RC network are matched in frequency, to the low frequency cutoff value of about 18Hz that I mentioned.
The component values to achieve the 18Hz cutoff are near the limit of what you could use and not introduce biasing issues with the op amp. It might be possible to extend that lower by using adjustable input offsets or some other modifications to the circuit, but first I would like to know the reason. Because, the phase by itself doesn't provide much information besides confirming general electronic theory (it shows a typical phase shift for this kind of circuit). More than that, would require some sort of wide band calibration with some "perfect" pickup in order to null out the phase effects of the test coil, generator and integrator as a unified signal path.
Amplitude testing isn't required below even 100Hz because the outcome is always so predictable due to the diminution of eddy currents and capacitive reactance. So, I used that as a design target and settled for a minor (~0.2dB) shelf at 100Hz. I expect that users will be at the same time aware of it, and also of its unimportance to the measurements.
If there is a specific reason for making special phase testing while also using an excitation coil, then I might be interested in thinking about, and presenting solutions. Otherwise, I am sorry but it seems like a waste of time. The integrator is not intended for phase measurements, and I never claimed that it is. Having said that, there is also the aspect (or question?) that assuming that you did want to capture the phase, why it would be important at the lower frequencies (just as, the reason for measuring amplitude at lower frequencies would need an explanation or experimental justification)?
If you are really, really interested in that kind of phase response and you want to measure it with the integrator, you could restrict the frequency range to something reasonable and perform a wide band calibration with a nearly perfect, almost theoretically correct DUT such as an air core inductor. It's possible, it might be interesting, but as you can guess, I'm skeptical of the value unless the underlying theory or idea is explained to me. Since you just pose it as a question, "would it reveal anything?", well the way I would treat that myself is to look at the existing plots in the light of that question.
However, in general, phase has no "stand alone" relevance in the context of the pickup circuit, outside of the relationship with impedance. That can be much more easily and reliably measured using the two terminal tests that I mentioned.
|
|
|
Post by stratotarts on Mar 17, 2024 7:20:54 GMT -5
The integrator can not reliably measure phase. It can only reliably measure amplitude which is what it is designed for. But it's not mainly because of any part of the implementation of the integrator circuit. It's because the test coil to pickup coupling is not controlled for phase. You found that out when you moved the arrangement and got different readings. There is some amplitude and phase shift at the high and low ends of the integrator response but this has been overcompensated as far as possible. The low end response was set at 18Hz, well below the required test range 100Hz. At that frequency the circuit phase shift is more noticeable than the amplitude variation which is a fraction of a dB. At the high end, the phase accuracy is limited by the bandwidth of the op amp. The JFET amp in use is quite fast, so errors should be very minimal below 20kHz.
If you want to seriously consider phase, you must perform two terminal network analysis, directly at the leads of the pickup. Because that kind of measurement gives you an isolated coupling to the pickup, the actual phase/amplitude values can be processed further mathematically and yield valid results. I once modified an integrator board to do that, but as usual got sidetracked with other work. Member here "ms" and others use this kind of approach and can generate Bode plots from it. Those kinds of plot follow closely what you would measure with a test coil and integrator, except that the cover losses will differ, since the geometry of the pickup coil, and the pickup coil test coil combination are different.
In a reactive circuit, phase and amplitude are intimately related, but it does not mean that you necessarily need to measure both, it definitely depends on the actual behaviour that you are trying to capture. The guitar output chain and human ear (in the limited context of a mono instrument played through an amp) are mostly insensitive to phase, the phase affects the amplitude along the way but it is the amplitude that is relevant at the output. Again, if you do use phase measurements, they have to take into account the phase behaviour of all your test equipment as well as the device under test.
It isn't the case that phase in the guitar circuit is not relevant. It could factor into calculations about the circuit. But you can not learn much from putting a "phase stethoscope" on the resulting signal and hope to learn what is going on inside the "body".
|
|
|
Post by stratotarts on Mar 11, 2024 15:30:23 GMT -5
Zexcoil Z-Series use metal pole pieces but then do use individual magnets at the bottom of each pole. Would this give a a result more similar to traditional alnico pole pieces instead of rails? No. The Zexcoil designer has explained that the poles material is the main determinant of the sound of those pickups. In fact, he offers different pole materials as options. But they are fundamentally like steel poles on other pickups, because they are all different formulations of steel (last time I looked at the Zexcoil site IIRC). The Neo's don't play much of a role, other than to effectively magnetize the steel.
|
|
|
Post by stratotarts on Mar 11, 2024 15:20:31 GMT -5
I'm not aware of any dual-rail pickups that use alnico magnets as the rails themselves. I assume this is for some reason not viable? Well, the Firebird pickups do. They are not rail pickups per se, but internally they have alnico bars inside the bobbins. There are multiple reasons why not rails. First, manufacturers hate to use multiple magnets because they are a big part of the pickup cost. But also, the magnets are brittle so would break easily if left exposed. You may say, what about the round magnets? I think those are harder to break because they are thicker and don't have corners, only edges. Aesthetics probably comes into it as well. Exposed bars wouldn't really look very good IMHO.
Forgot to mention, manufacturability. That's really what limits the Firebird design. It is difficult to position all that stuff without a frame, and hold it in place. The Firebird PU uses the cover as a kind of fixture to hold everything together, assisted by the subsequent potting.
|
|
|
Post by stratotarts on Mar 11, 2024 15:03:37 GMT -5
The proximity to the poles gained by the thinner wire, also improves the efficiency. Refer to Zollner's EM simulations of steel pole pickup string fields.
|
|
|
Post by stratotarts on Mar 11, 2024 15:00:47 GMT -5
I don't think this would give you an honest representation of the effect. You would need to compare the instantaneous phase/amplitude of each pole set. It is only possible to capture that in real time from both pole sets. With this technique, the phase between the two poles doesn't have any significance because it's arbitrary since it's taken at arbitrary times and created with different pick attacks also.
|
|
|
Post by stratotarts on Mar 9, 2024 14:02:16 GMT -5
It's true. In the 20th century, most interference was from mains frequency. That is low frequency, except where there is arcing or chopping like in a light dimmer. In the latter years, more and more high frequency circuits made their way into consumer and industrial circuits. So there is a much greater range of frequencies that have to be protected.
As the frequency increases, the magnetic and electric effects increasingly interact, which makes it all the more difficult to fix and to explain. Small wonder that there is some confusion.
|
|
|
Post by stratotarts on Mar 9, 2024 13:48:43 GMT -5
Side note: Impedance is moot if you have an active preamplifier and a mixer for both the pickup and balance coil. Because then, there's no need to put them in series or parallel combination that would introduce the need for an undesired pickup impedance or some unwieldy passive compensation. Also the set input gain to the mixer, would dramatically loosen constraints on the balance coil.
I do fully understand the desire to avoid active electronics in a guitar.
|
|
|
Post by stratotarts on Mar 9, 2024 13:38:15 GMT -5
The pure cancellation at short wavelengths seems unlikely. Yes, the instrument would be frustrating to play if that were so. Because of the fretting in different places, notes would have effectively arbitrary volume. I think it's reasonable to believe that the effect exists, but is diffuse due to the phase spreading that you mentioned. In other words, affects all notes almost equally (with a gradation from very low to very high). Tillman's diagrams show the behaviour of an extremely simplified mathematical model.
|
|
|
Post by stratotarts on Mar 8, 2024 20:31:58 GMT -5
No, it's that when the signals from the two coils is added, signals of opposite polarity cancel out. In reality those are harmonic nodes on the string. Suppose the fundamental is A, the signal is A+A, the same in each coil because the string is moving in the same direction in relation to each coil. So, 2A. But harmonic B is a much higher vibration of the string, such that the crest of a wave is opposite one coil, and the valley is opposite the other - so one is B and the other is (-B). When added, B+(-B) = 0. They cancel each other. This action depends on short wavelengths, so the effect is that generally, higher harmonics cancel out more than the fundamental and the lower harmonics.
Consider the article by Tillman, linked in the reference thread:
|
|
|
Post by stratotarts on Mar 8, 2024 8:49:52 GMT -5
Like this?
|
|
|
Post by stratotarts on Mar 8, 2024 8:31:57 GMT -5
Yeah, but one reason why the Micro Coils have less noise is in the small internal radius of the coils. The larger the radius, the more inductance all other things being equal, and also a reduced degree of coupling to external magnetic fields. I'm not sure if Bill came to that with a full understanding or by experiment but it certainly works that way. I know of no other designer that recognized that opportunity and leveraged it. At the end of the day, he was doing the right things to minimize the noise - both electric and magnetic - at least in the Micro Coils and I think with the others too.
EMI is a general term that could apply to either electrostatic or magnetic interference. It can also refer to interference from electromagnetic waves, but as ms mentioned, it's not applicable in this case.
The topic is not restricted to pickups. Many, many electronics devices have needed some kind of protection from external fields and so the field is pretty full of solutions. For example, toroidal inductors, which offer exceptionally good isolation from external fields and also contain their internal fields from outside circuits very well. Some early humbucking pickup patents reference existing patents for humbucking transformers, inductors and so on that were for general electronic use. It was not a new idea.
|
|
|
Post by stratotarts on Mar 7, 2024 17:50:05 GMT -5
I've never seen a Strat sized quad. The ones I saw were humbucker sized.
|
|
|
Post by stratotarts on Mar 7, 2024 17:42:30 GMT -5
It's possible your generator is radiating RF. Is this in reference to the increase in 60 Hz? <abbr>No, the 510 Hz. You said it was unexpectedly strong. If you don't think so, I'll go along with that.
|
|
|
Post by stratotarts on Mar 7, 2024 16:14:10 GMT -5
It's possible your generator is radiating RF.
|
|
|
Post by stratotarts on Mar 7, 2024 13:28:06 GMT -5
By electrical sources, do you mean only that which travels through the wiring to a given device? No, I'm talking about the dynamic electrostatic fields that exist in the (mostly technological) environment. You are sensing them every time you touch the disconnected end of a guitar cable that's plugged into an amplifier. The BZZZZZZZZT. You can tell that it's electrostatic (high impedance) in nature, because if you ground some other part of your body to the amp ground, you have created a medium (~2k to ~500k ohm) impedance path that short circuits the signal.
Another bit of evidence from this, if you connect the plug tip to a much larger object, the capacitance between the electrical source and the input is greatly increased, which greatly increases the buzz. The operation of a capacitor is fundamentally electrostatic, it's an electric field displacing electrons that are held captive in the plate conductors.
|
|
|
Post by stratotarts on Mar 7, 2024 12:36:06 GMT -5
So if I slide a sheet of mu-metal between the strings and the pickup, the sound should go dead quiet? Maybe. I'll be honest and tell you I don't know what is special about mu metal. But, with respect to aluminum (or plated copper as we see in cheap covers) the situation with eddy currents is comparable when a sheet is placed nearby, except that if it is behind the pickup it is far away from the string field and so has minimal effect. But yes, a small part of the same effect.
However the string field and an external interference field are not of the same character. A conductor in the near vicinity of a string field has induced voltage in opposite directions, which only create eddy currents (and therefore somewhat block the field variations) when there is a low resistance path through the conductor which connects the induced regions. This is just the well known effect of lossy covers.
But, a uniform alternating external field does not excite regions of the conductor in different directions, because the field lines are uniform. A voltage is induced by Faraday's Law. However in this case the voltage differential is also uniform across the conductor and therefore there are no regions with opposing voltages, and no current will flow. Thus, there is no diminuation or blocking of the external field.
This is why I think the attributes of aluminum as a material that will block interference, must be mistakenly attributed to magnetic interference when it is really electrical. It's the lack of experimental rigour that made that mistake possible. Also I think, a sketchy understanding of electromagnetism, in fairness it's pretty weird stuff.
Also, it's unlikely that a plate could "disturb" the field. It's recognized that such materials are transparent to externally applied magnetic fields. I explained why there is no (or extremely minimal) electromagnetic interaction, so you can't look there (unless you want to debate that specific point). Simply put, there isn't any specific theory that would predict it. I'm open to suggestions, but I haven't encountered any that are fully fleshed out.
Installations of mu metal that I have seen, fully enclose the shielded object, perhaps that accounts for a difference... we do know that steel will concentrate magnetic field lines, but it is far from magnetically transparent.
|
|
|
Post by stratotarts on Mar 7, 2024 12:12:58 GMT -5
Aha, so it's a reference point for the forthcoming posts. Got it.
|
|
|
Post by stratotarts on Mar 7, 2024 11:37:45 GMT -5
What is the guitar configuration? Is this an RWRP pickup pair in the middle position? With no cancellation coils installed?
|
|
|
Post by stratotarts on Mar 6, 2024 13:55:47 GMT -5
FYI, quad rails are available on the import market. Based on the dual rails I have looked at, they should be pretty normal, i.e. no big skimping on wind count, very comparable to similar name brand offerings. I had one (dual coil blade HB) in the middle position of an Ibanez RG, it sounded great all by itself. If you're just experimenting you could save some big money there.
I came across those just after the end of the times when I would buy pickups, just to test them. So I never owned any. The Tele has the SuperSwitch for playing around, for other guitars you need something like a rotary wafer switch.
|
|
|
Post by stratotarts on Mar 6, 2024 13:42:26 GMT -5
Bill L. was very smart and observant. But he lacked the engineering background to understand the experimental rigour that is needed to really verify or invalidate theories. He never published enough details in his experimental techniques to give them any scientific validity. Largely, what he did do with those investigations is to confirm the more obvious (to an engineer) aspects, and to somewhat steer people away from really wacko theory and pickup folklore.
The main experimental risk, is that magnetic and electric sources must be isolated so you don't attribute results from one phenomena to the other, which may not be the root cause.
Errada's lab had a very carefully constructed magnetic field generator, it was a large tube with field windings around it. By turning the field currents on and off, he was able to isolate the magnetic from the electric field effects. That is what it takes to be sure.
|
|
|
Post by stratotarts on Mar 6, 2024 9:40:39 GMT -5
No, a higher wind count does not in itself, mean that it will be more prone to interference. That has more to do with size of the coil's internal area. However in a noise cancelling design, that may not matter very much if the noise cancelling method is well engineered.
You need to look at parameters in the context of the entire guitar circuit including controls. Yes, the capacitance makes a difference. But the self capacitance of a typical pickup is much less than that of the circuit it's in. Thinking again - for your example, the change in capacitance would be the biggest difference between them, because parallel wiring has about 4 times the capacitance as series.
The part about sensing a larger string area is controversial, or at least subtle. It does not automatically follow that it would affect the sound audibly. It has been proven that the sensing area does make a difference, but I think not on the scale of a few millimeters such as would be the case with comparing a low wind and high wind on the same pickup frame.
|
|
|
Post by stratotarts on Mar 6, 2024 9:32:08 GMT -5
The inductance is the primary factor influencing the response. However, the factors that determine inductance do not only change the inductance. Because for example, if there are more wire turns, the physical geometry is changed, parts of the coil intersect different parts of the string field and so on. Capacitance will increase with more wire turns, and so on. But those side effects are very subtle compared with the main factor, inductance. DC resistance will increase with the wire turns.
The problem with your example, the ratio of the wind counts needed to differentiate the two models are so extreme, that you are kind of black box testing the minor factors but in an aggregate way. To understand those factors, you need to design experiments that will separate them out and test them individually. Because, for a "pure" implementation, i.e. "perfect" pickups, there would be NO difference between the two models.
By the way, a stack will always have more wire turns simply because there are two coils, and limited space to wind on. Since the bottom coil doesn't contribute to the output, the overall output level will be less, and so designers must add more turns (usually with finer wire) to compensate so the pickups aren't perceived as weak sounding.
|
|
|
Post by stratotarts on Mar 5, 2024 8:22:30 GMT -5
I can attest to that because in 1965 or very close to that, I was a kid interested in electronics and overheard that a friends's mother was a coil winder in an transformer factory. I considered that a bit awesome and asked her about it. She got angry, her face turned red and she told me not to tell anyone else.
|
|
|
Post by stratotarts on Mar 4, 2024 18:36:03 GMT -5
You have to find the premise very important (a pickup that has multiple voices). It's at least interesting and may be useful to someone. If that feature is valuable to you, they might be the right product for you, notwithstanding the hype. There's no mystery about coil switching, it's been going on for many decades. Many people find it unnecessary, if you want different voicing, it's possible to just collect more guitars. A lot of people do that anyway, and as a bonus you can match the pickups to a unique body/neck in each case. Even from big name manufacturers, you could get great conventional pickups for less than that kind of money. Often the real reason for something new is that the manufacturer has run out of differences with other manufacturers products, need something to differentiate their products. Pickup design is so worked to death that it's hard to do anything really new that is really worthwhile.
|
|