some hints on ARTA
In Setup Menu you have to select the appropriate audio input and output devices depending on your hardware available.
Now push "FR1",
push red start button.
Audio output of your soundcard should now deliver noise and you see the spectrum plot on your screen.
This noise can be fed via 1meg to hot terminal of pu or coupled by a transmit coil as described.
In Setup Menu you have to select the appropriate audio input and output devices depending on your hardware available.
Now push "FR1",
push red start button.
Audio output of your soundcard should now deliver noise and you see the spectrum plot on your screen.
This noise can be fed via 1meg to hot terminal of pu or coupled by a transmit coil as described.
Hi,
This thread is all over the place. Fact is cheap single coil pickups mainly
sound too thin, too weedy and too bright, they don't have proper "tone".
Your chasing your tail thinking you can measure your way to success.
Measurements won't reveal well staggered pole heights, the type and
quality of the magnets used in the pickup, which affect sound, the type
of potting of the coils if used. There is loads more to a good pickup.
Its been years since I looked at budget Strat pickup sets, but there
are some great value pickups out there, its just hard to keep up.
However that is what I'd search on, Strat pickup sets and the
differences between the 3 pickups and sets, is a good clue to
a decent manufacturer that can provide a pickup to suit.
rgds, sreten.
This thread is all over the place. Fact is cheap single coil pickups mainly
sound too thin, too weedy and too bright, they don't have proper "tone".
Your chasing your tail thinking you can measure your way to success.
Measurements won't reveal well staggered pole heights, the type and
quality of the magnets used in the pickup, which affect sound, the type
of potting of the coils if used. There is loads more to a good pickup.
Its been years since I looked at budget Strat pickup sets, but there
are some great value pickups out there, its just hard to keep up.
However that is what I'd search on, Strat pickup sets and the
differences between the 3 pickups and sets, is a good clue to
a decent manufacturer that can provide a pickup to suit.
rgds, sreten.
I think you guys might be interested in this thread over at Stratalk:
Measured electrical differences between a Fat 50 and a Texas Special | Fender Stratocaster Guitar Forum
Its a bit like this one, exploring testing and visualisation of pickup response but has been running for a few months now, up to page 40. In about the last 10 or 15 pages the guys have developed very good test procedures using signal sources and driver coils, combined with circuitry. These capture very closely the kind of frequency response plots that one might generate from LRC analysis models, but modified by particular extra damping characteristics etc. More pickups are being tested all the time in this and other threads. They test unloaded and also under a typical guitar/cable load.
My interest has been to take some of the test data and derive more complex models that better capture the electrical response, and it is getting quite successful.
Worth a look.
Measured electrical differences between a Fat 50 and a Texas Special | Fender Stratocaster Guitar Forum
Its a bit like this one, exploring testing and visualisation of pickup response but has been running for a few months now, up to page 40. In about the last 10 or 15 pages the guys have developed very good test procedures using signal sources and driver coils, combined with circuitry. These capture very closely the kind of frequency response plots that one might generate from LRC analysis models, but modified by particular extra damping characteristics etc. More pickups are being tested all the time in this and other threads. They test unloaded and also under a typical guitar/cable load.
My interest has been to take some of the test data and derive more complex models that better capture the electrical response, and it is getting quite successful.
Worth a look.
Well here's what I found so far. First, I had high hopes for Gnobuddy's method, because I liked the idea of better simulating what the pickup would actually do. I made my test coil and setup almost exactly as he described, and made a jig to hold a sample pickup and test coil in near proximity. I should point out that for my tests right now, I'm using two single coil pickups... one from a box of identical cheap e-bay pickups, and one other that was from an old D'ADdario product, designed to be mounted in an acoustic. These two are already known to me have "muddy" and "bright" response, respectively.
So anyway, feeding the test coil with the pink noise signal from the ARTA software did allow me to see something like a frequency response trace of each pickup. But unfortunately, even employing various averaging to smooth out the jitter, I was hard pressed to actually see a difference between the way my two test pickups responded, even printing out the two curves and holding them together up to a strong light.
Next I tried the method described by Voltwide. Here I could see peaks similar to those in his diagrams. Now the differences were still a lot more subtle then those in his comparison... my cheap pickup seemed to have a "peak" around 6Khz, while the better pickup's peak was clearly higher, at 7.5kHz. How significant is that? I don't know, as I obviously don't have enough test cases. I would have felt more satisfied with this method had I seen a difference like 6kHz and 10Khz, but at least it was a visible difference.
I'm glad at least one of these methods revealed a difference, but can't say I'm really confident that the difference account for what I'd call a very significant audible difference between these pickups. As some of you have pointed out, there are DC characteristics here such as the strength of the magnet and coil resistance that are probably not being measured at all by the impedance check, but surprisingly made little graphic difference using the induced signal test coil method. I will have to revisit that method and see if there's something i could do differently.
So anyway, feeding the test coil with the pink noise signal from the ARTA software did allow me to see something like a frequency response trace of each pickup. But unfortunately, even employing various averaging to smooth out the jitter, I was hard pressed to actually see a difference between the way my two test pickups responded, even printing out the two curves and holding them together up to a strong light.
Next I tried the method described by Voltwide. Here I could see peaks similar to those in his diagrams. Now the differences were still a lot more subtle then those in his comparison... my cheap pickup seemed to have a "peak" around 6Khz, while the better pickup's peak was clearly higher, at 7.5kHz. How significant is that? I don't know, as I obviously don't have enough test cases. I would have felt more satisfied with this method had I seen a difference like 6kHz and 10Khz, but at least it was a visible difference.
I'm glad at least one of these methods revealed a difference, but can't say I'm really confident that the difference account for what I'd call a very significant audible difference between these pickups. As some of you have pointed out, there are DC characteristics here such as the strength of the magnet and coil resistance that are probably not being measured at all by the impedance check, but surprisingly made little graphic difference using the induced signal test coil method. I will have to revisit that method and see if there's something i could do differently.
Very interesting!
Were both pickups feeding 1M, purely resistive (no cable capacitance) loads, as in your guitar?
When I did my initial measurements, I didn't use a guitar cable (my guitars are all conventional passive setups, not an active one like yours. Pots and tone controls were connected, but no guitar cable was plugged in.)
I found the self-resonant frequency of the pickups was too high in frequency to play any role in the guitar's sound. The highest frequency note you can play even on a 24-fret guitar is around 1.3 KHz - everything above that is harmonics or unmusical noises like picking and fretting noises. The self-resonance frequencies I was seeing were all up above 4 KHz, in some cases as high as 6 KHz.
Looking at this, I also couldn't make any sense of what I was seeing, in terms of being able to correlate it with what I thought I was hearing. Adding the usual 10-foot (roughly 3 metres) cable I usually use changed the frequency response substantially, and now there was a peak at a frequency that would actually make it through a guitar amp and guitar speaker.
I no longer have my old measurement data, unfortunately. But, for the purpose of discussion, and maybe to stimulate some ideas, here is an LTSpice simulation I did recently.
For the simulation, inductance and resistance values for the pickup were taken from a table of measurements I found somewhere on the Internet. The self-capacitance and cable capacitance values are just my guesses.
Whether or not those values are accurate, the simulation shows some general features of interest: the pickup acts as a low-pass filter, with some peaking on resonance, and a roughly 12 dB/octave (2nd order) rolloff above it.
Both measurement methods we've been discussing on this thread multiply the curve in the LTSpice simulation by frequency (voltage output is proportional to dphi/dt, where phi is the magnetic flux, and for a sinusoidal change, d/dt is proportional to frequency). This is equivalent to tilting the whole LTSpice plot counter-clockwise, multiplying it by a rising +6dB/octave response.
The test-coil measurement method could be modified to generate (hypothetically) the same curve as the LTSpice simulation, if the constant current coil drive were replaced by a current falling at 6 dB/octave. This could be done by using constant current drive from the output of an integrator, for instance.
The same thing could be done mathematically too, by simply dividing each data point by its corresponding frequency.
It would be interesting to see if the measured data in any way matches up to the simulation, which is based on electrically modelling a guitar pickup as an inductance, series resistance, and parallel winding capacitance.
-Gnobuddy
Attachments
I do not think either of those frequencies has much audible significance when actually used with a guitar and guitar amplifier.
Most electric guitar amps (and speakers) heavily roll off output above 5 - 6 KHz, and often begin that roll off starting with a big peak as low as 3-4 KHz.
We do know that metal-string acoustic guitars generate a wider bandwidth than the typical electric guitar does. Your acoustic pickup shows that this was considered by the designer - it has a higher self-resonant frequency, which probably comes from fewer turns in the winding, and reduced self-inductance.
One other thing to consider: if you feed pink noise into a graphic EQ, listen to the result, and then raise first the 6 KHz slider, then the 7.5 kHz slider, can you hear a difference?
I've never tried exactly this experiment, and the graphic EQ I use most often doesn't have any bands higher than 6.4 kHz (it's a guitar effects pedal). But on this seven-band graphic EQ, raising any one slider sounds quite noticeably different than raising one of the sliders adjacent to it.
So, if you were to try the 6kHz vs 7.5 kHz EQ peak test, my bet is that you would indeed hear a difference, and a very substantial one. (This may not be relevant to guitar per se, but speaks to the abilities of the human ear.)
-Gnobuddy
Yes... exactly as you described. There was some AC hum pickup due to test leads, but that was obviously unimportant for this test.
Its a separate subject, but I'm surprised you'd count harmonics or even picking noise as "non musical". Surely you understand that without harmonics, a guitar string would sound no more interesting then a damped sine wave. In fact, one of the things that makes a vibrating string sound so interesting, and one of the reasons its hard to simulate electronically, is that the harmonics of a vibrating string tend to be sharp in comparison to what they should be from a pure mathematical standpoint. But anyway, I can assure you the ability to respond to harmonics up around 4khz and 6khz make a big difference in the sound of a guitar.
Measured electrical differences between a Fat 50 and a Texas Special | Fender Stratocaster Guitar Forum
This guy says a lot, but that's about it. He is wrong on a lot of stuff and the crew over there are following him like he is the answer to all.
"The geometry of the coil, and the insulation thickness have no effect on the inductance" What?
This guy says a lot, but that's about it. He is wrong on a lot of stuff and the crew over there are following him like he is the answer to all.
"The geometry of the coil, and the insulation thickness have no effect on the inductance" What?
Thats a 40 page thread, and like any long thread there is a lot of things said by many people that deoend on context. I linked it here because there is very good testing happening, particularly in the second half. Diss it if you like, or learn from whats good. Nobody is really saying that coil geometry is not important.
By the way Gnobuddy, I found this interesting article that shows a pickup tester, that seems to work a lot like what you suggested. And I also realize one thing i did wrong, was I used a pickup bobbin with a steel insert core. I'll have to try again without it. I think it also might be useful for me to set up an OP-AMP as as high impedance buffer into my sound card, so i won't pick up so much noise from all the 1 Meg HIGH-Z input.
BuildYourGuitar.com :: The Secrets of Electric Guitar Pickups
BuildYourGuitar.com :: The Secrets of Electric Guitar Pickups
I see you are involved in the thread over there. Believe it all if you like, i don't. I know some things the original poster said are not true. I'm not dissecting everything all the other poster are saying either. Some are right but many others are just sheep and agreeing to EVERYTHING that is said.
The owner of that site and the other 20 websites he owns is a ssshole. You can't get in any type of disagreement or you are considered a trouble maker, low and behold if you are not a "dealer" and pay him $25/month. I'm a member there but once you read if for a while, you are better off staying on the sidelines.
In reality, i don't even see the point of the original post, he takes 2 pickups out of a pile and says they are alike yet they test different? Just had to rant.
Well not having anything specific to do with the project I originally wrote about, I have for some time been keenly aware that pickup loading can change the output response of any of my guitars. For that reason, I've been strongly considering building a tiny pre-amp board to be installed inside various guitars, specifically to offer as near to "no load" as i9s possible with the best in modern op-amps, and thus offering the player a situation where his/her guitar is going to sound its brightest into any load.
But as the OP, I just want to point out that my purpose here has been to discern some kind of standardized setup which will reliably help me predict the best I can expect from various pickups, especially the low cost kind that doesn't come with charts, graphs, and a significant bill. I think we are all keenly aware that the closer you can get to a no-load situation, the better your high end response. But being able to determine that, at least approximately on a bench without having to physically install is something that I'd find super valuable.
All that said, lets try to remember that all arguments likely have some intrinsic value, and remember that psychoacoustics has been debated for near a half century now. So I think that what is most constructive here are actual personal experiences and test set ups. All of them might be flawed in their own way, but at least none of us is trying to "sell" their ideas here.
But as the OP, I just want to point out that my purpose here has been to discern some kind of standardized setup which will reliably help me predict the best I can expect from various pickups, especially the low cost kind that doesn't come with charts, graphs, and a significant bill. I think we are all keenly aware that the closer you can get to a no-load situation, the better your high end response. But being able to determine that, at least approximately on a bench without having to physically install is something that I'd find super valuable.
All that said, lets try to remember that all arguments likely have some intrinsic value, and remember that psychoacoustics has been debated for near a half century now. So I think that what is most constructive here are actual personal experiences and test set ups. All of them might be flawed in their own way, but at least none of us is trying to "sell" their ideas here.
about 48 years ago, I rewound a bobbin over an alnico bar magnet with relatively heavy wire then used a cheap tiny mic matching transformer - it was quite "hi-fi" compared to say a regular stock P-bass pickup. I had no way to measure inductance and really didn't know what I was doing at the time - but it worked. My bass amp in those days was the Traynor YBA-3 Custom Special and IIRC, an 18" Goodmans's loaded short folded horn Traynor cabinet. Two of those cabinets (about 12 cubic foot each) side by side with that little head really sounded good in the local store. Do or did any players here experiment with quadraphonic bass pickups and rigs? My fingers have recently become "numb" so no guitar fun anymore.
Last edited:
Copyrighted material removed.I'm simulating what the pickup coil actually does in a guitar, not the measurement method you're using.
When a coil is exposed to a fluctuating magnetic field, a tiny voltage is induced in every tiny length of wire in the coil, adding up to a much bigger voltage across the two ends of the coil. (The relevant Maxwell's equations being: curl of E = - dB/dt).
This simulation should show the frequency response of the voltage generated by a pickup in response to a fluctuating AC magnetic field with, with dB/dt having a constant magnitude. That in turn is equivalent to a sinusoidal string vibration, with an amplitude falling linearly with frequency (- 6 dB/octave).
I am not sure if that last part is the proper frequency dependence or not, but anyone who has played a stringed instrument will have noticed that the bass strings vibrate through a much larger amplitude than the treble ones. We know that in reality the amplitude does fall with frequency, but I'm not sure if this is a -6 dB/octave rate or not.
(If all strings were equally thick, I think the amplitude would fall at -12 dB/octave - just look at the formula for total energy of a harmonic oscillator, which is proportional to the square of the amplitude. A constant energy input (pick stroke) should therefore result in an a vibrating amplitude that falls as the square of frequency. However, we don't use equally thick strings for the entire piano - and putting equal energy into a thinner string involves a larger amplitude of vibration. I think the use of progressively thinner strings has the effect of causing vibration amplitude (for a given picking impulse) to fall at -6 dB/octave.)
-Gnobuddy
Last edited:
I entirely agree with you that harmonics are crucial to the sound of a musical instrument. I just did a poor job of describing what I was trying to say earlier.
So let me try it again, in completely different words this time.
I quite regularly play plugged-in electro-acoustic guitars, and for some time now, I've been running my guitar straight into the mixer, via a 7-band graphic EQ pedal. I use the EQ pedal to subjectively improve the sound of my guitar through the PA.
The pedal I use is a Danelectro Fish-n-Chips, which has it's highest frequency slider centered at 6.4 kHz.
The guitars I'm discussing have onboard piezo pickups at the bridge, and onboard preamps that attempt to deliver a full-bandwidth guitar signal. The preamp manufacturer will likely have already incorporated some frequency-shaping into the preamp, to reduce the harsh treble that piezo pickups are known for.
Remember the guitar is being played through a normal full audio bandwidth P.A. system in this case. There are actual tweeters involved, unlike an electric guitar amplifier.
The first time I used this EQ pedal, I went through the exercise of raising one frequency slider at a time, and listening to the result. The idea was to educate my ears, so that in actual use, I would have a better idea what slider to move to achieve a specific change in sound.
What I heard - and still hear - is that raising the 6.4 kHz slider has a very dramatic effect on the guitars sound. But, it is a very bad effect. Even a good guitar sound nasty when EQ'd this way. I hear lots of scratchy, irritating transients and unwanted noises. Pick noises, fret noises, finger squeaks - all of it unpleasant when emphasized by an EQ peak in this frequency range.
I have tried the same experiment with my purely electric guitars. The 6.4 kHz slider has somewhat less audible effect there, because the guitar itself puts out less signal at these frequencies, and because the electric guitar amplifiers I use have a much more restricted treble bandwidth than a P.A. system does. There are no tweeters in electric guitar amps, and the speaker(s) are usually designed to roll off treble above 5 - 6 kHz very steeply.
Despite this, to me, raising the 6.4 kHz slider still makes the (electric) guitar sound worse. More scratchy, more harsh, etc.
Now, we have now had most of a century of electric guitar pickup development, with early one-of examples known to date from the 1920s if not earlier. During this time, pickups have evolved to make guitar sound good, among other things.
This is why it makes no sense to me for a significant number of electric guitar pickups to have been deliberately designed with a built-in EQ peak in the vicinity of 6 kHz. Why on earth would a majority of commercially available electric guitar pickups have evolved to emphasize a particularly bad-sounding part of the guitar's frequency spectrum?
What I've seen is that, when a typical 10' or 15' guitar cable is plugged into an electric guitar (with passive pickups), you see a resonant peak that is much lower in frequency, within the range that includes actual musical content from the guitar, and not just irritating, incidental sounds created by the playing techniques normally used.
I believe this is the frequency peak that actually influences the guitar sound in a beneficial way, brightening up the treble and adding some "shimmer" to the sound. Without this peak, electric guitars would sound rather dull and lifeless.
All of this discussion is somewhat complicated by the subjective nature of "good" sound. There may be people who like a very extended frequency response on their electric guitar - Les Paul himself was one of them, and he went to the trouble of winding his own low-inductance, low-impedance pickup coils for every guitar he used regularly. He also played only with a "clean" sound, with little or no added distortion from the amplifier.
But Les Paul walked a lonely road. I know of no other successful electric guitarist who imitated his sound. I myself find Les Paul's musical chops and guitar technique very impressive, but his guitar tone repulsive.
Perhaps you, Peter Pan, share Les Paul's taste in electric guitar sound. That is entirely your prerogative. If you like the sound of a 6 kHz EQ peak applied to electric guitar, so be it. Music is ultimately about personal enjoyment, not about uniformity.
However, for most guitarists and musicians, I can say with some confidence that they do not want to hear a guitar whose signal has been boosted in the 6 kHz region.
-Gnobuddy
The problem here is that to avoid complicating things, I began the thread with the single question of how to predict the high frequency characteristics of inexpensive pickups. I appreciate all you've added to the discussion, including your test methods. But these comments, however constructive, are straying from the question and this (understandably) is because, you don't know the details of my end project.
BUT... again, that's an aside. As it turns out, this all relates to a home made guitar project, in which one single coil pickup is employed, along with my own pre-amp circuitry. That pre-amp contains a +/- 20dB boost/cut control for the treble frequencies, starting at a moderate 1Khz (I've since lowered that to 800hz). During development, I noted that one pickup clearly had a brighter and more crisp sound when this treble EQ was raised, while the other gave me more of a "squeezing blood out of a rock" feeling when it came to treble. So as I attempt to productize this home brewed guitar, and knowing that all components (including pickup obviously) come at a cost, I simply wanted a way of predicting which pickups would do better in the only area I found SOME to be lacking... and that is high end. So if the pickup that "sounded better" (subjective, yes!) had a "peak" at around 7.5kHz, and the one that sounded "duller" peaked at around 6kHz, that is a piece of data, and I'm hoping its repeatable.
In the grand scheme of things, I have no particular intention of boosting actual musical frequencies around 6khz. I'm just noting that in my limited tests here, pickups with a higher peak sounded "better", at least in terms of having more of what I want to work with. Frankly, I like it better when i can get the sound I'm after without having my nice active treble EQ control boosted to max.
- Status
- This old topic is closed. If you want to reopen this topic, contact a moderator using the "Report Post" button.