|
Post by straylight on Jul 4, 2018 18:00:52 GMT -5
UPDATE: I've improved how I do this, scroll down to my post of 8th September 2018 for revised plots and numbers
I happen to have this early example of a DiMarzio Super2 on my bench. It was bought second hand by an associate in the late 80s under the impression it was a Super Distortion and whilst it has increased output over the stock pickups in the Jedson (a brand Hoshino Gakki pareent company of Ibanez) guitar it was installed in, it led to the guitar being neglected as it really did not compare well against an Ibanez Rocket Roll II with a factory fitted Dual Sound (3-conductor super distortion).
It should be noted that the DiMarzio/IBZ RG neck pickup is "very closely" modelled on this (I have one, will do a comparison shortly) and further similarites across Ibanez' own brand pickups suggest that the SuperD/Super2 combination with a basswood body and slim maple neck has been the template for The Ibanez Sound for a significant proportion of the brand's history*
It's in a plastic cover with some now quite corroded hex-socket pole pieces. These are the ubiquitous set screws with a 5/64" socket that are typical of Dimarzio "Super" family pickups. They do wind out with care, but note the pickups are filled with something resembling hot-melt glue so care is needed. It's difficult to see in any of the photographs, but the bobbins are DiMarzio's signiture double-cream.
Removing the plastic backplate reveals an oversize (i think 5/32") magnet and a strip of what looks like mild steel acting as a spacer. Later Super2s do not have this steel strip, instead having a wider magnet and I suspect this cheap slice of guillotined steel has been used to alter the inductance slightly to compensate for there being no backplate or screws present. Whilst the backplate proclaims "DiMarzio Pickups, Made in USA" I had to contact DiMarzio to identify the model, they were very helpful explainign that the earliest Super2s were all of this construction, but they switched to a more traditional backplate and open-bobbin appearance "very quickly".
It follows the Super Distortion pattern of a big magnet and larger polepices for output but isn't wound particularly hot. This one measures in at 8.35kOhm, 8.7 being quoted on their website as the current figure. I suspect that these pickups are specified in turns and this is a "standard" or gibson spaced model at 2 15/16" not an "F" spaced at (is it?)* 2 1/32" This is consistent with about 5300 turns of 42AWG by my calculations.
This particular example only has 80mm (3") of cable so apparent capacitance may be a little lower than expected.
* There several other voicing combinations that are quite recognisable and even more polarising but I think this is the sound most instantly recogonisable as being related the brand as a whole than a particular model, artist, or genre.
** Apologies if I'm not giving Imperial measurements in the proper form. I am part of the generation that was purposefully schooled never to work in anything but Metric/SI
|
|
|
Post by stratotarts on Jul 4, 2018 23:05:25 GMT -5
Thanks, that's a really interesting data point. It would be nice to have more information about the plot to help interpret it. It's hard to read the peak frequencies (loaded and unloaded, red and blue lines) from the chart, as the divisions are quite large. It's good to mention the value of any test load as it is impossible to compare with any other plots unless they are the same.
It's great to have another publishing tester in the field. We always need more.
You can really see the effects of the eddy current losses in the oversized steel poles, in the unloaded plot.
|
|
|
Post by antigua on Jul 5, 2018 20:03:46 GMT -5
A recently made Super Distortion has the spacer bar also. I'm not sure why they don't just make the magnet wider. I think from an efficiency standpoint, if you're going to increase the inductance, you want it to come from coil windings, or permeable pole pieces that magnetically couple with the strings more strongly. If the inductance increases for other reasons, the tone becomes darker but the voltage produced isn't increased.
|
|
|
Post by wgen on Jul 6, 2018 3:51:27 GMT -5
Thank you very much. I was expecting much more of a difference from the regular Super Distortion. This Super 2 has an inductance of 5.9H, the regular Super D has 6.4H, from the analysis of Antigua in this forum. It makes for a quite dark pickup anyway..! The Super 2 is often referred to a very trebly pickup on guitar forums. Anyhow, the Super 2 is just slightly above 8k DC resistance, while the Super D is 14.6k... Shouldn't the Super 2 have a lower inductance with that much difference of DC resistance? Or, the difference between different awg wires does have a role here, instead, so that the 2 pickups are more similar than how they might appear...?
|
|
|
Post by straylight on Jul 7, 2018 3:31:05 GMT -5
The Super2 works well as a neck pickup with a Super Distortion at the Bridge, it's brighter and cuts through the mix better than a traditionaly voiced PAF, it's quite bright compared to hot and fashionable 'neck' pickups like the Air Borton. I'm not sure how to describe it against a traditionally voiced paf after i just A/B tested, conventiently the bode data is waiting for me to analyse, but it's definatley got more midrange presence and sits in the mix differently. I confess most of the humbuckers I've owned have been hot and modern-voiced. Or just cheap and nasty.
I've been using the Dimarzio/IBZ RG set (which I understand the neck is based on the Super2) and will be analysing the set when the pickups on my bench go in to one of my Ibanez S Series. It's really clear in said Ibanez and cuts through when using a lot of gain so I'm not worrying about technical parts being lost in the mud. I had an Air Norton / Tone Zone set in the guitar previously which I found a bit too warm and muddy in a mahogany S series, both are better in brighter instruments. Conversely, the Super 2 is better behaved than the Evolution neck which really leaps out of the mix.
That said the IBZ pairing do not have the spacer on the magnet, so it's going to be interesting measuring them. Thanks for the modern Super D picture. I can't really drop the poles out of this to demonstrate just how much inductance comes from the poles, but I have a few pickups wound similarly kicking around that I can take apart, so we can investigate this.
Load is 470p/200k which i believe is your standard. Should have mentioned that explicitly. I think i should pretty-print a datasheet to go along with the plot.
I shall be modifying the peak-find code to return an x,y tuple rather than just the x value, so we can have the loess predicted magnetude which could denoise my sampling a little. Should i be valley-finding on the unloaded plot to build eddy-loss models?
Note the coloured smoothed line for each plot has a grey confidence band, i shall configure the plot so it's a white background with denser divisions. I'm a bit of an R/ggplot noob, and this is somehting of an exercise in being competent with R before I select it as the tool of coice of choice for other purposes. I used SPSS and Matlab when i was in mechEng which seem to be configured for extracting data from ugly charts. I think the R way (and the right way to do it if you can) is to extract the data programatically and print a plot that looks nice to impress mangement, shareholders and journalists. R has the bonus of being open source so I'm not having to resort to abusing a dayjob computer, spend big money or find alternate sources for Matlab/SPSS
|
|
|
Post by straylight on Jul 7, 2018 14:40:42 GMT -5
I may have to resort to a different plotting library as I'm having isues with getting enough y-axis gridlines. The problem might be between the the keyboard and the chair though. Should not have to take any readings from the plots though as I can calculate peaks and valleys. Need to deal with the valley finder as it's not as straightforward as finding a minimum, I need to look for gradientsgoing from -ve to +ve and find the minimum value this occours at.
R lets you shoot yourself in the foot with neat user-configurable multi-round burts and do statistical analysis on your accuracy, as long as you don't want to shoot yourself in the left foot, why would anyone want to do that?
|
|
|
Post by JohnH on Jul 7, 2018 17:15:49 GMT -5
Your plots look great! Just to respond about the information to be listed as specific data beyond what can be seen from plots:
Ive been a user of these data sets, rather than a maker of them, to derive equivalent analysis models based on L, R and C. These go into my spreadsheet GuitarFreak, see Reference section. The models I use have 6 parts, being three R's two L's and a C.
Obviously there's a heap of assumptions about whether the model has an inherent validity, but it works quite well in practice. To capture the 6 component values I use 6 pieces if info. Four of them are the amplitude and frequency of the two peaks. No 5 is the dc resistance (Request: please list this on plots). The 6th is one more point from the graphs to get a 'best match'. This one is harder to home in on. It tends to be a bit wooly, but close enough, to capture the dip and sometimes valley. But just for what I do, using the graph is ok, so I wouldnt worry too much about the exact valley values, which dont occur in all pickups anyway.
Pickups like this Distortion model, with significant eddy effects, are the hardest to match with such a model, but it usually gets within about 0.5 to 1 db across the frequency range.
|
|
|
Post by straylight on Jul 7, 2018 19:16:23 GMT -5
8.37kOhm at 22C, it's buried in the text. I think I'm going to have to produce them as a plot and block of text rather than trying to get all the data on the plots as my wishlist keeps growing and it's getting more and more cluttered. The more I automate plot production, the more I need to do things like the valley find anyway and it's a nice mental challenge to find ways of getting reliable information out of noisy data.
|
|
|
Post by antigua on Jul 8, 2018 12:32:36 GMT -5
Thank you very much. I was expecting much more of a difference from the regular Super Distortion. This Super 2 has an inductance of 5.9H, the regular Super D has 6.4H, from the analysis of Antigua in this forum. It makes for a quite dark pickup anyway..! The Super 2 is often referred to a very trebly pickup on guitar forums. Anyhow, the Super 2 is just slightly above 8k DC resistance, while the Super D is 14.6k... Shouldn't the Super 2 have a lower inductance with that much difference of DC resistance? Or, the difference between different awg wires does have a role here, instead, so that the 2 pickups are more similar than how they might appear...? The Super 2 must use 42AWG. A loaded peak of 2.3kHz makes it sort of a typical PAF bridge type of humbucker. The hotter ones such as the JB tend to be closer to 2.0kHz with a load. The near 6H inductance is high for a humbucker with 8k of 42AWG, so I suspect the large hex screws push the inductance up higher than would be had otherwise.
|
|
|
Post by antigua on Jul 8, 2018 12:41:54 GMT -5
I happen to have this early example of a DiMarzio Super2 on my bench. It was bought second hand by an associate in the late 80s under the impression it was a Super Distortion and whilst it has increased output over the stock pickups in the Jedson (a brand Hoshino Gakki pareent company of Ibanez) guitar it was installed in, it led to the guitar being neglected as it really did not compare well against an Ibanez Rocket Roll II with a factory fitted Dual Sound (3-conductor super distortion).
... I hope you'll think about getting a magnetometer, they're going for a little over $100 on Amazon www.amazon.com/dp/B07DCN3Z2G/ref=twister_B07DCPL7KL?_encoding=UTF8&th=1 I even see a couple new models that were not available a few months ago, including one called the TD8620 that even lets you display mT or Gauss . It would be interesting to see how much flux density there is at the pole tops, to see if it's about the same as the modern Super Distortion. Also you noted that details are buried in the text, it's helpful to have a "key: value" type list of all the measured specs if you're planning to survey a selection of pickups you have on hand.
|
|
|
Post by straylight on Jul 8, 2018 13:05:56 GMT -5
That particular item won't ship to england. Yes, eventually. Now I'm satisified with the CGM it's probably viable, and releveant as I'm buying alnico magnets uncharged. Curently on a polepiece material tangent as 14.9 steel appears to have all the inductance and and a big frequency spike before the cutoff, cheap potmetal polepieces appear to be responsible for the gentle curve into oblivion and I'm testing all the black oxide M3 and 4-40 pole screws I have because there's a nugget of usefulness here. Infodump coming.
|
|
|
Post by antigua on Jul 8, 2018 13:52:35 GMT -5
I'm not sure what the CGM is, or how it relates to flux density, but for example, I measure closer to ~200 gauss with humbuckers with AlNiCo 2 magnetic, and closer to 300G with AlNiCo 5, and closer to 400G with ceramic. I measured 450G (with a margin for error of about 15%) with the Super Distortion. I had thought these differences were trivial at first, but the evidence is suggesting it has a meaningful impact upon out the strings vibrate, and how guitarists perceive the tone of the pickup. The effects are so complex though that people tend not to describe them with any consistent terminology, such as "bright" or "dark", instead there just ends up being the "AlNiCo 2 sound" or the "AlNiCo 5 sound".
|
|
|
Post by straylight on Jul 9, 2018 8:01:09 GMT -5
CGM = Syscomp CGM-101, the usb osciloscope/bodeplotter/datalogger thing I shelled out for.
Alnico blends do things to the induction and eddy losses as I'm sure you know, so it's not just about magnet strenght. It's next on the list of things to buy along with a good L/C meter.
I think the steel strip is in the Super series because ceramic on it's own sounds different. I think it's deliberate inductance tuning rather than packing. I've got preliminary data on the similar IBZ pickups and they aren't as similar as claimed. The only visual difference is the steel strip is missing. I'm going to check the pole pieces aren't substantually different steel as that's a really sneaky way to get different sounds.
This morning's experiments seem to suggest that 14.9 steel is the sound of djent. Not quite what I was going for but I have a feeling my quick study into the screws in my parts tray is going to get extensive and may involve messing with heat treatment and then cold-blue. And a lot of trips to the metal fabricators. I think the key to having really good and hard to replicate pickups is having a consistent supply of well specified materials. Note DiMarzio and Seymour Duncan slugs have marks from being lathe-turned and appear steel-grey, yet the ubiquitous supply is smooth (presumably ground) and plated.
Antigua, have you got access to that SuperDistortion to run some plots? If you could help me out with an idea i might just be able to do something super cool with the data? If you have and can dump CSV or or similar of the bode plots done without an integrator I can do cool analysis. I'm working on producing good comarison analysis now I've (almost) got good single pickup analysis. This is the kind of raw data I'm working with
|
|
|
Post by straylight on Jul 9, 2018 8:18:16 GMT -5
Oooh, sneaky. The IBZ set screws are the same dimensions to within accepted tolerances for a screw thread, but stick to a magnet a little better. I don't know if that's varieance between batches or a different specification of steel.
|
|
|
Post by antigua on Jul 9, 2018 9:29:44 GMT -5
I don't mind producing a new plot for the Super Distortion, it's the one tested here guitarnuts2.proboards.com/thread/7737/dimarzio-super-distortion-analysis-review . The Velleman produces a tab delimited text file with output like this, imagine there are tabs in here: Hz Vrms dBV deg
100.0 11.306 21.045 200.0 11.357 21.084 300.0 11.279 21.025 400.0 11.064 20.857
I dont think the metal spacer is for inductance tuning, because I bet if you test this, you'd find that the bar only modifies the inductance by maybe 100mH, and my research into pickups tells me that makers of pickups don't really care about inductance to such a fine degree. For example, people talk about how base plates are placed on Tele bridge pickups to increase the inductance, but there again, it's in the are of a 100mH depending on the thickness of the steel, and it's more likely its intention was to act as a keeper for the AlNiCo, and/or shielding for the coil. I'm sure in the case the idea was to preserve the magnetic circuit while making use of a magnetic with "standard" width. I'm not sure what you're observing about the steel, but my own testing showed a variety of common alloys didn't produce any real differences guitarnuts2.proboards.com/thread/8036/addiction-steel-alloy-electrical-evaluation , and this is to be expected because of how much of the magnetic circuit consists of air gap. When it comes to eddy currents and steel parts, since most pickups on the market use the same overall design, you can generally tell just by looking at the construction what degree of damping is to be expected. I've gathered data for most of the popular pickup layouts, and a handful of odd ball Gretsch types from TV Jones, and there's usually a rather direct relationship between the damping and the volume of steel, and AlNiCo to a lesser extent, and how wholly it overlaps with the magnetic field of the coils, so of course the cores are most decisive, followed but steel around the perimiter.
|
|
|
Post by stratotarts on Jul 9, 2018 22:33:59 GMT -5
DiMarzio has patented a lot of stuff that they do. Often the explanations in the patent, when you strip away the colourful legal language, are dubious or vague. Not to directly accuse them of it, but a lot of companies that can afford patents engage in the acquisition and trade of "patent real estate". The idea is to claim patents on anything you can, regardless of how well it works. If someone comes along and creates something similar that does work, or has to work in another way while incorporating the same feature, bingo! you own it and you can cut them down.
|
|
|
Post by straylight on Jul 16, 2018 20:06:33 GMT -5
I need to double and triple check my numbers as I've been measuring the inductance and capacitance off the integrated plot and I don't think that's giving me consistent numbers. Need to math, and see what correlates well with my meter and and published data. Sorry I'be been quiet on this, have been away. But It looks like i'm getting some really big swings in inductance by using hardend steel hex-key machiene screws.
I've just been to see the client whose super 2 has been on the meter. He's had my test guitar with some prototpe pickups in and he's sent me away with a car full of guitars whilst he's on holiday. There's a Rocket Roll II in just to put on the meters. I think it's got a Dual Sound which is a 3-wire Super Distortion but I need to check specs, will be handy to correlate my data against another known thing, Also an Maverick X1 with those alnico rod magnet humbuckers.
Yes, I can use the velleman data in tab delimited format, and having what I think is a super distortion to hand as well as an IBZ almost super distortion, I think I can get some really good comparisons.
I'm running plots unloaded, with a 200kOhm and 470pF load, and with a 10nF load without an integrator if you can duplicate that, it's enough. If you want to compare hardware maths to software maths, I could do something with integrated plots as well in terms of validating my maths but on their own I have to write more code to get a comparison out. I'd really appreciate it if you could post those. I'm using the smallest frequency step possible to get the most data points, it makes a straight plaot less smooth, but I'm more confident in a plot I hit with math.
When my R code is less awful, works on velleman data, and produces individual plots, comparisons and datasheets without having to tweak plotting parameters each time, I'll happily make it available so what I've done can be reproduced, right now it's a mess some fiddling about is required.
|
|
|
Post by straylight on Sept 8, 2018 13:07:28 GMT -5
Ok, I've rewritten the code some in preparation for producing comparison plots. I'm now calculating inductance and compacitance from the raw plots and marking and calculating resonant peaks and cutoff frequencies from the integrated data. I'm including raw peaks for comparison with other analysis produced in the same way. I've chosen to output numbers seporately rather than on a plot. The plots need some legends and maybe some further display tweaks, but I'm happy with the data production.
[1] "Dimarzio Super 2" [1] "Calculated Inductance: 5.49 H" [1] "Calculated Capacitance: 56.6 pF" [1] "Loaded Cutoff: 3655 Hz" [1] "Loaded Resonant Peak: 2364 Hz 0.647 dB" [1] "Raw Loaded Peak: 2998 Hz 11.8 dB" [1] "Unloaded Cutoff: 9089 Hz" [1] "Unloaded Resonant Peak: 7605 Hz 0.934 dB" [1] "Raw Unloaded Peak: 7757 Hz 21.4 dB" [1] "Raw Induction Peak: 679 Hz 6.1 dB"
In each case the solid vertical lines intersect points identified from the plotted data and the dashed ones are produced elsewhere.
Red=Loaded 200k, 470pf Blue=Unloaded Green=Inductance test
The integrated plot shows cutoff frequencies and resonant peaks.
The raw plot is not particularly useful other than to verify peak detection for capacitance and inductance calculation.
|
|
|
Post by antigua on Sept 9, 2018 18:01:05 GMT -5
This is a really good analysis, thanks for posting this. Your method of taking raw CSV data and plotting it separately has a lot of advantages over relying on the plotters build into the USB oscilloscopes.
|
|
|
Post by straylight on Sept 11, 2018 21:11:01 GMT -5
Single pickup analysis is up on github, runs in R. If you've ever used matlab or can set a few values in a text file you can replicate this for any Syscomp produced data or data you edited in excel to look like the syscomp data. If I have some files from a velleman I will make the code read tht as well.
I'm not sure it's a good idea to release experimental data through github, but if anyone wants I can probbly find somewhere to put my CSV online.
|
|
|
Post by straylight on Sept 11, 2018 22:03:58 GMT -5
I'm still not happy with how multi-pickup comparisons are done, as each pickup is currently gain-normalised individually which isn't the behaviour I'd like (I think). But this shows the Super 2 in context with some vintage style humbuckers. The output should be more across the board as the super2 has a huge ceramic magnet but the cutoff is quite similar to a Prototype 22, a vintage style humbucker with 5150 turns of 42AWG, a regular size A5 magnet, and a nickel silver baselate.
Prototype 20 is similar but with 5000 turns of 42AWG and a nickel silver cover. The Super 70 is an Ibanez replica of a 70s gibson humbucker, unspecified alnico magnet, open with a brass backplate.
The Jedson pickup is a Hoshino/Fujigen abomination from the early 70s with a brass baseplate and cover.
Yeah, I need to get my hands on good plots of something i consider a good reference.
|
|
|
Post by antigua on Sept 12, 2018 10:28:57 GMT -5
In my plots made with an external exciter coil, I would usually not normalize the output, except to make the loaded and unloaded plots start out the same. The load drops the output by about 1dB, so that was normalized for better looking plots, but the bridge versus neck comparisons were not normalized, so that if the higher wound bridge has a higher voltage output, it would be more apparent. In most cases a bridge pickup that was overwound relative to a neck pickup would only have an output that was anywhere from 1dB to 3dB greater, and in most cases the pickups are of the same construction otherwise, same magnet, same pole pieces. The intent of a hotter bridge pickup is more likely to be about rolling off treble than it is about achieving balanced output levels.
And while the permeability of the magnet and pole pieces plays a role in the output levels that are plotted, this testing method completely excludes the B value of the moving string, as there is no string. If this testing were done directly, as a simple impedance test of the pickup, the relative outputs would be even less meaningful. I think in that sort of test, you would show less output, due to a higher series resistance, even though the pickup would be louder in situ. In that case it might be best to normalize the output in order to avoid misleading reads of the plots.
These plots are ultimately mostly useful for comparing transfer functions. IMO, it makes more sense to estimate the anticipated output by observing the reluctance of the magnetic circuit, the strength of the magnet at the tops of the pole pieces, and the density of coil in close proximity to the strings. Of course that's all estimation, which is not very scientific, but in order to get a real value you'd have to incorporate and actual guitar string and a rigid set of standards to say that for fixed input A with the qualities B,C,D.. you get output X. From a pragmatic standpoint, because the output does vary so much due to the pickup's distance from the strings, high precision with respect to output versus input is not required. Since most pickups of a given type have identical magnetic circuits, and only vary in terms of coil turn counts, you could probably draw a close correlation between turn count and output, where the turn count has been estimated by taking the DC resistance and wire gauge into account.
|
|
|
Post by straylight on Sept 12, 2018 13:29:24 GMT -5
Yeah, I took a big shortcut by taking the mean output across the flatish section of the output and taking this value away from the entire data set to get a plot that neatly crossed zero, allowing me to use a single plot setup when I had wildly differing data from a variety of sources. I should be able to store and extract this gain offset from an individual pickup and then when i'm running a comparison rather than compring a list, the behaviour should be load a reference pickup and then use the gain offset from that in the comparison pickups loaded. It's not too tricky, but it's been on the back burner as my current work has been about reproducing the desired frequency response with a reducced output as I'm trying to get the shred sound from a covered pickup without obscene string-pull to address isues with sustain, wolf tones and enthusiastic playing on light strings destroying pickups. There will be more to come out of this.
|
|
|
Post by wgen on May 25, 2019 7:40:16 GMT -5
I'm updating this analysis only to ask about something related. I'm actually interested in the Super 2 or the Super Distortion , so I checked out the analysis bere for the super 2, and the one of the Super distortion here by Antigua : guitarnuts2.proboards.com/thread/7737/dimarzio-super-distortion-analysis-reviewFact is, from the graph right above my message here, and the graph from the thread of Antigua...they almost seem to have the same response! Maybe, only 1db brighter at 4Khz for the Super 2 when compared to the Super Distortion. Interested in hearing from you about this subject...Thank you in advance! Edit: I'm specifically considering the graphs with load, when comparing
|
|
|
Post by antigua on May 25, 2019 14:07:04 GMT -5
I'm updating this analysis only to ask about something related. I'm actually interested in the Super 2 or the Super Distortion , so I checked out the analysis bere for the super 2, and the one of the Super distortion here by Antigua : guitarnuts2.proboards.com/thread/7737/dimarzio-super-distortion-analysis-reviewFact is, from the graph right above my message here, and the graph from the thread of Antigua...they almost seem to have the same response! Maybe, only 1db brighter at 4Khz for the Super 2 when compared to the Super Distortion. Interested in hearing from you about this subject...Thank you in advance! Edit: I'm specifically considering the graphs with load, when comparing The inductance calculated for the Super 2 is 5.49H, and the I measured my Super Distortion, showing 6.570H, so the Super 2 is still on the hotter side for a humbucker, what would be considered a "bridge" pickup. Even so, they shouldn't overlap that closely. The way Straylight estimated the loaded peak, the reference marker is a little bit left of where I place it, which is closer to the drop off. I think if it were my estimation, the peak would be close to 2.4kHz. Also, in my plot there was no observable resonant amplitude with the Super Distortion, because the Y axis is not as high of resolution as Straylight's, and so I might have placed the marker to far to the right, and maybe it more realistically would cut off at 2.2kHz if there was an observable resonance. tl;dr, the Super D might be closer to 2.2kHz, the Super 2 might be closer to 2.4kHz. Straylight plot My Super D plot:
|
|
|
Post by straylight on May 28, 2019 14:24:37 GMT -5
I'm using R's built in algorithms for finding peaks, cutoff frequency is where the 0dB referenced line crosses -3dB which I think it a reasonable approximation to finf the level before the peak.
I'm not spotting the peak from the plot, I'm doing a loess prediction on the data for a bit of smoothing and noise removal (the coloured line, the data is in the grey band or black line somtimes seen in addition, often the coloured line completely hides this if I have noise free data. I'm then programatically scanning the data for peaks. It might not be perfect, but it's pretty close and it's repeatable giving me the same result every time. I'm working hard on having R batch process data so I can get a comparison plot for a lot of pickups for fresh data done the same way each time. For me, consistency of data is really important.
The super 2 is monstously loud and thick, designed by dimarzio to pair with a superD in the bridge. The Dimarzio/IBZ RG set is the modified superD first licensed to maxxon for the early 80s V2 pickup as fouind in the Rocket Roll II and early Destroyer (like the red one made famous by Iron Maiden's Adrian Smith). The original Evolution has a very simialr frequency response. (I need to publish that don't I). The Dimarzio/IBZ RG neck pickup is very close to the super2 and the evolution neck pickup is again very similar. Whilst it was used in the beidge of the guitar it came out of, I didn't feel it was particularly suited for bridge use where guitarists expect the full on superD/Evo/Ibanez style thick sound. Conversely I use an Air Norton as a bridge pickup in one of my stratty guitars, despite it seeing most installation as a neck pickup paired with a tone zone. Your mileage may vary.
|
|
yanyan
Meter Reader 1st Class
Posts: 52
Likes: 2
|
Post by yanyan on Nov 16, 2021 2:31:48 GMT -5
Hi straylight if it's not too much trouble may i ask for a favor? I would be very interested to see a detailed comparison between your Super 2 from 1977 and a brand new Super 2, in particular differences (if any) in inductance, gauss, and resonant peak.
This is of interest to me now because i recently converted my Strat to HSS with a brand new Evolution bridge in the bridge. The neck and middle pickups are a Fast Track 1 and Chopper, respectively -- which i bought in 1996 and are now 25 years old. I'd like to see how much pickups change as they age, and relatedly how much of an imbalance exists between my old and new pickups.
|
|
|
Post by antigua on Nov 17, 2021 13:58:54 GMT -5
Hi straylight if it's not too much trouble may i ask for a favor? I would be very interested to see a detailed comparison between your Super 2 from 1977 and a brand new Super 2, in particular differences (if any) in inductance, gauss, and resonant peak.
This is of interest to me now because i recently converted my Strat to HSS with a brand new Evolution bridge in the bridge. The neck and middle pickups are a Fast Track 1 and Chopper, respectively -- which i bought in 1996 and are now 25 years old. I'd like to see how much pickups change as they age, and relatedly how much of an imbalance exists between my old and new pickups.
If you mean age in the physical sense, I think the plastic might yellow, but the ceramic magnets, steel parts and copper wire wouldn't change over time. The insulation on the magnet wire and break down over time, but I don't think it fails outright. A lot of pickup seem to spontaneously break over time, I think what happens there is that the copper coil expands and contracts due to hot and cold cycles, and it eventually breaks internally, but aside from this breaking, the inductance of the coil would be unchanged up until that moment, since the inductance relates to the number of turns, not the condition of the copper wire. The company might have made a decision to tweak the design during the past forty years, I think that would be a more likely cause of a difference.
|
|
|
Post by wgen on Feb 20, 2024 8:12:51 GMT -5
This video made me think about this thread. I was listening to the Super 2 in this comparison, it seems so trebly, almost twangy, Fendery type of sound. In the analysis here, it should behave like a loud PAF bridge humbucker, but I'm not hearing that. It almost reminded me of the EJ custom Dimarzio humbucker I had in the past, which was voiced to provide a sound similar to Filtertron pickups.
I was wondering what could be the reason, if the larger magnet and higher Gauss values these Super Distortion pickups have, when compared to more traditional PAF pickups, have a role here. The influence of higher string pull really seems to be important. Do they emphasize higher order harmonics?
|
|
|
Post by antigua on Feb 20, 2024 14:07:45 GMT -5
The influence of higher string pull really seems to be important. Do they emphasize higher order harmonics? String pull and magnetic saturation in the guitar string both happen with stronger magnetism, and both promote higher harmonics. The string pull does this by moving the energy from the fundamental into the partials, just like a pinch harmonic. In the case of saturation, the magnetic polarization of guitar string becomes increasingly acute towards the centers of the pole pieces, making higher harmonics more magnetically visible to the pickup.
|
|