|
Post by antigua on Mar 13, 2023 12:40:26 GMT -5
The EMG 81 and 85 should have nearly the same curve, the only difference afaik is ceramic versus AlNiCo magnets, and the response is dome shaped when plotted out, with the top of the dome being in the area of 2kHz, so 1.87kHz could be about right. Given the low Q factor and the dome like profile, the resonant peak of an EMG isn't as significant as it is when comparing passive hi-Z pickups. When people say EMGs have a modern sound, that dome like response curve is likely to blame, and I think EMG has newer models which I haven't looked at, which more closely mimic as passive high Z pickup, and it would be interesting to observe the resonant peak of those pickups, to see what sort of passive pickup spec they aim to sound like. Wait, I don't understand. From the below graph, should I assume the resonant freq of EMG 85 is 643Hz instead of 1.87k? Thanks. Technically it's at a peak around 643Hz, but it's rather flat, from 200Hz up to almost 2kHz.
|
|
|
Post by ms on Mar 13, 2023 14:50:39 GMT -5
I think that falls off too slowly on the high frequency side to qualify as a resonance.
|
|
dako
Rookie Solder Flinger
Posts: 9
Likes: 0
|
Post by dako on Mar 14, 2023 1:19:07 GMT -5
Wait, I don't understand. From the below graph, should I assume the resonant freq of EMG 85 is 643Hz instead of 1.87k? Thanks. Technically it's at a peak around 643Hz, but it's rather flat, from 200Hz up to almost 2kHz. I see. In this case, I think their number might not accurately show how their freq response curve gonna look like. Btw, Fishman stated that the resonant freq of their modern ceramic pickup is "Voice 1: 720Hz, Voice 2: 1.8kHz". And based on the measurement you made before, Fishman's numbers are much more close to the peaks of freq response curve and make more sense to me. I hope EMG could adopt this method.
|
|
dako
Rookie Solder Flinger
Posts: 9
Likes: 0
|
Post by dako on Apr 2, 2023 2:00:16 GMT -5
Hi Antigua, great information! From the plot, I see that the VLPF retained the high frequency signals and flattened the peak. In the past, I have tried 1M vol pot without tone pot for both single coil and humbucker. While this wiring retained more high frequency signals, it also makes the resonant peak too high and the treble harsh, I believe. You mentioned what VLPF does can be accomplished by passive circuit, could you elaborate on that? I am not aware of a way to retain high frequency signal while tame down the peak like what VLPF does here. I was just referring to the fact that it rolls off the treble. It has the upside of buffering the pickup(s), but you could just use some sort of buffer besides, and use regular tone controls instead. Personally, I use a wireless transmitter when I play, the kind from Boss and Line 6, which removes the guitar cable capacitance, so I get a nearly unloaded resonant frequency that way also. Hi Antigua, have you tried to measure the capacitance of the input of wireless transmitters you own? Do you think cheaper transmitters like from Joyo would also remove the guitar cable capacitance?
|
|
|
Post by antigua on Apr 4, 2023 0:03:41 GMT -5
I was just referring to the fact that it rolls off the treble. It has the upside of buffering the pickup(s), but you could just use some sort of buffer besides, and use regular tone controls instead. Personally, I use a wireless transmitter when I play, the kind from Boss and Line 6, which removes the guitar cable capacitance, so I get a nearly unloaded resonant frequency that way also. Hi Antigua, have you tried to measure the capacitance of the input of wireless transmitters you own? Do you think cheaper transmitters like from Joyo would also remove the guitar cable capacitance? Yeah I did measure the resulting resonant peak with one unit guitarnuts2.proboards.com/thread/8242/capacitance-line-6-relay-g10 , but I haven't tested all of the ones I have on hand. In general, if they don't specifically state that they include simulated cable capacitance, or have a switch to add line capacitance, then there is no added capacitance, and you basically have a zero length guitar cable. The Joyo model makes no mention of cable simulation, so I'd assume there is none. To my ear, that's what the lower end wireless units tend to be, bright and clear. I have to usually roll my tone controls down to six or seven to shed some unwanted high end, which is less of a problem when using a real cable. I can't just turn the treble down on the amp, because I still want my overdrive clipping to have some bite in the treble end. If you wanted to add in your own parallel capacitor, it would be a pretty easy mod, you just gave to put about 470pF across the circuit. Like you could get a sin inch male to female 1/4 extension cable, put it in between the wireless unit and the guitar, and solder the capacitor to the male or female connector, inside where the cable is soldered to the jack. It would be like a virtual 10ft cable.
|
|
dako
Rookie Solder Flinger
Posts: 9
Likes: 0
|
Post by dako on Apr 9, 2023 21:47:08 GMT -5
Hi Antigua, have you tried to measure the capacitance of the input of wireless transmitters you own? Do you think cheaper transmitters like from Joyo would also remove the guitar cable capacitance? Yeah I did measure the resulting resonant peak with one unit guitarnuts2.proboards.com/thread/8242/capacitance-line-6-relay-g10 , but I haven't tested all of the ones I have on hand. In general, if they don't specifically state that they include simulated cable capacitance, or have a switch to add line capacitance, then there is no added capacitance, and you basically have a zero length guitar cable. The Joyo model makes no mention of cable simulation, so I'd assume there is none. To my ear, that's what the lower end wireless units tend to be, bright and clear. I have to usually roll my tone controls down to six or seven to shed some unwanted high end, which is less of a problem when using a real cable. I can't just turn the treble down on the amp, because I still want my overdrive clipping to have some bite in the treble end. If you wanted to add in your own parallel capacitor, it would be a pretty easy mod, you just gave to put about 470pF across the circuit. Like you could get a sin inch male to female 1/4 extension cable, put it in between the wireless unit and the guitar, and solder the capacitor to the male or female connector, inside where the cable is soldered to the jack. It would be like a virtual 10ft cable. Thanks for the link. The information is really useful. About the extra treble, I actually modded my guitar to have no-load pot for tone so I generally roll-off some treble all the time. Do you think a buffer or a wireless units will have similar capacitance?
|
|
|
Post by antigua on Apr 11, 2023 0:08:09 GMT -5
Do you think a buffer or a wireless units will have similar capacitance? A buffer that's built into the guitar would have the lowest capacitance most likely, since there's such a short distance between the pickups and the buffer. Fender pickups usually have unshielded wire, the added capacitance is effectively zero in that case, but Gibson style guitar usually has a shielded cable connecting the pickups to the switch, and the switch the control pots, I measured 75pF for 1 foot of Gibson style braided wire, so if you assume 1 foot from pickup to controls, that's 75pF, but if you have a Les Paul, there's a three way trip from pickup to controls, controls to switch , switch to jack, that's probably about two and a half feet, or ~200pF capacitance just between the pickups and the output jack. Usually when I've had to account for capacitance in a buffered input for the sake making a calculation, the capacitance is between 10pF and 20pF, due parasitic capacitances in the circuit which can't be eliminated, I think you'd see about the same range with a wireless unit that plugs into the jack of the guitar. If it's the type of wireless unit that runs to the unit you hand off your pants / belt, a two foot patch cable from the jack to the wireless unit probably adds 60 to 80 pF capacitance, depending on how thick the patch cable is.
|
|
dako
Rookie Solder Flinger
Posts: 9
Likes: 0
|
Post by dako on Apr 11, 2023 15:50:13 GMT -5
Do you think a buffer or a wireless units will have similar capacitance? A buffer that's built into the guitar would have the lowest capacitance most likely, since there's such a short distance between the pickups and the buffer. Fender pickups usually have unshielded wire, the added capacitance is effectively zero in that case, but Gibson style guitar usually has a shielded cable connecting the pickups to the switch, and the switch the control pots, I measured 75pF for 1 foot of Gibson style braided wire, so if you assume 1 foot from pickup to controls, that's 75pF, but if you have a Les Paul, there's a three way trip from pickup to controls, controls to switch , switch to jack, that's probably about two and a half feet, or ~200pF capacitance just between the pickups and the output jack. Usually when I've had to account for capacitance in a buffered input for the sake making a calculation, the capacitance is between 10pF and 20pF, due parasitic capacitances in the circuit which can't be eliminated, I think you'd see about the same range with a wireless unit that plugs into the jack of the guitar. If it's the type of wireless unit that runs to the unit you hand off your pants / belt, a two foot patch cable from the jack to the wireless unit probably adds 60 to 80 pF capacitance, depending on how thick the patch cable is. Oh yeah, I saw your other post listing capacitance of different components like selector, shielded wire, etc. I didn't know shielded wire had that much capacitance. In the past I had thought about using shielded wire to improve noise performance but now think back on it, it's a terrible idea. I checked your previous post regarding the wireless unit's capacitance, I remember it was around 100pf. I used LTSpice for some test and found out that 100pf can decrease the resonant frequency from 10k to 7k for a single coil, which is kind of significant. And comparing it to the active tone control from EMG, which doesn't decrease resonant frequency much, I wonder if the buffer's input is significantly different from a wireless unit. My naive thought is that a wireless unit is basically an amp with ADC and encoder. The amp input should be a pure resistant input. So why a buffer can maintain the resonant frequency so perfectly but a wireless unit less so? Could you share any thoughts on this? Thanks!
|
|
|
Post by antigua on Apr 11, 2023 18:27:39 GMT -5
I checked your previous post regarding the wireless unit's capacitance, I remember it was around 100pf. I used LTSpice for some test and found out that 100pf can decrease the resonant frequency from 10k to 7k for a single coil, which is kind of significant. And comparing it to the active tone control from EMG, which doesn't decrease resonant frequency much, I wonder if the buffer's input is significantly different from a wireless unit. My naive thought is that a wireless unit is basically an amp with ADC and encoder. The amp input should be a pure resistant input. So why a buffer can maintain the resonant frequency so perfectly but a wireless unit less so? Could you share any thoughts on this? Thanks! I forgot that the Line 6 unit brought down the resonant peak so much, having approximately 120pF capacitance, but it's possible that they designed it to add capacitance on purpose, or its possible that the parasitic capacitance is just very high because of how they designed the enclosure and circuit, if they were willing to tolerate what might usually be bad circuit design for the sake of making it smaller and cheaper. 10kHz to 7kHz isn't real significant insofar as the guitar amp speaker's range drops off around 5kHz, which you see if you look at specs on the Celestion or Jensen sites. Even if you could hear it, it would be a difference of very high sibilance, but once the frequency gets below 5kHz, it starts making a noticeable difference in the transient sound and the overall sound. Like the difference between Texas Specials and 57/62's, or 500 ohms worth of wire added or removed, has guitarists preferring one or the other. It's not just the frequency alone though, if the Q factor is high it will draw a lot of attention to the resonant peak, but if the Q factor is low, it will emphasize a broader range of the treble frequencies and have a softer roll off, and so a lower resonant peak with a lower Q factor might end up seeming as though it sounds clearer. I think I'm sensitive to the resonant frequency, because one by one I've been swapping in pickups with higher resonant peaks, and working the tone control to flatten the response out, and when I plug in a guitar with pickups having lower resonant peaks, I notice I'm fighting with it more, if I turn the tone control up, the higher Q lends to shrillness, but going the other way can make it too dark. When I first got into pickup swapping I was mostly using solid state modeling amps, I also noticed that when I started using more vintage style Fender tube amps with less sensitive speakers, pickups that once seemed very different, became harder to tell apart, like the amp and speakers were providing more of the color and overriding the curve inherent to the pickup. It's a series of successive filters after all.
|
|
timtam
Meter Reader 1st Class
Posts: 53
Likes: 24
|
Post by timtam on Apr 12, 2023 2:44:56 GMT -5
A buffer that's built into the guitar would have the lowest capacitance most likely, since there's such a short distance between the pickups and the buffer. Fender pickups usually have unshielded wire, the added capacitance is effectively zero in that case, but Gibson style guitar usually has a shielded cable connecting the pickups to the switch, and the switch the control pots, I measured 75pF for 1 foot of Gibson style braided wire, so if you assume 1 foot from pickup to controls, that's 75pF, but if you have a Les Paul, there's a three way trip from pickup to controls, controls to switch , switch to jack, that's probably about two and a half feet, or ~200pF capacitance just between the pickups and the output jack.
Zollner recently highlighted a 44cm length of Gibson braided-shield wire to the pickup selector that had a capacitance of 514 pF, or 1165 pF/m ! So that very much belongs on the list of things that might contribute to the supposed sonic differences between LPs and SGs, even with the same pickups ... while the LP's pickup selector is a long cable length away from the pots, the SG's is right next to the pots.
(turn on CC and auto-translation to English)
|
|
dako
Rookie Solder Flinger
Posts: 9
Likes: 0
|
Post by dako on Apr 16, 2023 0:14:40 GMT -5
I checked your previous post regarding the wireless unit's capacitance, I remember it was around 100pf. I used LTSpice for some test and found out that 100pf can decrease the resonant frequency from 10k to 7k for a single coil, which is kind of significant. And comparing it to the active tone control from EMG, which doesn't decrease resonant frequency much, I wonder if the buffer's input is significantly different from a wireless unit. My naive thought is that a wireless unit is basically an amp with ADC and encoder. The amp input should be a pure resistant input. So why a buffer can maintain the resonant frequency so perfectly but a wireless unit less so? Could you share any thoughts on this? Thanks! I forgot that the Line 6 unit brought down the resonant peak so much, having approximately 120pF capacitance, but it's possible that they designed it to add capacitance on purpose, or its possible that the parasitic capacitance is just very high because of how they designed the enclosure and circuit, if they were willing to tolerate what might usually be bad circuit design for the sake of making it smaller and cheaper. 10kHz to 7kHz isn't real significant insofar as the guitar amp speaker's range drops off around 5kHz, which you see if you look at specs on the Celestion or Jensen sites. Even if you could hear it, it would be a difference of very high sibilance, but once the frequency gets below 5kHz, it starts making a noticeable difference in the transient sound and the overall sound. Like the difference between Texas Specials and 57/62's, or 500 ohms worth of wire added or removed, has guitarists preferring one or the other. It's not just the frequency alone though, if the Q factor is high it will draw a lot of attention to the resonant peak, but if the Q factor is low, it will emphasize a broader range of the treble frequencies and have a softer roll off, and so a lower resonant peak with a lower Q factor might end up seeming as though it sounds clearer. I think I'm sensitive to the resonant frequency, because one by one I've been swapping in pickups with higher resonant peaks, and working the tone control to flatten the response out, and when I plug in a guitar with pickups having lower resonant peaks, I notice I'm fighting with it more, if I turn the tone control up, the higher Q lends to shrillness, but going the other way can make it too dark. When I first got into pickup swapping I was mostly using solid state modeling amps, I also noticed that when I started using more vintage style Fender tube amps with less sensitive speakers, pickups that once seemed very different, became harder to tell apart, like the amp and speakers were providing more of the color and overriding the curve inherent to the pickup. It's a series of successive filters after all. Yes, I agree with you. The high resonant peak is very hard to fix in the post processing, so I prefer pickups with flatter response. Btw, when you use modeling amps, do you use cab simulations? They generally rolls off a lot of high frequency contents. I don't think there is significant difference between a tube amp vs solid state amp, the diff to me is coming from the tone stack and the cab/speaker. Have you tried using your modeling amp with middle output speakers like G12M?
|
|
|
Post by mikecg on Aug 2, 2023 19:02:01 GMT -5
Hello Antigua, This is only my second post here, so let me start by saying that I think you are doing a great job on shedding some light on the 'black art' of guitar magnetic pickup design! With regard to the Chinese active pickups (EMG look alikes?) you feature in this post - I believe your hunch concerning their construction is quite correct. I have assumed that these 'things' are - like most Chinese made products - built with a firm eye on cost. I have developed an electronic model that accurately simulates your measurements for the version with the 'flat' frequency response. The model is based on your hunch that the pickup consists of two (cheap) 'under-wound' humbucker coils, connected in parallel, and buffered by a single (cheap) battery powered opamp. The model shows that the 'underwound' humbucker, buried inside the black potting compound, is likely to have two coils with an inductance of around 1 H, a correspondingly reduced DCR of around 1.4 k Ohm, and a shunt capacitance of around 120 pf. The opamp is most likely a TL061 or equivalent - a cheap and cheerful part that is happy working at 9 Volts - and with a miserly current consumption. Here is the simulated frequency response assuming your standard load of 200 k Ohm, and 470 pf: And here is the circuit diagram: But note that these active pickups are designed to be used with 25k volume control pots, and so, because they have a 10 k ohm series output resistor in circuit, the output level will be correspondingly reduced. When the 200 k Ohm load is replaced by a 25 k Ohm load, the the output level drops to the 'Jazz' level. This may explain why one of your respondents claims that his (or her) EMG sounds quieter than the JB. Here is the simulated frequency response assuming an active pickup load of 25 k Ohm, and 470 pf: And here is the modified circuit diagram: I think this is an interesting pickup design - a wide flat frequency response is often regarded as the 'Holy Grail' in the audio world, and could perhaps be regarded as the perfect starting point for arbitrary downstream EQ frequency response shaping.
|
|
|
Post by mikecg on Aug 3, 2023 10:37:55 GMT -5
I've just spotted a couple of small errors on the previous circuit diagrams - luckily - they don't make any noticeable difference to the frequency plots. The first is the DCR value of the humbucker coils and they should be 2.8 k Ohms for each coil, not the 1.4 k Ohms, as shown in the diagrams. The second is that C7, the output coupling capacitor in the first circuit diagram is shown as 0.1 uf, and this is adequate for the standard load of 200 k Ohm and 470 pf, whereas C7 in the second circuit diagram is shown as 1 uf, and this is also adequate for the lower resistive load of 25k and the standard capacitive load of 470 pf.
From my basic understanding of the inherent hum cancelling properties of the humbucker magnet configuration, it would appear that wiring the two coils in parallel, start to start, and finish to finish, as shown in the circuit diagrams, should preserve the hum cancelling properties - i.e. the inherent magnetically coupled common mode rejection of the humbucker configuration.
|
|
|
Post by ms on Aug 3, 2023 14:23:08 GMT -5
From my basic understanding of the inherent hum cancelling properties of the humbucker magnet configuration, it would appear that wiring the two coils in parallel, start to start, and finish to finish, as shown in the circuit diagrams, should preserve the hum cancelling properties - i.e. the inherent magnetically coupled common mode rejection of the humbucker configuration. So the coils are wound in opposite directions?
|
|
|
Post by mikecg on Aug 3, 2023 17:03:50 GMT -5
No - sorry for the confusion - I guess it depends on how one defines the start and finish of each coil - no doubt there is a convention among those in the know? I hope this drawing helps: 'A' connects to 'D', and 'B' connects to 'C'
|
|
|
Post by mikecg on Aug 3, 2023 18:04:17 GMT -5
My definition of 'start' and 'finish' for humbucker coils:
|
|