Announcement

Collapse
No announcement yet.

Variable Resonance Pickups - Fun with Buffers

Collapse
X
 
  • Filter
  • Time
  • Show
Clear All
new posts

  • Variable Resonance Pickups - Fun with Buffers

    Everyone,

    I just got back from the Arlington guitar show where I demo'ed my Strat (1997 NC Affinity) with the variable resonance tone control I have been working on. The reaction was overwhelmingly positive with multiple players. Sounded fabulous through high-end tube amps.

    The goal of this demo was to demonstrate the concept using only off-the-shelf parts any guitar tech could install. The Tone control gives one-knob control of the pickup resonant frequency from about 1.4 KHz to over 5 KHz. This covers Gibson humbucker tonality through Tele through bright Strat. In positions 2 and 4 the resonance shifts up a bit because of the parallel pickups and covers about 2 KHz to over 6 Khz, which gives incredible "quack." The resonance can be switched out to provide a flat pickup response past 8 KHz.

    The 3 pickups were generic Chinese dual rail single coil-sized humbucking pickups which were about $8.50 each shipped. The key reason for using these was because in parallel coil mode these had low enough inductance and stray capacitance to make the technique work. The surprise was that these pickups sounded fabulous when in this circuit.

    The key electronic component was a Creation Audio Labs Redeemer buffer. This both buffers the pickup signal and provides a low impedance output to work the variable capacitor feature.

    Besides the buffer, the tone pot, and volume pot, there are two capacitors and two resistors needed for this to work. The schematic shows three caps but the Redeemer has an input cap so I deleted the 100 nF cap. I put a switch on the volume pot for lowpass/resonant mode with an extra resistor to allow hi-fi flat response to over 8 Khz, for recording/plugins.

    Here's the schematic.

    Redeemer Buffer with Resonance- White Rail.pdf

    If you want some ideas on how this works, please refer to my two posts. One is on variable resonance pickups Variable Resonance Pickups - Some Prior Art and Links,
    and one is on medium Z pickups with gain http://music-electronics-forum.com/t41094/. For this build I used conventional pickups and no gain.

    Not all of you read the Pickup Makers forum, so I thought this would be of general guitar tech interest.

    To make this technique work, you need to know the inductance, DC resistance, and stray capacitance to ground of the pickup windings. Using Circuitlab, you can simulate the response and pick the correct resistors and caps to get the resonance range you want. But you can generally make this work with any pickup with a reasonably low inductance, below 1.5 Henries. Many "vintage" or low-wind humbuckers will fit the bill when the coils are paralleled.

    Take care and questions/comments are welcome,

    Charlie

  • #2
    Very cool, thanks for the links. As it happens, I've recently been working on a DIY onboard bass preamp with a variable frequency/variable resonance format for the treble tone control section. Nothing new or earth shattering about that, of course, and no bass pickups I typically work with have inductance anywhere near under 1.5H. But I do also happen to have a CAL Redeemer kicking around in a rarely used active DI box, and a possible candidate guitar...hmm. Click image for larger version

Name:	Yel_wink.gif
Views:	2
Size:	950 Bytes
ID:	843867
    --Charlie Escher

    Comment


    • #3
      Passinwind,

      You can still use the technique with higher inductance pickups if they have very low wiring capacitance to ground. For example I can do this with Bill Lawrence L-90's which you can get at 2.0 and 2.8 Henries, because they have very low capacitance of about 90 pF. Using separate wires instead of shielded multi-conductor cable also helps; a lot of capacitance is in the cable.

      The real target is to have the self-resonance of the pickup very high -- 8 to 10 KHz or more. You can always bring the resonant frequency down with capacitance but you can't bring the resonant frequency above what the pickup itself limits you to.

      If you wind your own pickups or can have them wound to spec, you can easily get in the right range. The tradeoff is fewer winds = lower output. That's why one of my posts included a 12 DB FET preamp to boost the signal along with a FET buffer to provide the variable resonance control.

      In general, "hot" pickups won't work for this even in parallel coil mode. So an 8K DCR vintage humbucker might work where a 15K "hot" bridge pickup might not work for getting a Fendery sound since the resonance can't go high enough.

      If you (or anyone else) needs to get correct component values, until I can figure out if a public CircuitLab URL will work for anyone without paying for the service, I can give you the correct values if you give me the inductance (ideally measured with an Extech or DER EE LCR meter at 100 Hz) , DC resistance, and capacitance to ground of the pickup wiring. Or, maybe it's a pickup I already have or have access to, to measure. Most published pickup specs are either incomplete, are total BS, or done with unknown test conditions.


      -Charlie
      Last edited by charrich56; 10-16-2016, 03:18 AM.

      Comment


      • #4
        Thanks Charlie, appreciate those insights. It'll be a while before I get to even contemplating working on my guitar, and what I'm doing right now for bass is working quite well and is pretty simple to work out in LTspice. Here are some sims of my current config's response curves, which I have confirmed by measurement:

        Click image for larger version

Name:	PW3BLPF.JPG
Views:	1
Size:	227.5 KB
ID:	843868

        The whole board build including prototype PCBs was cheaper to build than buying a Redeemer, and does bass and mid EQ as well. My goals are a lot different than yours in this case though; I'm not trying to make my Travis Bean sound like an Alembic or Wal bass, even though I'm loosely copping their general sweepable resonant filter formats. But to make that work optimally, it's actually helpful to kill the pickups' native LRC resonant peaks, which makes me all the more interested in your work.
        --Charlie Escher

        Comment


        • #5
          That looks like a very versatile preamp/filter you have there. I've always admired bass players and builders for being out front in using active onboard electronics . Guitar players are a traditional finicky lot for the most part in going active, but things are coming around.

          You can kill the LRC peak with just the right value of load resistance as you probably know, but it doesn't help you with losing the highs above the peak very much. My "sneaky trick" is to put the native peak at about 8-10 Khz so when the peak is there and I'm not bringing it down with capacitance to use for voicing, it's just basically a moderate treble boost which can be dialed back to "flat", or a moderate cut. I don't see much use in having response past 8-10 KHz for guitar.
          Last edited by charrich56; 10-16-2016, 12:31 PM.

          Comment


          • #6
            Originally posted by charrich56 View Post
            That looks like a very versatile preamp/filter you have there. I've always admired bass players and builders for being out front in using active onboard electronics . Guitar players are a traditional finicky lot for the most part in going active, but things are coming around.

            You can kill the LRC peak with just the right value of load resistance as you probably know, but it doesn't help you with losing the highs above the peak very much. My "sneaky trick" is to put the native peak at about 8-10 Khz so when the peak is there and I'm not bringing it down with capacitance to use for voicing, it's just basically a moderate treble boost which can be dialed back to "flat", or a moderate cut. I don't see much use in having response past 8-10 KHz for guitar.
            Buffering is a very good and easy way to isolate the pickup from the additional capacitance loading of the guitar cable and the amp input load. A very easy way to hear the difference is to make the "Tillman buffer". Just web search the words in quotes to get the construction details. Then you just plug the buffer, located in the guitar end of the plug cable, and you can hear what effect the coax cable and amp loading have on your sound. It was very educational for me when I tried this about 10 years ago.

            Joseph J. Rogowski

            Comment


            • #7
              Joseph,

              Thanks and good to hear from you.

              The Tillman buffer will work for the resonant tone control, but you need to make sure that the source resistor is set fairly low, under 5K and maybe under about 2.5K. One of my circuits actually uses the 5K tone pot as the source resistor for the FET buffer. Generally you don't want to put DC through the wiper of a pot but if you are using the two fixed terminals it works OK, especially since you are just connecting a capacitor to the wiper.

              Here's a link for some buffer circuits. I don't agree with the article author's comments on buffers, but the links are OK:
              Electric Guitar Preamp Buffer*Circuits

              In certain situations with trying to set low frequency range resonance, too low pickup inductance, etc. you will need very low output impedance for the buffer, say less than 200 ohms or so, and will have to use a fancier buffer, or an opamp, to get the resonant frequency control to work right. Op amps make great buffers BTW as well (as long as they're unity gain stable), and you can use those.

              Unity gain is critical to getting the resonant control to work right in bootstrapping the capacitor. The closer you are to unity gain, the better the circuit will work and the more tweaking range you will have on the resonance of the pickup.
              Last edited by charrich56; 10-16-2016, 07:04 PM.

              Comment


              • #8
                I just made a URL for the Circuitlab simulation of the variable resonance control/buffer. Here it is:

                https://www.circuitlab.com/circuit/4...ffer-v2-opamp/

                This will allow you to look at and modify the circuit for a limited time, in demo mode without save or export capability. I don't know how long the time limit is for now.

                I am simulating the Redeemer with a generic opamp buffer.

                Comment


                • #9
                  A great idea

                  Sounds great! I hadn't heard of the CAL Redeemer buffer so I had to look it up...

                  Creation Audio Labs

                  Kinda pricey especially when you figure in the $16.87 shipping charge for UPS ground...



                  With all of the various buffering circuits in the public domain perhaps someone can come up with a single board solution incorporating the circuitry you came up with.

                  OOPS! I see that you already did exactly that... My bad!

                  https://www.circuitlab.com/circuit/4...ffer-v2-opamp/

                  Thanks!

                  Click image for larger version

Name:	Screenshot_2016-10-17-12-31-01_20161017125405376.jpg
Views:	1
Size:	258.9 KB
ID:	843889

                  Steve Ahola

                  P.S. I really like the idea of using a buffer not just for "long cable runs" but for doing something creative with the sound of a pickup. Can you post a link to the particular $8.50 pickup you used?
                  Last edited by Steve A.; 10-17-2016, 07:40 PM.
                  The Blue Guitar
                  www.blueguitar.org
                  Some recordings:
                  https://soundcloud.com/sssteeve/sets...e-blue-guitar/
                  .

                  Comment


                  • #10
                    Steve,

                    Here's the same one on Ebay, it looks like exactly the same thing. A bit cheaper than the ones I bought. I preferred the type with a single plastic cover rather then the "open rails" look, and I preferred the single copper tape strip grounding the rails on the bottom. There's lots of listings for these and I think they are all about the same. Amazon has some too.

                    New Dual Rail Humbucker Neck Pickup White 4 Wires for Electric Strat St Guitar | eBay

                    The other type has open rails and uses a grounding copper tape ring around the windings. These work OK IF you unwrap the covering tape, vertically cut the copper tape so it doesn't make a complete loop around the pickup, make sure the copper tape break is insulated from re-touching, and tape it back. They have slightly more capacitance but still work for the purpose.

                    New 1pcs Black Dual Rail Humbucker Strat Guiter Neck Pickup 4 Wire | eBay

                    The main spec is that the pickup is rated at about 9K - 10K resistance; a lot of them say that they are 4.98 K per coil.

                    I have a buffer circuit which I haven't built yet, which uses a hybrid of MOSFET and JFET in a mangled White cathode follower configuration. It looks great on headroom in CircuitLab but the MOSFET might be a bit noisy. I'll post it if there's some interest. Simple JFET buffers would work OK for this in theory and I will try one out very soon.

                    There are so few parts to this depending on the buffer, that it would be really simple to do a single sided PCB for it. That's probably the preferred direction, but the Redeemer is very nice albeit expensive as you pointed out. I'm moving on to humbucker territory to see how that will go.


                    -Charlie
                    Last edited by charrich56; 10-18-2016, 05:52 AM.

                    Comment


                    • #11
                      Steve,

                      Here's a Circuitlab link for the discrete buffer I designed. Again, hasn't been built yet but in simulation has incredible headroom, doesn't look like it distorts with +-3.5 Volt swings which is surprising for 9V. Current draw is good too, 550 - 565 microamps. I have high hopes for this on sound quality. Noise may be an issue though on the MOSFET.

                      https://www.circuitlab.com/circuit/k...r-v2-discrete/

                      The LSK170 is a high-end super low noise JFET. Could use a J201 probably with some tweaks. 1 uF coupling caps are polypropylene, or Elna Silmic II, which are dirt cheap from Mouser.

                      Part of the reason for posting all this stuff is to generate some interest and experimentation. If someone could come up with a better DIY buffer that is about $5.00 in parts, with a PCB which can be stuck to the back of a pot, has good battery life, and sounds great to the most discerning ears, that's great.

                      I'm going to be working with some local builders to whom a $50.00 buffer (cheaper for dealer/OEM) isn't going to be a huge expense on a $2.5K guitar. Also, might be able to save quite a bit of money on pickups.

                      But I'm still all in for the DIY and open source thing.
                      Last edited by charrich56; 10-18-2016, 04:59 AM.

                      Comment


                      • #12
                        Originally posted by charrich56 View Post

                        Part of the reason for posting all this stuff is to generate some interest and experimentation. If someone could come up with a better DIY buffer that is about $5.00 in parts, with a PCB which can be stuck to the back of a pot, has good battery life, and sounds great to the most discerning ears, that's great.
                        I have a few different opamps on hand that I've been planning to try out in that role, and I already have a pretty nice 2SK170 based buffer/preamp board and a Redeemer to use as benchmarks. If I manage to come up with anything good and cheap I'll be happy to share, but a lot of my candidate parts are well over $5 just for the chip. Many are SO-8 SMDs, so fitting something on a pot would be no problem at all.
                        --Charlie Escher

                        Comment


                        • #13
                          Passinwind,

                          Thanks! $5.00 was a stretch goal anyway. There are so many parts out there, that it would be a real service for someone to wade through them and suggest some good choices. I do like the old school methods of discrete FETs, etc. but modern opamps are very good. This is a somewhat stringent requirement because we want both low power consumption and low noise in the same part on a single 9V supply. It's no problem to get absolutely fantastic specs in an opamp which draws 5-7 mA per amp.

                          On second thought, maybe the whole thing could run on 5V, so we can build in one of those USB phone charger/battery modules that I see on Ebay a lot. I do worry about headroom but those modules will regulate the 5V until the battery is almost dead. Then just put a micro USB jack on the instrument control cavity cover or something, and plug in a USB charger to charge it up. Using an internal phone charger/battery would work fine for battery life because very small LiPo batteries can have a large maH capacity and we really want to get at least 200-500 hours battery life ideally.

                          Fishman has a 9V solution like this in both a Les Paul control cavity cover, and in a Strat trem cavity cover, but they want 100 bucks for them. Very nice and well engineered though.

                          I used a Microchip op amp some time back (MCP6021) for a low impedance preamp, which was a single DIP-8 and also available in SO-8 and it was for 5.5 V max supply. This one has good noise specs , but 1 to 1.35 mA draw per amp. This part also has an internal VDD/2 reference voltage pin which saves 2 resistors.

                          http://www.microchip.com/wwwproducts/en/MCP6021

                          It just might turn out though, that simple JFET buffers will work just fine for this application. I am going to try J201 and LSK170A.

                          I'm really interested in a good compact solution that would be simple for a small volume luthier/builder to install in their instrument.
                          Last edited by charrich56; 10-19-2016, 02:03 PM.

                          Comment


                          • #14
                            Passinwind,

                            I'd love to get a peek at your bass preamp design and/or the 2SK170 buffer/preamp sometime. I already showed you mine, will you show me yours? 8-) 8-)

                            Comment


                            • #15
                              Originally posted by charrich56 View Post
                              Passinwind,

                              I'd love to get a peek at your bass preamp design and/or the 2SK170 buffer/preamp sometime. I already showed you mine, will you show me yours? 8-) 8-)
                              I'll PM you. Some of the IP for the buffer is not mine, so I don't feel comfortable posting the schematic for that. Once I get the bass preamp fully dialed in I may do a board share for it on OSHpark, and will probably also make the design open source, including the LTspice model. It's really just cookbook stuff though, simple Sallen-Key low pass filter with the only trick being getting the alignment tweaked for good intuitive control feel. The board is 2.2" X 1.66" in through-hole config, I could definitely cut it down a good bit further with even just using SMD opamps and resistors and the same polypro and polyester film caps. Here's board #2:

                              Click image for larger version

Name:	PW3B_LPF_2.jpg
Views:	1
Size:	102.4 KB
ID:	843924

                              The 5 hole control format for the bass it's going in will be V-V-B-M-Treb Resonance-Treb Freq, with Bass and Mids on a stacked pot. Quiescent current draw is < 800uA. Here's my stompbox test fixture:

                              Click image for larger version

Name:	LPFStomp.JPG
Views:	1
Size:	201.9 KB
ID:	843925
                              Last edited by Passinwind; 10-20-2016, 08:01 PM. Reason: Added picture
                              --Charlie Escher

                              Comment

                              Working...
                              X