Low Impedance Pickup Research

[trzesnk]

Thank you, bbsailor, for ALL the hints! The content in this thread is an encyclopedia for low impedance pickup makers!
Krzysztof.

[trzesnk]

I'm not familiar with the Shure transformer, but maybe this helps:
1kHz is a widely used reference frequency in audio.
For comparison the 1kHz impedance of a typical PAF-type Hi-Z humbucker having a DCR of 8k and an inductance of 4.5H is around 30 kOhm.

More importantly your PU essentially is an inductive source which forms an LR low pass filter together with the load resistance.
E.g., if you connect your PU to a 700 Ohm load, the frequency reponse will be down by roughly 3dB at 20kHz

Thank you, Helmholtz, for a practical hint!
Krzysztof.

[bbsailor]

I'm not familiar with the Shure transformer, but maybe this helps:
1kHz is a widely used reference frequency in audio.
For comparison the 1kHz impedance of a typical PAF-type Hi-Z humbucker having a DCR of 8k and an inductance of 4.5H is around 30 kOhm.

More importantly your PU essentially is an inductive source which forms an LR low pass filter together with the load resistance.
E.g., if you connect your PU to a 700 Ohm load, the frequency reponse will be down by roughly 3dB at 20kHz

Hemholtz is correct about the audio impedance rating of Shure transformers. However,
I want to add how the guitar pickup sound is governed by the following.
1. Guitar pickup impedance and effect of pickup loading by:
1a. Passive volume and tone controls
1b. External loading by cable capacitance, amplifier input impedance loading or line matching input transformer loading.
1c. Pickup noise is affect by the pickup design, high impedance single coil, humbucker, coil covered by a metal cover or the use of a low impedance design.
2. Guitar amplifier speaker size and frequency response.
2a. A 12 inch diameter speaker has a peak response at 3K Hz and is down by 10 dB at 6 KHz.
2b. A 8 inch diameter speaker has a peak response at 3.5 KHz and is down by 10 dB at 8 KHz.

Making low impedance pickups requires having about 10 to 12 times less turns and wire about 10 times thicker than normal pickup wire. AWG 42 will now be about AWG 32. The large impact of a low impedance pickup output is the pickups minimal reaction to the cable capacitance and noise pickup through the cable. High impedance pickups react to the typical 25 pf to 30 pf of capacitance per foot of the guitar cable. Any high impedance pickup measurements you do should be done at the end of the coax cable length you typically use to get realistic results.

The highest guitar primary frequency, meaning not a harmonic, is 1,174 Hz at fret 22. Typical guitar string harmonics occur at 2, 3, 4 and 5 times the primary string frequency. This puts the upper harmonic range for the highest primary frequency at 5,870 Hz.

Doing measurements of guitar pickup frequency response up to about 6 KHz should cover all the possible guitar harmonics. Extending the measurements to 10 KHz would be practical to better see the very high frequency downward slope.

The final test is your ear. Where you stand in front of your guitar amplifier can affect how you hear higher frequencies with some amplifier speakers beaming very narrow bands of high frequencies.

I hope this helps those MEF members who are exploring and measuring low impedance pickups.

Joseph J. Rogowski

[Helmholtz]

So I am not asking here how to change my design, but how to understand the concept of output impedance given as a single value. I tried to find the answer myself, as described in previous posts, but I still do not understand which frequency should be used to measure output impedance. See the measurement in my post #502. What is an output impedance of my pickup? At which frequency should I measure it?

I'm not familiar with the Shure transformer, but maybe this helps:
1kHz is a widely used reference frequency in audio.
For comparison the 1kHz impedance of a typical PAF-type Hi-Z humbucker having a DCR of 8k and an inductance of 4.5H is around 30 kOhm.

More importantly your PU essentially is an inductive source which forms an LR low pass filter together with the load resistance.
E.g., if you connect your PU to a 700 Ohm load, the frequency reponse will be down by roughly 3dB at 20kHz

[trzesnk]

Dear Joseph!

Thank you for the input. Obviously this thread, started by you, is devoted mainly to pickups with single or couple thick copper loops and usage of curent sensing transformers. If I have wrongly posted here - sorry!

In spite of the fact that I have finally decided not to go this way, this thread was the place where a lot of my design decisions evolved. The main reason I decided to make more traditional design (if a 60 Ohm humbucker can be called traditional) was esthetics. I just couldn’t manage to make a design esthetically appealing and accommodating the current sensing transformer, or better two transformers into a very small place designated for the pickup. Yes, there is a place, but I do not want to clutter this space. So I designed a more traditional pickup generating in principle similar electric signal as a dynamic microphone. As the pickup meets my design principles (to support acoustic sound of the guitar) and is very well accepted by the players I do not want to go much away from the current design.

So I am not asking here how to change my design, but how to understand the concept of output impedance given as a single value. I tried to find the answer myself, as described in previous posts, but I still do not understand which frequency should be used to measure output impedance. See the measurement in my post #502. What is an output impedance of my pickup? At which frequency should I measure it?

BTW: You write in one of the posts in this thread, that 1 MOhm load (typical input impedance of guitar amps) on secondary windings of Shure A95 will reflect on the primary windings roughly 3.8 kOhm at 1kHz and 2.9 at 100Hz, so 49x more than 78 Ohm of my pickup impedance at 1kHz and 49x more than 59 Ohm of my pickup impedance at 100Hz. So the minimum 10x rule is met. But what nominal output impedance is, remains an open question to me.

Thanks!

Krzysztof.

[bbsailor]

Krzysztof,

When you reduce the AWG wire size by 3 sizes, you reduce the resistance by one half assuming you are using the same length of wire.

Since your guitars are using a new body and neck design that raises the neck above the body, you can easily use a Triad CSE186L current transformer with the three primary turns removed leaving about a 0.125 inch square opening to use AWG 8 as the string loop.

Measure in input impedance of your XLR input transformer loaded by the input impedance of the amplifier that it is plugged into. You should try to target the output impedance of your AWG 32 twin coil pickup to be about one tenth less than the transformer input impedance. This is called a bridging impedance and will maintain a higher voltage level than if it were the same impedance and then have half the output level.

The most vital measuring instrument is your ears. Look up the Fletcher Munson Curve to see where the human ear is most sensitive. Once you find what sounds good, then you can take measurements to begin to optimize your pickup design to move the coil, magnet, and current transformer variables in the right direction.

Here is a current transformer CSE186L example.

Calculate the resistance of 8 inches of AWG 8 solid copper wire. Look up the frequency of the skin effect for this wire. Clean the ends of the wire before soldering using a matching copper tube or drilled copper bar to make a very good low resistance joint. The CSE186L has 500 turns. Square this number: 250,000 and multiply by the string loop resistance to get the approximate output impedance of the current transformer. You can try various solid wire sizes or stranded wire sizes and listen for tonal changes.

Order a few Triad CSE186L to build a few current transformer designs. Ground the string loop and the metal frame of the transformer to reduce noise. Use two conductor shielded cable to the input transformer.

This should get you moving in the right direction.

Joseph J. Rogowski

[trzesnk]

Sound Samples of my pickup to augment my previous post.
Here are some recordings of my archtops with the same pickup by Kinloch Nelson, Tony
McManus, Tim Lerch, and Mark Dziuba:
https://youtu.be/yhIJemVg9N0?si=sgX5bphS5o_kQuNO
https://youtu.be/msH48bBv_ec?si=wFDkS452XBKTDbuA
https://youtu.be/SEEsRQW2zSg?si=QGbZ2ehzVIam8JuQ
https://youtu.be/zj1X4yMzw-8?si=eRvXAT7s2n65NZju

If anybody knowledgeable can help me with the question in the previous post - it would help me in the development of the next pickup version. I can not go too much up with the inductance as low frequencies starts to distorts (saturation of the iron core?). I am now experimenting with the same/similar inductance but lower DCR by using AWG 32 instead of 35 wire and slightly enlarged coil size to accommodate increased wire dimension. I assume that keeping the inductance the same and reducing the impedance will not negatively influence voltage on secondary winding of Shure A95 transformer.

Krzysztof.

[trzesnk]

Hello,

This is my first post, but I have read this thread for some time and I am very grateful for sharing your research. Thank you all for your invaluable input.
Thanks to this thread, among others sources, I designed a low impedance pickup that I use in my archtop guitars. Sounds beautiful. It is a humbucker with a little over 400 turns per coil of 38AWG wire. I connect the coils in parallel and the DCR is 60 Ohms. Since the pickup is balanced by nature, the signal goes to 20 feet XLR cable and only at the amplifiers it connects either to XLR mic input or to an instrument input through an impedance matching transformer (Neutrik NTE 10/3, Shure A85, or Shure A95, and the latter is most often chosen by the musicians). I measured the impedance of the humbucker using an acoustic signal generator, a voltmeter and a resistor R connected in series with the pickup P in this way: SignalGen-R-P-Gnd. For different frequencies I have measured voltage across resistor R and pickup P. The Impedance was calculated for each frequency this way: Z=(Vz/Vr)*R. Here are the results:

Hz			Ohm
20			59
40			59
80			59
100			59
200			60
400			60
800			72
1000			78
2000			115
4000			194
8000			339
10000			403
12500			478
15000			549
17500			623
20000			716

And now a lame question about the definition of output impedance and a request for clarification:
One output impedance value is given for microphones and in this thread for pickups. Sometimes the description talks about the average output impedance in the frequency response band, sometimes at 1kHz.
How should output impedance as a one value be interpreted in such a wide range of variability with frequency? What is the output impedance of my humbucker?

Do I understand correctly that if for e.g. 1kHz the output impedance of the pickup is 10x lower than the input impedance of the amplifier, then everything should be OK, because the reflected impedance of the impedance matching transformer with e.g. 1MOhm load on secondary will present an input impedance which will rise with frequency? So the 1:10 ratio will also be maintained for higher frequencies?

Thank you again!

Krzysztof.

[ncblue]

Last night I brought the new guitar to band practice for some real world experience. Had to use a clean boost or Eq pedal to get the output up to useable levels with the amp I was using. Once boosted the guitar performed well clean as well as with low or moderate amounts of overdrive. Even with higher gain it was pretty quiet, about on par as a noisy humbucker, but well below that of a typical single.

[ncblue]

Interesting update: In the new guitar I had the 2 pickups in middle switch position wired in parallel which was ok, but tried wiring the transformers in series and I really like it. Gives a nice boost to the mids.

[ncblue]

Thanks for the feedback Joseph. I haven't quite finished the guitar and I hadn't done a good setup. The nut was too low and I've been trying various ideas for the bridge. I think the output is acceptable passive going to an amp, but playing with my band will tell the tale. I like the jangly sounds I'm getting. I have put a Tillman preamp in another guitar, not much space to do it on this one.

[bbsailor]

Nick

Nice job in trying something new!

To quell the microphonic noise, put some packing foam in the current transformer hole where the wire goes through. This will prevent the wire vibrating near a loud amp or vibrations from any noise created with an contact or impact on the guitar body.

Using thicker and a shorter length of wire around the magnets will enhance the lower guitar frequencies. What test equipment do you have to do electrical measurements?

What is your current goal for improving the sound of you low impedance pickups?

I listened to your audio recording and heard some string fret buzzing. Either adjust the neck, raise the string height or go up one gauge in string thickness to make a cleaner sound.

Whatever gauge wire you choose to go around the magnets and through the current transformer, check the AWG wire table to see the frequency of the skin effect where the output of higher frequencies starts tapering off through less of high frequency energy being converted in the current transformer.

When designing pickups let your ears be the guide then measure what sounds good and take good notes about your measurements. In my previous posts I covered what variables affect the low impedance pickup sounds.

The mechanics of a low impedance current driven pickup have some practical physical design issues to consider.
1. Keep the pickup thickness small so it can slide under the guitar strings with no hole needing to be cut. Look for magnets that are 2.25 inches long, 0.25 inch to 0.375 inch wide and no thicker than about 0.25 inch tall to easily fit under the strings.
2. Locating the current transformer close to the pickup and magnet to keep the wire as short as practical. Bending the heavy wire loop down 90 degrees locates the current transformer near the pickup and only requires about a one cubic inch hole to fit the current transformer in. The wood in line with the neck and bridge stays uncut and provides a better mass for the strings to vibrate.
3. Estimate the output AC resistance of the current by following information that I have posted before. Try to make the volume pot value about 10 times higher than the current transformer output AC resistance when the copper or other metal string loop is attached.
4. Sample current transformers that have different published frequency ranges to determine what your ears say sounds better.
5. You can decide if you want active electronics mounted in the guitar but you will need to change the battery when it goes dead. You can make a buffer-amplifier mounted in the plug end of the guitar jack by going to the web site till.com where Tillman shows how to construct this device where the batteries or power source is located outside the guitar body.

Keep up your good and creative work. Please keep us forum members updated about your progress or design issues needing some insight.

Joseph J. Rogowski

[ncblue]

Hi, I have been busy building my new guitar to utilize the low Z pickups.


In the neck is the Aluminum single loop with ceramic strat magnet and with the 1:1000 transformer. I added a steel plate under magnet which shifted the resonant peak and improved output. I used hot glue on the magnet and plate and it's keeping the microphonics to a minimum. I had tried some neodymium magnets which were far too strong. The current transformer is installed from the back of the guitar with short 18ga wires

The bridge pickup is a 3 loop single coil of 18ga wire wrapped around a ceramic magnet and with 2 primary loops thru the transformer to give a 1:500 ratio. To make the pickup I glued two plastic flanges to the magnet, wrapped the wires around the magnet and then fed the wires thru an access hole to the transformer. The pickup is thin enough not to require routing on the top of the body. It is a little microphonic, so potting or glue might be needed.

I wanted a very light guitar, so a headless construction made sense for balance. The neck is a beheaded strat copy. The body is from a repurposed pine bed frame with a center strip of oak to give it a little strength. The thickness is 1.35 inch which makes it very light. I was concerned about the strength of the pine and rightly so as during shaping of the body contours it developed a large crack that had to be glued. I then added some areas of hardwood in the neck pocket, and electronics cavity. Routing in the center of the body was kept to a minimum. I also added a steel plate under the tuners. It seems to hold the 9 gauge string tension well. With careful design I was able to mount the standard tuners with straight string pull past the bridge and a sufficient break angle. Tuning access is a bit cramped, but seem to hold pitch well. Besides the cost savings of this headless tuning setup, it has the advantage that no clamping is needed at the nut end.

The electronics are simple, a 250ka volume pot and two mini switches. One switch for pickup selection and a tone switch adds capacitance to ground. Currently using a 4.7nf to cut off sometimes piercing highs and a 15nf to get in humbucker range.

The guitar and pickups produce clean and bright sounds and are quiet unless using high gain. Since the output is a little lower than even a Strat, higher gain setting are needed to get overdriven tones. More real world experience is needed to see if these pickups are a success.
I uploaded a sound clip played into my Fender HRD, recorded into my phone's mic. Bridge pickup then alternating with the neck.
https://www.dropbox.com/scl/fi/gb7bp...mq40er217&dl=0

[bbsailor]

Nick,

Here are a few things to consider when selecting the right size, shape and coating for your magnet selection on your current transformer based pickup. Consider using a neo type magnet due to its strength, especially important when using a thin height magnet.

1. Size. The magnet height should be based on how high the strings are from the guitar top when presses at the last fret on the neck. This is usually based on the fingerboard thickness plus the fret height. I have found that getting a magnet that is 0.25 inches thick makes testing on existing guitars very easy. Since string spacing width is about 2 inches wide, select a magnet that is 2.25 inches long to ensure that the two E strings have full magnetic coverage for string vibrations. A magnet width of 0.375 inches is a good starting point but a 0.25 inch wide will work also.

2. Shape. A rectangular shape with the magnet magnetized through the thickness is a good start for your design. However you can get what is called a stepped in design magnet that contains a semi circular groove along the length of the pickup magnet on each side about 0.125 inch wide will make mounting the string loop wire to the magnet easier using a heat shrink tubing. If you want to fine tune the individual string output due to the height difference, a flat (not stepped) magnet side wall will make shaping the string loop to follow the arch of the fingerboard curve easier.

3. Magnets come with many selections of coatings. I would select an epoxy coating to avoid shorting out a bare copper wire AWG 8 string loop going through the current transformer. If you choose to use a metal coated magnet, you will be generating eddy currents in the metal coating and affect the pickup tone. If you choose to try it, just place a piece of tape over the side of the magnet to prevent shorting out the string loop to the conductive magnet surface.

Many magnet supply vendors will allow you to select a magnet type, size and coating in addition to the common sizes they offer on their web sites.

Another key design consideration is the current transformer turns ratio. Try using CTs with 1000, 1500, 2500 and 5000 turns to get a good high impedance output. Balancing a strong guitar sound signal with noise and tone is a balancing act that your ears can select. However, when trying to fine tune some alternatives, you will need access to some test equipment.

Before selecting your final magnet size, type and coating, plan on how you will mount the pickup to the guitar. Typically it should fit under the strings with no cutout needed under the strings. The only cut out should be about a cubic inch for the current transformer. Plan on stuffing some foam in the current transformer opening to prevent any movement or vibrations between the string loop wire and the current transformer. To minimize noise, make sure that you ground the string loop.

Let us know what type, size, coating and shape magnets you are using. There are many magnet vendors on the web. Experiment with different magnets and wire sizes and let your ears point you in the right final design direction.

Joseph J. Rogowski

[ncblue]

Ok guys I'm starting to get a better handle on these CT pickups. It's doesn't take many, more than two, winds to blow up the AC resistance and drag down the resonant freq. Case in point I took a ceramic strat PU and wrapped 7 winds of 16 awg stranded around the bobbin. The result was very powerful, very mid rangy. If it hadn't be extremely noisy it may have been useful. I took this as a sign to work on humbuckers. Using the sketch in Joseph's first post I made a version with 12awg and two center legs. This arrangement appears to me to be a parallel humbucker which would lower the resistance by 1/4, I calculate it to be around 528ohm. The output is a bit low with a pretty flat eq. Very quiet.


Next I reconfigured to have a series humbucker. As expected the power went up as the tone got some what less bright. Resistance about 2100ohm. Also very quiet. It's a little hard to see, but the electric tape is there since the wires cross to give a reverse wind.



I wanted to try the 2 winds through the CT idea. I made a parallel humbucker with 18 awg stranded, two loops of 2 winds. Calculated resistance 1200ohm. Good power and good highs. I changed it to one wind through the CT and the power was stronger and the highs slightly less, so this gives us another option to shape the tone.




Contrary to my early experiments, which used a single magnet, these humbucker arrangements do work very well. The balance of power to tone₁ isn't quite there yet. Looking at the Fletcher Munson curves has got me wondering where I want to get the resonant freq for my taste.
Next up I plan to build a parallel humbucker with heavy gauge wire or Aluminum. I've been designing the guitar to mount these pickups and should start cutting wood soon. You guys know of any good inexpensive magnets on Ebay or even at craft stores?


1. "Power to Tone Ratio", a registered trademark of NC Inc.