More Low Impedance Pickup Research

[bbsailor]

Here is my latest low impedance pickup design. It is very simple to construct. I have drawn a schematic as well as placed the finished pickup so you can see how simple it is to make.

Parts List:

2 CSE187L low frequency current sensing transformers
2 lengths of AWG 10 electrical wire 2.5" long with .25" insulation striped off each end.
1 magnet .25" wide X 2" long X .125" thick, polarized through the thickness
4 pieces of .125" OD X 3/32 ID copper tubing cut .5" long
1 matching transformer, 500 ohms to 50K
1 10ft length of 2 conductor shielded microphone cable
1. 3-pin mic connector to match the 500 ohm to 50K ohm transformer
3 short lengths of stranded hook up wire
2 pieces of electrical tape to secure the magnet to the sensor coil

The .125" OD X 3/32 ID copper tubing makes a good mechanism to connect the wire to the transformer. See the photo. Use a tubing cutter to cut the tubing so you don't deform it too much. Use a 3/32 drill bit to keep the tubing fully open after cutting as it compresses slightly. You will need to use a flat file on the AWG 10 with .25" striped ends to allow the copper tube to be forced on to these ends. The tubing will more easily fit on the large AWG 12transformer primary leads. Connect as shown. Check your local hobby shops for the copper tubing on the K&S Engineering display racks. They have other metal products that might be useful as well.

I choose to leave the wire insulation on the wire to prevent the metal magnet from shorting out the primary turn. If you use ceramic magnets you may not need insulation on the wires so the wire can be a little closer to the magnets.
You may want to lay a strip of .25" conductive adhesive copper foil on the metal magnet and ground this copper foil to the ground connection of the coax to further reduce noise.

The transformer primaries are in series and form the low impedance sensor loop. The transformer secondaries are in series and form a 375 ohms output impedance according to my measurements.

I connected a single CSE187L transformer to the 500 ohm to 50K ohm transformer and the output was noticibly less than when connected with two transformers in series. This 2 transformer design has a similar output as a commercial single coil pickup.

I mounted this pickup in my acoustic guitar using duct tape and it sounds quite good although "quite good" is not a scientific term. By grounding each of the transformer frames to the shielded cable ground (see photo), the output is balanced and is very quiet even near my computer.

Anyone looking to experiment at a low cost should consider making one of these to quickly explore low impedance pickups. The pictures show everything; nothing is hidden.

If you bend the ends of each wire 90 degrees you can mount the transformers under the magnet and fit the whole assembly in a commercial single coil pickup housing/cover. You can even use 4 transformers and mount two sensor coils in a commecial humbucking housing/cover.

I hope this stimulates some others to try it.

Jospeh Rogowski

[David King]

Joseph,
perhaps if you put together an experimenter's kit at a reasonable price we would take up this challenge? Right now I don't have an extra $10 to my name so don't do it on my account.

[bbsailor]

Joseph,
perhaps if you put together an experimenter's kit at a reasonable price we would take up this challenge? Right now I don't have an extra $10 to my name so don't do it on my account.

The CSE187L low frequency current sense transformer is available from Mouser Electronics for less than $3.00 each. The 500 ohm to 40K to 50K ohm output transformer is available from Radio Shack at about $15.00. The other stuff you can scrounge up locally. The magnet is available from K&J Magnetics for around $2.50 each. You can use six short rod magnets being held in a place by a piece of .25" thick plastic. Keep the single loop coil close to the same height as the top of the magnets for maximum sensitivity.

The total cost of trying this is about the same as buying a small roll of AWG 42 magnet wire and less if you already have some parts in your junk box.

I am not interested in supplying parts or kits.

Joseph Rogowski

[bbsailor]

For those interested in technical data, below is the resonance measurement taken with a 10 to 1 scope probe to minimize capacitance loading and to obtain better accuracy.

This measurement was done at the high output of the 50K ohms side that normally plugs into the amplifier input. This transformer is attached to the end of a 10 ft, 2-conductor, balanced microphone cable that is attached to the two series connected CSE187L low frequency current sense transformers that are connected to the single turn string sensing loop (per the photo). The total turns ratio of all the transformers is 10,000 to 1.

Resonant peak: 12.83KHz
Half voltage points: 6.94KHz and 22.2KHz

Joseph Rogowski

[DrStrangelove]

Note: these are very low impedance pickups since the entire sensor
loop with transformer primaries only reaches .001 ohms.
I can't measure it -- I can only calculate it.

Over the weekend, I made a couple low impedance pickups for the cello
since shaping/bending the sensor over the fretboard arch is facile.
The irregular arch of the cello's fingerboard is what makes this design
an easy choice.

These pickups are very high fidelity and utterly immune to hum.
Making a humbucker version for low noise is redundant -- do it
for the higher output.

Several details: You need strong magnets.
Cheap Ceramic 1 is barely adequate, C8 is tolerable.
No news on Alnico8 yet. A thin stick of neodymium
magnet works fine.

Bigger wire for the sensor loop is better.
1/8" copper refrigeration tubing is easier to work with than AWG10.


more later.
-drh

[David Schwab]

1/8" copper refrigeration tubing is easier to work with than AWG10.

I was thinking the same thing today.... just popped in my head.

[salvarsan]

The sensor loop resistance is too low to measure.

Here is a calculator for arbitrary lengths of rectangular conductors
and AWG sized wire. You can also select among a several conductors.

http://www.salvarsan.org/magwire/ohmcalc.html

It is all client-side javascript and may be used,saved, and
distributed with wild abandon.

-drh

[Béla]

Could this pickup design benefit from an active electronic circuit? Could an opamp current to voltage converter be placed instead of the transformer? I think this would make the pickup have a more even response - but will it be more noisy this way?

Sorry for acting like a child and asking seemingly stupid questions, but this is the most amazing pickup design I have yet seen, and I'm very excited about it.

Consider this setup: A pair of these pickups, outputting the full frequency range of the string movement. Still inside the guitar, let's add a second order low-pass filter, that has an adjustable Q factor and cutoff point. A state-variable filter can do that, with the help of two potmeters, in the place of the regular tone controls.
One could have any sound that one feels appropriate: classic pickup responses, full range, almost acoustic sounds, and even high Q resonances for a wah-like sound.

For my own application I'll also add a balanced line driver + phantom powered operation, running into a computer sound interface that has phantom power (E-MU 0404 USB). A notebook computer on stage for guitar effect processing - that's something you don't see every day.

[bbsailor]

Could this pickup design benefit from an active electronic circuit?

Yes, but after the first transformer conversion from the single string loop to the output of either the CSE187L or the CST1005 current transformers.


Could an opamp current to voltage converter be placed instead of the transformer?
No. There are no circuits that I am aware can effectivly operate at this extreemly low string loop impedance. But, after the current transformer, ICs or transistors can work as you suggest.

I think this would make the pickup have a more even response - but will it be more noisy this way?

Using a transformer to boost the string loop to about 150 to 300 ohms puts this pickup in the range of commercial microphone outputs and very processable and low noise.

Sorry for acting like a child and asking seemingly stupid questions, but this is the most amazing pickup design I have yet seen, and I'm very excited about it.

Look at the Lace Alumitone pickups as they use the same design. I have just identified an easy way to use commonly available current transformers to easily experiment with current-based pickup designs. The current sense loop is very sensitive to the size of the string loop and the wire thickness. Thicker is better. The design goal is to use an inductance meter like the Extech and measure the output impedance of the CSE187L or the CST1005 with the string loop installed and obtain a transformer output impedance that is a close match for the Shure 97U series matching transformer input of either 150 to 300 ohms on one setting or the 50 to 75 ohms setting with the jumper changed inside the Shure 97U case.

Consider this setup: A pair of these pickups, outputting the full frequency range of the string movement. Still inside the guitar, let's add a second order low-pass filter, that has an adjustable Q factor and cutoff point. A state-variable filter can do that, with the help of two potmeters, in the place of the regular tone controls.
One could have any sound that one feels appropriate: classic pickup responses, full range, almost acoustic sounds, and even high Q resonances for a wah-like sound.

Yes, if this circuit has a gain of between 10 and 20 the output directly from the current sense tranformer could match passive guitar pickups and bypass the need for the second passive Shure A97U transformer at the amplifier. Stacked concentric pots offer what you want to do with two controls in one hole.

For my own application I'll also add a balanced line driver + phantom powered operation, running into a computer sound interface that has phantom power (E-MU 0404 USB).
You could to this but EMG makes a bass varaible EQ control with High and Low frequency shelving as well as a variable mid 300 to 3Khz with boost and cut that operates on either a single 9V or dual (18V) battery system. Your guitar needs to be about 1.25" deep for the circuit cards to fit in the control cavity. You could externally power the EMG active circuit using the second conductor as a phantom power connection.

A notebook computer on stage for guitar effect processing - that's something you don't see every day.

As micro processor speeds increases, real-time audio processing offers more possibilities.


I hope I have answered your questions?

Joseph Rogowski