Low and Medium Z Pickups and Onboard Electronics

[charrich56]

This thread is for discussion of low and medium impedance pickups and onboard electronics in the instrument which uses them.

There have been some great posts in the past on the low-Z approach in this forum, and several commercial instruments, guitars and basses, have used low-Z pickups. One advocate and pioneer in this area was Les Paul himself.

Let me start off with a quote from Bill Lawrence, may he rest in peace. What a great gentleman.

To achieve the widest variety of sounds, all parts of the chain ( guitar, strings, controls, cable, amp and speaker) must be considered as a total, capable of reproducing a wide range of frequencies. There are two ways to achieve this: We either keep both the inductance of the pickups and the cable capacitance as low as possible or we introduce 2-way pickup systems where the high impedance part reproduces the lows and the low impedance part reproduces the highs, just like a 2-way speaker system.

We are paying today for the biggest blunder that the guitar industry made.

In the mid-30s, in Berlin, at the Telefunken Research Lab, Neuman and his team of engineers designed the condenser mike and the first tape recorder with magnetic pickups. To achieve a true reproduction of sound, they had to use low impedance. In 1947, the low impedance tape recorder was introduced in the US, and a new company called Ampex started to produce them. A few years later, the major recording studios were using tape recorders; by the early 50s, the whole world, with the exception of the guitar industry, had switched to low impedance, and it was called High Fidelity -- or just Hi Fi. The guitar industry did not recognize the advantage of low impedance and did not make the switch. Today, a change to low impedance is virtually impossible. There are too many amps and guitars floating around, and we are stuck with all the disadvantages of high impedance.

Let’s see how we can make the best out of it.

http://www.billlawrence.com/Pages/Al...hImpedance.htm

This thread is about not being satisfied with "making the best out of it" and moving forward.

[bbsailor]

Charrich56,

How to make the best of it and move forward. The problem seems to be that high impedance pickups with 6,000 to 10,00 turns per coil and a coaxial cable with a capacitance of about 350pf for a 10 ft length, makes a good resonant circuit with a steep drop-off after that resonance. The guitar is not a high fidelity instrument in that its bandwidth is limited by the string frequencies and harmonics and the frequency response of the guitar amp itself. That is not to say that treating the guitar like a low impedance microphone with an actual impedance of between 150 to 300 ohms could not help transmit the signal using a balanced line approach with minimal high frequency losses or noise pickup.

Typically you want the input impedance of any low impedance microphone to be a bridging impedance of about 10 times the source impedance. Most Shure A95U series mic matching transformers have about 3K ohms input impedance measured at 1Khz with the Extech LCR meter. What this means is that any pickup footprint type (SC, Humbucker, etc) with one tenth the number of turns in the 600 to 1000 turn range can be made to match the 3K ohms input impedance of the mic matching transformer or the actual impedance of a mic mixer. Using one tenth the number of turns makes the impedance ratio the square of the turns difference, or in this case, one hundreth. A 250K high impedance pickup pot becomes a 2.5K ohm low impedance pickup pot while a 500K ohms pot becomes a 5K ohm for a low impedance version. Scaling the capacitor values by 100 is also a good place to start selecting tonal component values.

To use a standard guitar amp the only required new component is a mic matching transformer at the guitar amp end of the connecting cable coming out of the guitar between 150 to 300 ohms. The Les Paul guitar put a decade switch in to make tonal shifts in the low impedance pickup circuit by changing the resonant points to sound like other popular pickup sounds of the day.

Most early guitar amps were a repackaging of PA type amplifiers where the speaker and amp were combined in a single enclosure. These amps uses a high impedane mic input circuit that were repurposed for a standard .25" plug type high impedane interface. For the guitar to put out enough signal to drive this input required near a few hundred millivolts and thus 6,000 to 10,000 turns of very fine wire.

If you target your low impedance pickup to be in the 25mH range (2.5H divided by 100), you will start to appreciate how some physical changes in coil construction techniques in the low impdance range can offer new design opportunities like in the new Fishman design. If you expand your thinking to include making active pickups with active midrange shift/boot/cut, you can tune the pickup to sound like traditional passive high impedance designs.

There is one more point that needs to be stated. The sound of a guitar string reflects the amount of energy that is left in the string while it is vibrating. Some energy is absorbed by the air around the moving strings, but much more energy is absorbed by the body/neck type of the guitar being used to host the pickups. Low impedance pickups with no classic resonant hump in the 3Khz to 5Khz range can be set up to hear the energy band that was highly damped by high impedane pickups. This will be an area where blending in the upper harmonics can control more subtle and subjective aspects of the guitar sound. Now we can have a widely accepted way to to this. However, someone will always want more cowbell!!!

Joseph Rogowski

[Mike Sulzer]

bbsailor: "If you expand your thinking to include making active pickups with active midrange shift/boot/cut, you can tune the pickup to sound like traditional passive high impedance designs."

You can do a lot without the active electronics in the guitar. It just requires switching different values of C across the pickup, although you do not have complete control over the Q. Medium impedance is better for running through a cable than high, so the preamp can be external to the guitar. (I prefer a preamp rather than a transformer.)

[bbsailor]

bbsailor: "If you expand your thinking to include making active pickups with active midrange shift/boot/cut, you can tune the pickup to sound like traditional passive high impedance designs."

You can do a lot without the active electronics in the guitar. It just requires switching different values of C across the pickup, although you do not have complete control over the Q. Medium impedance is better for running through a cable than high, so the preamp can be external to the guitar. (I prefer a preamp rather than a transformer.)

Mike,

What is your definition of medium impedance?

Thanks

Joseph Rogowski

[Mike Sulzer]

Joseph,

Order of a thousand turns, more or less.

Mike,

What is your definition of medium impedance?

Thanks

Joseph Rogowski

[charrich56]

Joseph,

Thanks for a simply great opening post. You have covered a lot of the design territory here that I fully expected to be discussed over many posts.

I want to also thank you for some of your earlier posts on the same subjects. Your posts among a few others restarted the design journey into this space for me.

Obviously, to be compatible with anything but a mic-level input, we are going to want to bring that low-Z signal up to "normal" guitar level, add voicing to at least be compatible tone-wise with a "normal" guitar when we want to, and be able to plug into into a normal guitar/bass amp. So the other part of this thread is on-board electronics.

There are several approaches to use for doing this, but for this post I am just going to talk about what I am playing with - using an onboard miniature audio transformer with a preamp/ buffer, with low-Z pickups. It would be great to see other approaches discussed (pure preamps of various flavors, etc.) but what follows is that approach.

Just to flesh out what Joseph is starting on, I have wound Strat single-coils, P-90's, and humbucker coils with the following, and done measurements.

The baseline if you want to start playing with this stuff with "regular" pickup magnetics and bobbins is: 600 to 700 turns of #32 AWG. A typical pickup wound with these (I am going to use the P-90s I did for an example): will measure in the range:

Inductance (1 KHz): 25 to 35 mH
DC Resistance: 30-60 ohms
Winding capacitance: 100-180 pF

These are typical of assembled pickup measurements, not just bare coil measurements.

I wound the P-90s sort of "hot" for signal level with the transformer I am using, and ended up with 34 mH inductance, 54 ohms DCR, and about 170 pF capacitance.

Generally we end up with an AC impedance at 1 KHz of 200 ohms or less. I think that low-Z pickups in this 150-200 ohm max range are the most versatile starting point for the following reasons:

1. Compatible with outboard and onboard mic-level preamp inputs, as Joseph stated.

2. Best noise figure with dedicated mic preamp IC's. If you look at the data sheet for the INA217 for example, the best noise figure is obtained with a low-Z input of 200 ohms source impedance.

3. 600-700 turns range used with a 1K:50K transformer gives OK bridging (not 1:10 but about 1:5) and 15 dB voltage step-up - and a good signal level so it only needs to be brought up 12-18 dB more for a very healthy signal to use with a volume control and send out of the instrument jack.

This "gain structure" if you will works quite well with a simple JFET booster such as the Fetzer Valve (runoffgroove.com), Stratoblaster, Bartolini TC-3, Artec VTB-1, etc. That's a big plus on battery draw depending on the booster, and the consensus is that JFET boosters sound great on guitar and bass. They aren't pure hi-fi but the slight even-order distortion is tasty and musical, and many players play with the outboard version of one in their signal chain.

That brings up an interesting point - theoretically we should need to bring the signal up 40 dBV (100X) since our AC voltage out should be 1/100th of the output of a high-Z pickup with the same construction and 10X the turns - but we don't need to, really. Overall 15 dB for the transformer, and about 12-18 dB for the preamp is fine. There's about 6-10 dB unaccounted for and my guess is that the resistance of the high-Z pickup is dissipating that much of what the "theoretical" signal would be with zero ohms and 6000 - 8000 turns. The eddy currents should be about the same for the same construction. Any thoughts on that?

Anyway, sorry to get so nitty-gritty but maybe the details will spark some discussion and questions in this thread.

[charrich56]

Mike,

That's a very valid approach. You end up with a lower signal level, but gain is almost free and most live guitarists in particular plug into a pedalboard, so that could be workable.

The big problem is that you can't control how and into what the guitarist is going to plug in and with what cable, and so it is devilishly hard to get a flat or even consistent frequency response if you want , say, pseudo-acoustic sounding tones, or going into a computer to get all the signal for recording with plugins.

I do agree that medium impedance would be better for this problem.

Making the guitarist always use the same cable and preamp box, everywhere, even for practice, might not go well with the guitar-slinger crowd, but I think bass players are in that sandbox right now. Except the ones with on-board active electronics in their basses, which might be a majority at this point.

The compromise is using a high-impedance in, low impedance out onboard buffer, but we have just introduced electronics back into the equation. Any self-respecting guitar tech would have to vote against putting a battery box and electronic fizzerdegibbits into a vintage instrument (unless reversible, and if there was a compelling sonic reason for it.) But the buffer or preamp does stand as the best way to mitigate the cable capacitance loss and "plugging into just about anything" problem.

We are just debating onboard versus offboard. Both approaches can work, but I would prefer a minimalist on-board approach myself that could be retrofitted without much or any pain, hassle, or sawdust flying. Even if you have to unscrew four screws to change the battery every year or so.

[Mike Sulzer]

"You end up with a lower signal level,"

The preamp that you would have used in the guitar is now outside, but it is not the same preamp since you can make a much better low noise preamp using more current than you would like to use inside, especially if you want the battery to last a year.

"The big problem is that you can't control how and into what the guitarist is going to plug in and with what cable, and so it is devilishly hard to get a flat or even consistent frequency response if you want..."

if you have reduced the number of turns by 6 to 8 times, keeping the coil geometry the same, then the inductance is down 36 to 64 times. The sensitivity to changes caused by different cable capacitance is much reduced. Here is what I calculate: Consider a 35 mH coil (100 ohm series resistance) and two total capacitances, 450 pf and 650 pf. Load with 7.5 K to damp the resonance, which is a bit above 40KHZ with the 450 pf C. Changing the C to 650 pf changes the response by a bit less than one db at 20KHz, which seems acceptable.

[charrich56]

Mike,

With your example you are right in the ballpark of the pickups I am playing around with right now. With 35 mH and 100 ohms DCR, the impedance of the thing at 1KHz is still around 200 ohms. I just wound a set with 35 mH and a bit over 50 ohms DCR.

So this is a good, microphone-type low impedance, and you are right, it is such low-Z that cable capacitance is simply not a problem for any sane cable length and decent quality mic cable.

You are also right that quality, low-noise, mic-level preamps with 30+ dB gain generally consume more current than you want to use for onboard, to have the battery (typically 9 Volt, 500 mAH) last for a long time.

I just threw out the "one year" because it is a desirable goal and also because my goal for the onboard hybrid transformer/JFET preamp I designed was to stay under 200 microamps current draw, so even with battery self-discharge, at least 2000 hours on a 9-volt battery. That's a lot of pickin'.

How I am "cheating" on the onboard electronics is using a small cheap audio transformer to get 15 dB of the 30 dB voltage gain that works in this design. (referred to in another of my posts on this thread.) That leaves about 15 dB which is just in the area that makes sense for a single transistor (BJT or JFET) to get that amount of gain with low enough noise and good to excellent current consumption depending on the device.

The small amount of distortion in these simple pre-amps might not make it in the hi-fi magazine review, but having built several of the JFET variety, I have played though and listened to them, and they sound really, really good.

There's also now some pretty low-power single-supply op amps that are probably (have to check more closely) under 1 mA which is not too bad for battery life. Generally there are some performance tradeoffs though for low power consumption, but there might be something good and quiet enough out there to use onboard if someone wants to go with that sort of preamp.

But yes, passive instrument and outboard electronics would work very well, too.

-Charlie

[bbsailor]

Mike,

With your example you are right in the ballpark of the pickups I am playing around with right now. With 35 mH and 100 ohms DCR, the impedance of the thing at 1KHz is still around 200 ohms. I just wound a set with 35 mH and a bit over 50 ohms DCR.

So this is a good, microphone-type low impedance, and you are right, it is such low-Z that cable capacitance is simply not a problem for any sane cable length and decent quality mic cable.

You are also right that quality, low-noise, mic-level preamps with 30+ dB gain generally consume more current than you want to use for onboard, to have the battery (typically 9 Volt, 500 mAH) last for a long time.

I just threw out the "one year" because it is a desirable goal and also because my goal for the onboard hybrid transformer/JFET preamp I designed was to stay under 200 microamps current draw, so even with battery self-discharge, at least 2000 hours on a 9-volt battery. That's a lot of pickin'.

How I am "cheating" on the onboard electronics is using a small cheap audio transformer to get 15 dB of the 30 dB voltage gain that works in this design. (referred to in another of my posts on this thread.) That leaves about 15 dB which is just in the area that makes sense for a single transistor (BJT or JFET) to get that amount of gain with low enough noise and good to excellent current consumption depending on the device.

The small amount of distortion in these simple pre-amps might not make it in the hi-fi magazine review, but having built several of the JFET variety, I have played though and listened to them, and they sound really, really good.

There's also now some pretty low-power single-supply op amps that are probably (have to check more closely) under 1 mA which is not too bad for battery life. Generally there are some performance tradeoffs though for low power consumption, but there might be something good and quiet enough out there to use onboard if someone wants to go with that sort of preamp.

But yes, passive instrument and outboard electronics would work very well, too.

-Charlie

Charlie,

Go to this web site to see how this Les Paul recording guitar was modified with active electronics. Les Paul Personal 2 If you use a totally passive system, the Shure A95U mic transformer has a 1:12 turns ratio so the impedance ratio is the square of that or 144. The high impedance side of the A95U, plugged into a 1Meg ohm guitar amp input impedance, reflects 1,000,000/144 or 6,944 ohms back into the primary side of the A95U transformer. The whole purpose of using 150 to 300 primary source impedance ranges is to select an impedance range where commercially available part values can be easily obtained. Once you get into the 600 to about 2,000 ohm range, you are into the medium impedance range. Consider rethinking about the guitar cable as simply being the transmission line and you want a flat response or predicted response at the other end of the transmission line.

Changes in the wire gauge will result in differing amounts of low cut depending on the DC resistance of the guitar pickup coil in relation to the input impedance of the mic matching transformer. Guitar pickups are typically low Q type coils unless you load the center of the coil with a large mass of ferrous metal which will increase the reactive part of the impedance. The upper pickup winding limit for a low impedance pickup using something like the Shure A95U mic matching transformer will be the the transformer input impedance staying about 10X higher than the source impedance across the frequency band of audio interest. With an A95U input impedance about 3K ohms, a 300 ohm coil impedance maximum is a good limit. This makes the limit on the A95U transformer pickup coil or 2 HB coils in series to be 48mH which has an XL of 301 ohms at 1000Hz.

Active electronics in the form of a high Z buffer only elevates the potential high frequency range of a pickup by eliminating the transmission line capacitance and input impedance from loading down the pickup. A 500K ohm volume pot in parallel with 1M ohm input impedance presents a 400K ohm resistive load plus the capacitance of the transmission line. Make the buffer input impdance 3M ohm to 5M ohm and the total load on the pickup will be near 500K ohms with no cable capacitance being reflected back into the pickup. A "Tillman buffer cable" (web search these words in quotes) allows your passive guitar to function like an active guitar with no mods and no on-board battery.

Once you have a higher frequency range of sounds you can choose to use it or attenuate it but you now have more choices. Totally passive high impedance pickups deprive you of this option.

I suspect that the wire gauge you will be using will be between AWG28 up to AWG34.

Joseph Rogowski

[Mike Sulzer]

"A 500K ohm volume pot in parallel with 1M ohm input impedance presents a 400K ohm resistive load plus the capacitance of the transmission line."

If you have a tone control on 10, don't forget to put another 500K in parallel. (The capacitor impedance is small compared to 500K at the frequencies that matter.)

[bbsailor]

"A 500K ohm volume pot in parallel with 1M ohm input impedance presents a 400K ohm resistive load plus the capacitance of the transmission line."

If you have a tone control on 10, don't forget to put another 500K in parallel. (The capacitor impedance is small compared to 500K at the frequencies that matter.)

Mike,

That is just one more reason why lower impedance pickups offer more potential upper harmonic content. The discussion issues then become:
1. Output level, mostly driven by coil turns and magnetic strength
2. Practical impedance range to match low input impedance devices
3. Method to boost voltage to a workable level. (A) Transformers with a step-up voltage ratio probably between 10 to 12 so the stepped up turns count will match passive pickup turns count (B) Active boost with gain to boost the 5mv to 10mv low impedance signal back to near the 150mv range where many passive pickups operate.
4. Active buffering (Tillman active pickup buffering) using high impedance pickups where the guitar end of the .25" plug has an active FET representing a 3 meg ohm to 5 meg ohm load on the guitar but also isolating the pickups from the 350+ pfs on the guitar cable that tends to shift down the resonant point.
5. Noise level. High impedance pickups also make good noise detection antennas. Sometimes reoriention of the guitar can reduce the noise but operating the high impedance guitar near a computer is a good test.
6. Manufacturing techniques and materials to produce consistent and desireable pickup results. These desired results and be catagorized into pickup types that have (1) Coil Q: in the high, medium, or low range; (2) Resonant Point: from about 2Khz up to about 10Khz (3) output level: to adequately drive effects or stay clean sounding.
7. Guitar construction technique that allows more energy to stay within the strings in the form of more sustain and more high frequency energy present while in the decay mode. This can be heard on guitars made from carbon fiber materials as well as traditional wood guitars that use (1) neck through body designs, (2) set necks, (3) well fitted bolt on necks.

The Alumitone pickups were a major depatrure from high impedance pickup thinking. However, they are still designed to drive a high impedance amp input and have two hum cancelling coils, made with very thin wire, under the metal frame body to convert the current being induced into the matal frame primary coils into a voltage as in a current transformer.

Finding a more efficient way to get low noise signals from the guitar to the amp or other input device requires looking at how other devices have solved this problem. Microphones in the 150 to 300 ohms range are now standard. Low impedance pickups (150 ohm to 300 ohm) seem to be within the one tenth the number of turns range also and are tunable by how much air or ferrous material is in the pickup core with from 500 to 1000 turns of heavier wire.

Just think how Fishman could optimize their new layered pickup concept. They could have a coil platform for all standard pickup footprints such as SC Strat, SC Tele, Humbucker, P90, Etc. and offer a coil stack arrangement that sounds like that classic pickup sound but with some more variable possibilities due to active tone shaping, gain and buffering.

When pickup makers start looking into the previous dark zone, maybe they will find some sounds worthy of listening to or adding to our palate of pickup sounds.

Joseph Rogowski

[charrich56]

Joseph,

With you on the points above.

The current approach I am working with is (with minimal parts count) in circuit simulation, giving a frequency response (with no voicing capacitors added) of -3 dB, 80 Hz to about 22.5 KHz at the amp input with about 450 pF cable capacitance and 1 megohm amp input resistance.

That was my original design goal, to have a truly flat guitar pickup system to run into the computer for recording and plugins.

The secondary goal was to be able to easily voice the pickup with any resonant peak desired to reasonably emulate a high-Z pickup (within limits due to construction, position, magnetic structure, etc .)

I am still in the process of verifying this design in "flat mode" but will do so soon with a driver coil and my new friend, the Syscomp CGM-101 in network analyzer mode. Also will finish up the humbucker and build a couple more preamps to test.

-Charlie

[bbsailor]

Joseph,

With you on the points above.

The current approach I am working with is (with minimal parts count) in circuit simulation, giving a frequency response (with no voicing capacitors added) of -3 dB, 80 Hz to about 22.5 KHz at the amp input with about 450 pF cable capacitance and 1 megohm amp input resistance.

That was my original design goal, to have a truly flat guitar pickup system to run into the computer for recording and plugins.

The secondary goal was to be able to easily voice the pickup with any resonant peak desired to reasonably emulate a high-Z pickup (within limits due to construction, position, magnetic structure, etc .)

I am still in the process of verifying this design in "flat mode" but will do so soon with a driver coil and my new friend, the Syscomp CGM-101 in network analyzer mode. Also will finish up the humbucker and build a couple more preamps to test.

-Charlie

Charlie,

Calculate your impedance of the pickup output at resonance. Multiply that impedance value by 10 to define the buffer load that will not affect the resonant peak ans isolate the pickup coil.

Now when you do simulations in this lower impedance range, the cable capacitance is not needed in your model and you can see the pickup in more isolation for a better comparison. You can add variable mid range frequency shift with boost or cut and active stacked treble/bass boost/cut like the EMG-Control.

Joseph Rogowski

[David Schwab]

For a number of years I made bass humbuckers with about 1,000 turns of wire on each coil, and then used a simple JFET preamp for each pickup. They had a very extended frequency response. The preamps ran off a 9v battery which lasted a very long time.

This is a sound clip from 2006:

https://soundcloud.com/davidravenmoo...-23-06/s-Igl2C

I believe this was the bridge pickup soloed.