Low Impedance Pickup Research

[fieldwrangler]

I have been reading and doing my best to mentally absorb as much of this 'alien pickup technology' as I can. I'm getting close to being able to wrap my head around it so for that, much thanks to you far more knowledgeable guys in this area (especially bbsailor) for enlightening the rest of us. Anyways the point of me responding to this thread is that I wanted to tell you all that I might have found an easily available way to get those types of laminations used in the Alumitone p/ups; from the transducers of reverb tanks! I was going through my stuff looking for something and an old reverb tank fell out of the box I had it in and landed top side down. Upon retrieving it from the floor I looked at the transducers and realized they were at the least similar as all the pics/drawings I've seen from the underside of Alumitones and their current transformer secondary. Do you think they would work with this type of thing? [ATTACH=CONFIG]32575[/ATTACH][ATTACH=CONFIG]32576[/ATTACH][ATTACH=CONFIG]32577[/ATTACH]

Capehead -

Indeed, I mentioned these reverb tank laminations in my first post on recent prototypes (around the middle of page 4 of this thread I think.) I used them on one of the earliest versions I tried after coating them with lacquer. Still have some around, but I'm now trying other materials; not sure where it will all end up.

Thanks, though, both for highlighting this resource and for your excellent illustrations!

While here, I'll add that I've been making transformer bobbins using rectangular model railroad tubing for the cores and nylon washers (slightly filed out) slipped over sections of the tubing for flanges.

[capehead]

I have been reading and doing my best to mentally absorb as much of this 'alien pickup technology' as I can. I'm getting close to being able to wrap my head around it so for that, much thanks to you far more knowledgeable guys in this area (especially bbsailor) for enlightening the rest of us. Anyways the point of me responding to this thread is that I wanted to tell you all that I might have found an easily available way to get those types of laminations used in the Alumitone p/ups; from the transducers of reverb tanks! I was going through my stuff looking for something and an old reverb tank fell out of the box I had it in and landed top side down. Upon retrieving it from the floor I looked at the transducers and realized they were at the least similar as all the pics/drawings I've seen from the underside of Alumitones and their current transformer secondary. Do you think they would work with this type of thing?

[bbsailor]

Bob,

To help you with your experiments check out this web sit for access to good low resistance wire. Copper Magnet / Enameled & Bare Wires

At the bottom of the page, sold by the foot are (1) squate AWG 7 copper wire .150" X .150" and (2) 6 AWG .250" diameter of many thinner strands.

The trick to efficiently using them is to make low resistance attachments to the CSE187L CT primary by drilling a hole in the wire, Heat the wire to expand it then insert the CT primary and then solder to keep it stable. One way to see if the connection is low enough is to measure the CT output impedance and see of the calculated value is close to what you are reading on the Extech LCR meter. Wiggle the connection to make sure the LCR reading stays stable. When using toroids, only one connection is needed to join the string loop together as it is the same wire is going through the toroid CT.

If you want to experiment with tonal variances, try this. Use a solid string loop listen and measure. Then use smaller wire like AWG 16 (good for up to 12KHz due to the skin effect) or AWG 18 (good for up to 18KHz due to the skin effect). Wire these loops as a a bunch of parallel string loops independently looped through either a toroid CT or a Prem Magnetics SPCT-251 with an open primary space of .160" square and each soldered together. This will form a Litz-like multi strand, insulated wire bundle. Measure the impedance, and listen as each loop is added to the bunch.

The reason why this is important is that you want to hear all the harmonics that are potentially present. Since this is a current based pickup, you want to make sure that any upper harmonics are given a fair chance to make their way into the amp or the test recording and not get lost due to the skin effect attenuation by the higher AC wire resistance of these upper harmonics.

I have done many of my experiments by using craft sticks (ice cream pop sticks or tongue deressors) to make "quick to mount" acoustic guitar pickups where these upper harmonics really contribute to the more acoustic sound vice the electric guitar sound. Here is where using various wire types (single large area or multiple smaller) and diameter of string loop(s) wires comes into play. Since acoustic guitars only have magnetic cores, the bass strings need augmentation to sound balanced. Here, I raise the lower end of the pickup and lower the treble end. The B string still sounds dominant so I then try to make about a .125" deep dip under the B string or even mount the CT to a string loop where the CT primary is directly below the B string. This makes a very large .375" wide deep dip as the string loop connect to the CT. Then you need to try adding a small dip on the opposite side of the string loop to be most effective.

The creativity of using these types of pickup is mostly practical and mechanical. Magnet strength, type and strength are easy variables to change with this type of pickup. Check out Magnets - Master Magnetics, Inc. for their part number 07001, a pack of 8 ceramic magnets .222" wide .875 long and .196 thick, magnetized through the thickness. Radio Shack even has part number 64-1895, 3/16" dia and almost 1/16" thick rare earth magnets that allow some very small pickups to be fabricated. I can even see the day when a small current based pickup can be placed just beyond the last fret on the fingerboard with the magnets and string loop hidden under the fingerboard with the CT being attached vertically (primary orientation) just in front of the 4" sound hole opening.

This is an area of pickup fabrication where learning by doing opens new doors of possibilities.

Joseph Rogowski

[fieldwrangler]

Bob,

Good to see you getting into this crazy new world of current-based pickups. See this web link:Les Paul Personal. Scroll through the pages and see how the Les Paul Recording guitar used a 2.5K ohm volume pot and other low ohm pots for tone control to have on board controls. It would be best to keep the A95 matching transformer at the amp end of the cable for less loss of high frequencies. Try making the LP recording alumium control test box that you can run the output of the CSE187L directly into either with 3 binding posts for the 3-pin XLR input connections or an XLR connector in the control box. Then you can see what controls you may want to migrate onto the guitar.

This approach is very likely to help me hear what happens with the insertion of this variable resistance in the setup.

You can take an old credit card and cut it into a .187 wide rectangle with a pointed arrow tip to poke between the CSE187L secondary coil and the E-I transformer frame so you can mount additional U-shaped primary turns (up to about 4 extra) that connect to the string loop to lower the impedance and reduce leakage inductance. You should see about about a 10 to 12 ohm reduction for adding two AWG 20 additional string loops in parallel. Remember the CSE187L will have an AWG 12 primary that is 250 micro ohms and that resistance will limit the lowest impedance the string loop can make and also affect the tone. The calculated output of a CSE187L current transformer with the existing primary winding can be bent and and aligned to form a very small primary loop soldered together. The Extech measurement will be about 80 ohms. The calculated impedance of this 1.8 inch long wire loop should be 250 micro ohms (.000250 ohms) times 250,000 (turns ratio squared) or 62.5.
Use this handy resistance calculator Resistivity Calc to accurately calculate the added impedance of the string loop DCR for a 5 to 6 inch string loop. A 5 inch long piece of AWG 11 will be 105 micro ohms per inch so a 5 inch length will be 525 micro-ohms or .000525 times 250,000 is 131.25 ohms added to the 80 ohms CT primary loop impedance to be measured on the Extech to be near 211 ohms as the output impedance.

What causes the 80 ohms Extech reading is the added leakage inductance of the added 17.5 ohms that forms a series inductance to raise the impedance of the primary loop slightly which is then reflected into the secondary reading.

Let you ear be the final judge of the effect of using various gauges of string loop wire. Note that very high frequency harmonics may have a harder time making its way through very thick string loop wire due to the skin effect where high frequencies do not fully penetrate the full wire area and only travel at the surface. Look for a skin effect calculator on the web.

At present the sense coil is behaving (and measuring) very well. In fact, I may try less cross-sectional area in an attempt to drop the low & low-mid response, partially to avoid potential feedback issues in live situations. Winding a small drive coil and playing music through it is going to be part of the process of voicing this rig.

You can use a vice to press out the CSE187L primary wire. Heat the primary to loosen the glue on the primary to make removing the primary easier. Make a spacer to shim the transformer frame firmly against the vice jaws to remove the primary without breaking the CT. This will now allow you to put in a little thicker AWG 11 wire but make it a 7 inches long to form the string loop so the CT primary and string loop are the same wire with only one joint to complete the loop (less resistance). You can then also try making your own Litz wire by using a bundle of AWG 18 or 20 wire to fill the primary CT wire space with multiple parallel string loops individually joined together.

In fact I used a vise (with a hex wrench placed behind the primary loop) to press the ends into ultra-snug precision holes in the copper sense loop.

If you want to keep the AWG 12 CSE187L primary look to poke a hole right next to each side of primary wire and the E-I metal core so you can easily squeeze two additional AWG 18 or 20 primary CT turns connected to the string loop to lower the impedance slightly. Lay these two thinner wires on top of the heavier wire and dip them down under the B string (sometimes the G string) to lower or balance the output of this typically dominant string. Here is where the creative manipulation of the hairpin string loop and be shaped to make about a .125" dip under the dominant string. You will need to make the string loop about .5 to .7 inches longer to account for the wire dip length (on each side of the string loop) needed to span the strings. Try it on scrap wire first and measure the added length to fit your specific design requirements.

Keep good notes as you alter these primary manipulations on your Extech measurements. Note the Q readings also.

There are other makers of current transformers to look up. Here are a few CT sources to research and play with. From my experience the most practical number of turns is near 500 turns turns 300 will also work up to about 750 turns. At 1000 turns the output impedance will be about 4 times higner than the 500 turn output. If you use a 1000 turn CT then you may need to put in loading resistors of about 300 ohms on each side of the secondary going to ground which is the metal frame of the CSE187L which is also connected to the shield (pin 1 of the XLR) with the two center conductors connected to pins 2 and 3 of the XLR. Try using small variable pot of about 1 to 5 K across the CT output and lower the resistance and listen for a change in the tone and or noise and measure at the desired rotation point and put in two resistors of one half that value on each side of the secondary to ground. If you are using a neo magnet with a metal coating, the run a wire under the magnet to ground it to the shield connection point to minimize induced noise pickup.

Zettler
Prem Magnetics SPCT-251
Minntronix

What is the gauge or size of the string loop wire?

Cross sectional area is about .04 square inches.

What type of magnets are you using?

Hasn't been fully established, at the moment they're Samarium Cobalt discs.

What is your Extech measured (both 120 Hz and 1KHz) AC resistance, inductance and Q of your CSE187L pickup?

I'm not at the shop at the moment but I seem to recall something like 64mh/164 ohms @ 120 and 31mh/196 ohms @ 1k measured at the secondary outputs of the CSE187L.

Post your results.

Joseph Rogowski

I'll be back in the lab next week.

I really appreciate the comprehensive & detailed information you've posted here, and will do what I can to add to the knowledge pool.

Bob Palmieri

[bbsailor]

Folks -

I've set up a known interface for my sense coil using the specified CSE187L and the Shure 95 series low-to-high Z transformers. Measurements of inductance and impedance with the Extech are exactly as predicted and output is quite good. I'm now using this as a benchmark for what this single turn sense coil is capable of.

I'm continuing to pursue an in-house single transformer solution (which is going surprisingly well) partially because with the CSE187 on the instrument and the Shure at the amp end of the cable the idea of an instrument-mounted volume control looks pretty daunting to me, seeing as how we're dealing with power transfer between stages, intermediate termination impedance issues, etc. I suppose I could try to put both transformers on the instrument (although I'm not sure there's gonna be room for that) and sacrifice the balanced cable run...

Anything I'm not considering here that I should?

Bob Palmieri

Bob,

Good to see you getting into this crazy new world of current-based pickups. See this web link:Les Paul Personal. Scroll through the pages and see how the Les Paul Recording guitar used a 2.5K ohm volume pot and other low ohm pots for tone control to have on board controls. It would be best to keep the A95 matching transformer at the amp end of the cable for less loss of high frequencies. Try making the LP recording alumium control test box that you can run the output of the CSE187L directly into either with 3 binding posts for the 3-pin XLR input connections or an XLR connector in the control box. Then you can see what controls you may want to migrate onto the guitar.

You can take an old credit card and cut it into a .187 wide rectangle with a pointed arrow tip to poke between the CSE187L secondary coil and the E-I transformer frame so you can mount additional U-shaped primary turns (up to about 4 extra) that connect to the string loop to lower the impedance and reduce leakage inductance. You should see about about a 10 to 12 ohm reduction for adding two AWG 20 additional string loops in parallel. Remember the CSE187L will have an AWG 12 primary that is 250 micro ohms and that resistance will limit the lowest impedance the string loop can make and also affect the tone. The calculated output of a CSE187L current transformer with the existing primary winding can be bent and and aligned to form a very small primary loop soldered together. The Extech measurement will be about 80 ohms. The calculated impedance of this 1.8 inch long wire loop should be 250 micro ohms (.000250 ohms) times 250,000 (turns ratio squared) or 62.5.
Use this handy resistance calculator http://www.salvarsan.org/pickups/ResistivityCalc.html to accurately calculate the added impedance of the string loop DCR for a 5 to 6 inch string loop. A 5 inch long piece of AWG 11 will be 105 micro ohms per inch so a 5 inch length will be 525 micro-ohms or .000525 times 250,000 is 131.25 ohms added to the 80 ohms CT primary loop impedance to be measured on the Extech to be near 211 ohms as the output impedance.

What causes the 80 ohms Extech reading is the added leakage inductance of the added 17.5 ohms that forms a series inductance to raise the impedance of the primary loop slightly which is then reflected into the secondary reading.

Let you ear be the final judge of the effect of using various gauges of string loop wire. Note that very high frequency harmonics may have a harder time making its way through very thick string loop wire due to the skin effect where high frequencies do not fully penetrate the full wire area and only travel at the surface. Look for a skin effect calculator on the web.

You can use a vice to press out the CSE187L primary wire. Heat the primary to loosen the glue on the primary to make removing the primary easier. Make a spacer to shim the transformer frame firmly against the vice jaws to remove the primary without breaking the CT. This will now allow you to put in a little thicker AWG 11 wire but make it a 7 inches long to form the string loop so the CT primary and string loop are the same wire with only one joint to complete the loop (less resistance). You can then also try making your own Litz wire by using a bundle of AWG 18 or 20 wire to fill the primary CT wire space with multiple parallel string loops individually joined together.

If you want to keep the AWG 12 CSE187L primary look to poke a hole right next to each side of primary wire and the E-I metal core so you can easily squeeze two additional AWG 18 or 20 primary CT turns connected to the string loop to lower the impedance slightly. Lay these two thinner wires on top of the heavier wire and dip them down under the B string (sometimes the G string) to lower or balance the output of this typically dominant string. Here is where the creative manipulation of the hairpin string loop and be shaped to make about a .125" dip under the dominant string. You will need to make the string loop about .5 to .7 inches longer to account for the wire dip length (on each side of the string loop) needed to span the strings. Try it on scrap wire first and measure the added length to fit your specific design requirements.

Keep good notes as you alter these primary manipulations on your Extech measurements. Note the Q readings also.

There are other makers of current transformers to look up. Here are a few CT sources to research and play with. From my experience the most practical number of turns is near 500 turns turns 300 will also work up to about 750 turns. At 1000 turns the output impedance will be about 4 times higner than the 500 turn output. If you use a 1000 turn CT then you may need to put in loading resistors of about 300 ohms on each side of the secondary going to ground which is the metal frame of the CSE187L which is also connected to the shield (pin 1 of the XLR) with the two center conductors connected to pins 2 and 3 of the XLR. Try using small variable pot of about 1 to 5 K across the CT output and lower the resistance and listen for a change in the tone and or noise and measure at the desired rotation point and put in two resistors of one half that value on each side of the secondary to ground. If you are using a neo magnet with a metal coating, the run a wire under the magnet to ground it to the shield connection point to minimize induced noise pickup.

Zettler
Prem Magnetics SPCT-251
Minntronix

What is the gauge or size of the string loop wire?
What type of magnets are you using?
What is your Extech measured (both 120 Hz and 1KHz) AC resistance, inductance and Q of your CSE187L pickup?

Post your results.

Joseph Rogowski

[fieldwrangler]

Folks -

I've set up a known interface for my sense coil using the specified CSE187L and the Shure 95 series low-to-high Z transformers. Measurements of inductance and impedance with the Extech are exactly as predicted and output is quite good. I'm now using this as a benchmark for what this single turn sense coil is capable of.

I'm continuing to pursue an in-house single transformer solution (which is going surprisingly well) partially because with the CSE187 on the instrument and the Shure at the amp end of the cable the idea of an instrument-mounted volume control looks pretty daunting to me, seeing as how we're dealing with power transfer between stages, intermediate termination impedance issues, etc. I suppose I could try to put both transformers on the instrument (although I'm not sure there's gonna be room for that) and sacrifice the balanced cable run...

Anything I'm not considering here that I should?

Bob Palmieri

[fieldwrangler]

Let's consider a thought experiment; maybe not take the results as 100% real but at least an indication of what to expect.

Measure the impedance of some normal pickup, as you normally would, but let's insert an ideal one to one transformer between the pickup and the measuring instrument. We lose the dc resistance, of course, but the ac measurements should be the same.

Now make another pickup exactly the same, except using half the number of turns with wire with twice the cross section. Wind it so that the coil geometry is the same. Both the inductance and the resistance will be four times lower than the original pickup.

Now measure, but use an ideal transformer that has a two to one step up turns ratio, or an impedance ratio of four to one. With this setup, it should measure the same as the original. (We are ignoring any possible differences in capacitance, so we will restrict the measurements to low and medium frequencies.)

You can keep on doing this until you get down to just one turn made out of a special piece of machined copper that preserves the coil geometry, and you should still measure the same.

The Lace pickup under discussion is not the same as any other pickup, and maybe there are ways to excite some currents in that "frame coil" that do not just go around as you want in a coil. But it is still a pretty simple thing, while a real audio transformer is a complicated beast. So I am betting on the transformer, but you would have to disect one and make some really careful measurements to find out for sure.

Mike -

This really is an excellent way to visualize the connection between the familiar world of high-Z sources and the wild & wooly one of transformed current sources.

Bob Palmieri

[Mike Sulzer]

Any chance these eddy current losses are originating in the unlaminated primary sense coil?

Let's consider a thought experiment; maybe not take the results as 100% real but at least an indication of what to expect.

Measure the impedance of some normal pickup, as you normally would, but let's insert an ideal one to one transformer between the pickup and the measuring instrument. We lose the dc resistance, of course, but the ac measurements should be the same.

Now make another pickup exactly the same, except using half the number of turns with wire with twice the cross section. Wind it so that the coil geometry is the same. Both the inductance and the resistance will be four times lower than the original pickup.

Now measure, but use an ideal transformer that has a two to one step up turns ratio, or an impedance ratio of four to one. With this setup, it should measure the same as the original. (We are ignoring any possible differences in capacitance, so we will restrict the measurements to low and medium frequencies.)

You can keep on doing this until you get down to just one turn made out of a special piece of machined copper that preserves the coil geometry, and you should still measure the same.

The Lace pickup under discussion is not the same as any other pickup, and maybe there are ways to excite some currents in that "frame coil" that do not just go around as you want in a coil. But it is still a pretty simple thing, while a real audio transformer is a complicated beast. So I am betting on the transformer, but you would have to disect one and make some really careful measurements to find out for sure.

[fieldwrangler]

HAPPY NEW YEAR readers and contributors to this Low Impedance Pickup Research thread.

It has always been my goal to share what I have learned from over 50 years of tinkering with guitars and pickups. This Music-Electronics Forum is a world-wide and highly respected source of technical information.

MEF members, thanks for the questions, comments, and challenges (including Mike Sulzer) along with the variety of World-wide low impedance pickup initiatives that were started based on this thread.

Right now I am working on presenting to the MEF members a low impedance pickup that can be easily mounted on an acoustic flat top guitar with about a 4" hole just below the fingerboad with no modifications to the guitar. This will be a common learning platform to collectively share and try the unusual world of very low impedance, single string loops and current transformer based guitar pickups.

I'll share more low impedance pickup technical information in 2015.

Stay safe and keep on tinkering!

Joseph Rogowski

I must say... HAPPY NEW YEAR!!!! And, in fact, further disclosures from the cognoscenti will certainly contribute to mine being so.

[fieldwrangler]

Your readings suggest that the eddy currents are large, larger than a normal humbucker. The construction suggests that these losses are in the transformer, since the magnets appear to be low loss ceramic, and I do not think there is any iron in the design except the transformer core. If so, from the view of normal transform design, it is a piece of crap. On the other hand, maybe it is exactly what is needed to give the sound he wants. Not sure how to find out except to dissect one; the word from the man himself is kind of dubious: he appears to claim that aluminum is a better conductor than copper.

Any chance these eddy current losses are originating in the unlaminated primary sense coil?

[bbsailor]

HAPPY NEW YEAR readers and contributors to this Low Impedance Pickup Research thread.

It has always been my goal to share what I have learned from over 50 years of tinkering with guitars and pickups. This Music-Electronics Forum is a world-wide and highly respected source of technical information.

MEF members, thanks for the questions, comments, and challenges (including Mike Sulzer) along with the variety of World-wide low impedance pickup initiatives that were started based on this thread.

Right now I am working on presenting to the MEF members a low impedance pickup that can be easily mounted on an acoustic flat top guitar with about a 4" hole just below the fingerboad with no modifications to the guitar. This will be a common learning platform to collectively share and try the unusual world of very low impedance, single string loops and current transformer based guitar pickups.

I'll share more low impedance pickup technical information in 2015.

Stay safe and keep on tinkering!

Joseph Rogowski

[Mike Sulzer]

And for someone like you who actually has a clue about how the Extech is arriving at its display figures my report on its readings of the Lace Alumitone aggregation might give a clue to this unknown.

Your readings suggest that the eddy currents are large, larger than a normal humbucker. The construction suggests that these losses are in the transformer, since the magnets appear to be low loss ceramic, and I do not think there is any iron in the design except the transformer core. If so, from the view of normal transform design, it is a piece of crap. On the other hand, maybe it is exactly what is needed to give the sound he wants. Not sure how to find out except to dissect one; the word from the man himself is kind of dubious: he appears to claim that aluminum is a better conductor than copper.

[fieldwrangler]

A physicist might be dismissive at first, but if he is any good, he will eventually see several important factors, such as:
1. What really determines the frequency response and the harmonic response?
2. What do those cores do? (eddy currents, maybe?)
3. Just how linear is this thing?

And so on.

Indeed he or she would.

And this is precisely how I feel when looking at a "simple transformer."

[Mike Sulzer]

Although I gotta say, impedance ratios with so many zeros that I favor scientific notation do scare me a bit.

Also, saying what I said about the Lace transformer is uncomfortably resonant with something a physicist said recently after glancing at an exposed bridge bucker on an Epiphone Les Paulish thing - "What is there to do, it just looks like a solenoid to me..."

A physicist might be dismissive at first, but if he is any good, he will eventually see several important factors, such as:
1. What really determines the frequency response and the harmonic response?
2. What do those cores do? (eddy currents, maybe?)
3. Just how linear is this thing?

And so on.

[fieldwrangler]

Also unknown is how good it is.

And for someone like you who actually has a clue about how the Extech is arriving at its display figures my report on its readings of the Lace Alumitone aggregation might give a clue to this unknown.