Low Impedance Pickup Research

[bbsailor]

Today i bought some copper wire, 6 mm**2 (between AWG 9 and 10). Now it is time to order the CTs and to decide which size magnets would be best (among many other things...)

Bea,

Make sure that the chosen string loop wire will fit inside the CT opening. If it is a little tight but close to fitting, you can file the opening a little. Start out with from 500 to 750 current transformer secondary turns to keep the output impedance in the XLR mic input impedance range of between 1500 ohms to 2500 ohms for the typical mixer input. Check your mixer specification to see what your actual input impedance is to make the best match of one tenth down to one fifth the actual input impedance of the mixer. However, at one fifth the input impedance, the loading might be audible on the high end.

Joseph Rogowski

[bea]

AWG 9-10, what do you do with wire that large?

That shall become a single loop. The current transformer i am going to use has a 5 mm opening, i.e. an area of about 19.5 mm^2. Plenty enough for the 6 mm^2 of the wire i think. But before i start with these the pair of more traditional LoZs (P90/Bisonic inspired bass pickups) i am currently working on must be finished.

[tedmich]

I wanted to mention Uli Jon Roths Sky Guitar with its (mighty/mega)Wing pickups in any discussion of low impedance designs. The newest Megawing Dommenget pickups are reportedly patented low impedance design that uses a preamp designed by John Oram
http://www.guitarmaker.de/Sky.html
Musical Electronics
R-O-D Systems? LOL

Almost every Uli fan agrees the Sky guitar sounds inferior to his old Strat/Super Lead setup, so go figure...

[bea]

Maybe, nothing for Rock'n Roll, but definitely those parts of Metal Rick Toone appears to have in mind with many of his creations - and he is using alumitones..

But i am planning something wholly different:

a) magnetic pickups for an acoustically fully usable archtop guitar which shall reproduce the acoustical sound of the guitar. That's also possible with traditional LoZ designs, but in this special case the unconventional approach discussed here appears to be the way to go. Please see above.

b) the sound characteristics of an archtop guitar with traditional HiZ pickups, which i hope to achieve with a state variable filter in the preamp.

c) bass guitars. I have plans to try to build an acoustic archtop bass.

[fieldwrangler]

Folks -

I've been looking for an opportunity to use this technology and now have the ideal commission. So, I formed a primary sensing loop from copper with a rectangular cross section that's basically .360" x .062". This application almost definitely needs to have a conventional high-impedance output from the instrument, so I'm trying to make an appropriate hum-cancelling transformer. My first approximation unit for this purpose was made from bobbins & laminations harvested from a couple of broken reverb tanks. I've wound 5,000 turns of wire around two of the bobbins and arranged them in hum-cancelling orientation at opposite ends of 5 layers of nested C-cores (.014" thick) and wrapped the single-turn primary sense coil around one of the ends of the resultant rectangle about 340 degrees.

However, the readings I'm getting from the mighty Extech are a bit puzzling. With the coils in series I'm reading the following results:

Unterminated:

Z@120hz - 7.5H

Z@1K - 4H

ACR@120Hz - 4.3 k

ACR@1K - 11.6K

Q@120Hz - 1.35

Q@1K - 2.2

Terminating with 250K results in slight proportional reductions in impedance of less than 10%.

Output (with well-arranged magnet arrays) is much lower than I had predicted and there's not nearly the nice, extended high end I had expected from this device.

Do we think that the coupling might be inefficient in the high frequencies? Possible eddy current losses in the transformer?

My next step is going to be to try to assemble a better transformer, with a combination of better silicon steel laminations (with better insulation in between them, along with some other bobbins of various specs.

I'd really appreciate some suggestions from those of you with far more experience than I in this area; Ill do my best to present my findings in a way that could benefit this thread.

Bob Palmieri

[bbsailor]

Folks -

I've been looking for an opportunity to use this technology and now have the ideal commission. So, I formed a primary sensing loop from copper with a rectangular cross section that's basically .360" x .062". This application almost definitely needs to have a conventional high-impedance output from the instrument, so I'm trying to make an appropriate hum-cancelling transformer. My first approximation unit for this purpose was made from bobbins & laminations harvested from a couple of broken reverb tanks. I've wound 5,000 turns of wire around two of the bobbins and arranged them in hum-cancelling orientation at opposite ends of 5 layers of nested C-cores (.014" thick) and wrapped the single-turn primary sense coil around one of the ends of the resultant rectangle about 340 degrees.

However, the readings I'm getting from the mighty Extech are a bit puzzling. With the coils in series I'm reading the following results:

Unterminated:

Z@120hz - 7.5H

Z@1K - 4H

ACR@120Hz - 4.3 k

ACR@1K - 11.6K

Q@120Hz - 1.35

Q@1K - 2.2

Terminating with 250K results in slight proportional reductions in impedance of less than 10%.

Output (with well-arranged magnet arrays) is much lower than I had predicted and there's not nearly the nice, extended high end I had expected from this device.

Do we think that the coupling might be inefficient in the high frequencies? Possible eddy current losses in the transformer?

My next step is going to be to try to assemble a better transformer, with a combination of better silicon steel laminations (with better insulation in between them, along with some other bobbins of various specs.

I'd really appreciate some suggestions from those of you with far more experience than I in this area; Ill do my best to present my findings in a way that could benefit this thread.

Bob Palmieri

Bob,

Post a photo of your design.

Also, try wiring the coils in parallel and then measure and post results. What is the Peak to Peak output? What type of magnets are you using?

Look at the Alumitone pickup design by Lace and see how they use the primary low Z string loop as two coils with a north pole magnet in one loop and a south magnet in the other loop. They place the C shaped laminated cores where the center piece joins the outer loops or where the two primary string loops are in series, then the two under frame coils on the lamination are wired in a humbucking arrangement. Lace uses two coils that use about AWG 44 (maybe a litte finer) and have about 10,000 to 15,000 turns per coil and I believe are wired in parallel to be rated at 2.5K ohms but can also be wired in series for a different sound.

With these low level signals you want to use the highest permeability laminated C cores that you can find. Use the MEF search feature and locate some posted photos of Alumitone pickups (under side view and side view) to visualize what I have said above. Another thing to consider is that the resistance of the two copper string loops needs to very very low to generate the most current and be the most efficient.

Do this experiment. Obtain and strip a 6" length of AWG 6 copper wire. Form a string loop wide enough to span the string width (usually about 2.125") and join with a piece of copper tubing, cleaned with fine sandpaper, crimped and soldered. The output impedance should be the calculated DC resistance of that wire loop about 35 micro ohms per inch (look it up here http://www.salvarsan.org/pickups/ResistivityCalc.html) times the turns ratio squared. Then, add about 15 percent for leakage inductance and you should be very close to the Extech LCR measured impedance of one of your secondary coils.

I hope this gets you going in the right direction.

Joseph Rogowski

[fieldwrangler]

Joseph -

Thanks so much for this. Can't shoot this at the moment, but will a bit further down the line. The shape of the primary sense coil is in flux, so to speak.

In the next 48 hours I plan to execute much of the plan you've outlined. Way better core material, more winds on the coils and a new shape that eliminates the need for a solder joint are high priorities, and should be addressed within a week.

Interestingly, I grabbed a Lace Alumitone Strat pickup for comparison; the Extech readings from that device are even more puzzling (to me) than the readings I'm getting from my own unit:

Lace Strat (unterminated):

Z@120hz - 28.3H

Z@1K - 4.5H

ACR@120Hz - 50k

ACR@1K - 64.5K

Q@120Hz - 0.44

Q@1K - 0.44

Once again, reasonable termination results in proportionate rescaling but still a much higher inductance reading than I'd expect to see at 120Hz and the same wide disparity between those readings and the ones at 1k.

Bob Palmieri

[fieldwrangler]

Incidentally, anyone have any favorite sources for transformer core material and little bobbins in little quantities?

[Mike Sulzer]

Output (with well-arranged magnet arrays) is much lower than I had predicted and there's not nearly the nice, extended high end I had expected from this device.

There is no reason to expect a low impedance pickup to have better high frequency response than a high impedance pickup if the impedance is transformed to high before passing through the cable from the instrument to the amp or board.

There are various reasons for using the singe loop pickup design, but if your primary objective is good high frequency response, the easiest approach is to wind a normal pickup with fewer turns if this meets the physical (space) requirement. Of course a single loop pickup can be brought to an intermediate (say microphone) impedance in the guitar and a transformer used at the amp, or a preamp designed for good SNR at low (microphone) impedance can be used in the instrument or at the amp.

[Mike Sulzer]

Incidentally, anyone have any favorite sources for transformer core material and little bobbins in little quantities?

Designing and building high quality audio transformers is not easy. A good approach with low impedance pickups for high fidelity is to design the pickup in such a way that you can use a good commercially available transformer.

[fieldwrangler]

Agreed on both counts. However, I'm not sure that a single transformer exists for this application, and the client would be MUCH happier if the connection to the amplifier followed the conventional 1/4" unbalanced configuration that has become the unfortunate industry standard.

And, although I know next to nothing about transformer design I must say that the little unit on the Lace sensor doesn't appear to be terribly exotic.

[fieldwrangler]

There is no reason to expect a low impedance pickup to have better high frequency response than a high impedance pickup if the impedance is transformed to high before passing through the cable from the instrument to the amp or board.

There are various reasons for using the singe loop pickup design, but if your primary objective is good high frequency response, the easiest approach is to wind a normal pickup with fewer turns if this meets the physical (space) requirement. Of course a single loop pickup can be brought to an intermediate (say microphone) impedance in the guitar and a transformer used at the amp, or a preamp designed for good SNR at low (microphone) impedance can be used in the instrument or at the amp.

In this case, for aesthetic reasons I need to give the single-turn unit set up for high impedance output conditions every chance before going another way.

[fieldwrangler]

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..."

[Mike Sulzer]

And, although I know next to nothing about transformer design I must say that the little unit on the Lace sensor doesn't appear to be terribly exotic.

Also unknown is how good it is.

[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.