Low Impedance Pickup Research

[bea]

Just continuing the speculation a bit: if i chose the active way, i will run into trouble as soon as i manipulate the treble resonance of the pickup (which i intend to do...); the dummy loop should then act as a hiss generator... that's of course independent on the details of the implementation of the loop itself.

[Helmholtz]

Just continuing the speculation a bit: if i chose the active way, i will run into trouble as soon as i manipulate the treble resonance of the pickup (which i intend to do...); the dummy loop should then act as a hiss generator... that's of course independent on the details of the implementation of the loop itself.

Not sure if I understand. Can you show your circuit?

I would wire additional resonance shaping RC components in parallel to the series arrangement of PU+ secondary of CT (and not across the PU only). Will require smaller cap values because of increased inductance.
This way opposing noise EMFs can cancel "before" the filtering.

[bea]

The circuit does not yet exist. The idea was using an "arbitrary" compensation coil and feeding that with inverted phase to an OpAmp based adder stage and doing the fine adjustment through the input resistance of that branch of the adder. . Which will lead to incorrect compensation if the treble resonance of the pickup is modified, just like in the passive case when You do NOT nose cancel before filtering.

At the moment i am using one half of a preci splitcoil, quite surprisingly with almost no audible impact on the treble resonance.

[bea]

Concluding: the idea should work. It might, however, be useful to simply use more turns on the transformer, maybe 1500, and just one single loop. In conncetion with a High-Z pickup the impedance and also the inductance of the transformed loop will always remain uncritically small in comparison to the values of the pickup.

Feeding the singal of the loop into a separate input of an amplifier - as originially done by Alembic - is useful only if You do not modifiy the properties of the pickup, especially its treble resonance. In that case, the passive approach for attaching such a loop is preferrable.

Other topic: in this project i actually considered using a CT - based Low-Z pickup. I decided not to do. Reason: it is difficult to make them as narrow as even a tele neck pickup, and it would be less compact - just because of the size of the CT.

[nickceulemans]

If you are targeting feeding this pickup into a high impedance amp, you will need coils with 5,000 to 10,000 turns of very fine wire, assuming that there is sufficient magnetic coupling to transform the current into a useable voltage. Based on my experiments, I do not think this will work as there needs to be good magnetic coupling to get a good signal and to overcome noise that poor coupling would create.

Hi Bbsailor and Helmholtz,

Just wanted to let you guys know that I was able to get all the materials together to actually make the prototype as described two weeks ago... And, what do you know. It works! Kind of. I get some very confusing first measurements and results, though. Having the two rows of magnets and two sets of coils does not make it transparent as the configuration options seem to be infinite.

Fot this initial test I got two custom air coils measuring 2,2kOhm resistance each, about 8000 turns. Little 4mm neodymium magnets and a 4mm thick aluminium profile.

Some surprising results (to me at least):
- a single coil produces about the same amount of sound than combining 2. Either in parallel or series. I know sound is not linear, but still imagined there would be some noticeable difference.
- a single line of magnets produces the same sound as a double (opposing) lines of magnets, no noticeable hum difference either...
- a 500k pot has lots of hum in between the 3 and 7 setting. The hum goes away near 1 and 10 setting.
- the sound quality can be described as thin and pluncky and some combination of the coils does seem to allow for a slightly richer sound: A random combination produced a slightly deeper bottom end... (need more experimentation)
- the two coils combined read 4,4kOhm, either in parallel or series (?)
- the coils independent orientation/direction does not affect the results (?)

What I think I want to do, while continuing experimentation, is order 2x bigger coils, aiming for 8,8kOhm resistance combined. My thinking is it will potentially allow for more lower frequencies in the mix (?)

I'll upload some pictures if I get around to it this weekend.

Nick

[Helmholtz]

Just wanted to let you guys know that I was able to get all the materials together to actually make the prototype as described two weeks ago

Are you speaking of the design shown in post#393?

Some surprising results (to me at least):
- a single coil produces about the same amount of sound than combining 2. Either in parallel or series. I know sound is not linear, but still imagined there would be some noticeable difference.
- a single line of magnets produces the same sound as a double (opposing) lines of magnets, no noticeable hum difference either...
- a 500k pot has lots of hum in between the 3 and 7 setting. The hum goes away near 1 and 10 setting.
- the sound quality can be described as thin and pluncky and some combination of the coils does seem to allow for a slightly richer sound: A random combination produced a slightly deeper bottom end... (need more experimentation)
- the two coils combined read 4,4kOhm, either in parallel or series (?)
- the coils independent orientation/direction does not affect the results (?)

Seems that only one of your coils works properly.
Do you get about the same signal level using each of the coils separately (each in its designated position and using both rows of magnets)?

[bbsailor]

Hi Bbsailor and Helmholtz,

Just wanted to let you guys know that I was able to get all the materials together to actually make the prototype as described two weeks ago... And, what do you know. It works! Kind of. I get some very confusing first measurements and results, though. Having the two rows of magnets and two sets of coils does not make it transparent as the configuration options seem to be infinite.

Fot this initial test I got two custom air coils measuring 2,2kOhm resistance each, about 8000 turns. Little 4mm neodymium magnets and a 4mm thick aluminium profile.

Some surprising results (to me at least):
- a single coil produces about the same amount of sound than combining 2. Either in parallel or series. I know sound is not linear, but still imagined there would be some noticeable difference.
- a single line of magnets produces the same sound as a double (opposing) lines of magnets, no noticeable hum difference either...
- a 500k pot has lots of hum in between the 3 and 7 setting. The hum goes away near 1 and 10 setting.
- the sound quality can be described as thin and pluncky and some combination of the coils does seem to allow for a slightly richer sound: A random combination produced a slightly deeper bottom end... (need more experimentation)
- the two coils combined read 4,4kOhm, either in parallel or series (?)
- the coils independent orientation/direction does not affect the results (?)

What I think I want to do, while continuing experimentation, is order 2x bigger coils, aiming for 8,8kOhm resistance combined. My thinking is it will potentially allow for more lower frequencies in the mix (?)

I'll upload some pictures if I get around to it this weekend.

Nick

Nick,

Yes, please post a photo so we can see:
1. The thick metal string loop, identify it's total length, size and metal type
2. The magnets within the string loops, plus identify their polarity relative to the strings
3. The coupling of how the string loop is going through the transformer lamination or toroid core opening.

Are your measurements DC resistance or impedance from using an LCR meter such as from using an Extech 38093 LCR meter?

Good photos and answers will allow any of us to offer better recommendations.

Joseph J. Rogowski

[bea]

Then your loop area needs to be at least 4 times larger than the PU coil area (with 2 turns).

Recently i tried to run half of a split coil PU with 3850 kOhms DCR in series with the pickup. not full cancellation, but it worked reasonably well with only a small effect on the voicing of the pickup. I need that coil as a guitar pickup and will therefore test the idea i sketched previously. But i learned the the impedance of that coil is not THAT critical as long as it is not too large.

As i did not find my Talemas (wer Ordnung hält ist nur zu faul zum Suchen) i ordered new ones. AX-500, AX-1000 and AX-1500. I think i'll start with a single loop of 6mm² wire and an AX-1500 in the bass i showed previously and if that's successfull i'll do a single loop with an AX-1000 in another recent bass with the same pickup. That bass is a LP JR DC shape, and its pickup cavity is much larger.

[bea]

Success. Here some pictures - one turn, 6mm², Talema AX-1500, due to the small cross section of the loop. Sound slightly darker than without the loop, but that might easily be a feint caused by the missing hizz.

[bea]

Unfortunately not success at all in the other bass. Same pickup, different coil geometry.


The noise even increases. On the scope the noise signals from the loop and from the pickup seem completely unrelated. A very slight reduction can be achieved when i connect the loop to the ground. So slight that you must really believe in an improvement ;-)

Still almost no correlation on the scope (btw: surprisingly weak also in the 1st example).

How to improve this?
The wavelength of the noise should be a non issue, shouldn't it? Or do i have to take into account some kind of interference patterns with a spacial wavelength of 1-2 decimeters which could explain different patterns of the noise field at the two coils?

Is there any possibility to improve the directionality of the compensation coil? Maybe by adding a "ladder" of wire pieces with an orientation of the pickup coil to the compensation loop?

[Helmholtz]

The noise even increases.

In this case I would try to reverse the phase/polarity of the transformer.

The wavelength of the noise should be a non issue, shouldn't it?

No relevance as the wavelength at e.g. 100Hz is 3000 km.

Magnetic (interference) fields are polarized/directional and their orientation, phase and amplitude may vary strongly with position depending on source and distance of source. Far fields are easier to "buck" than near fields.

Maybe by adding a "ladder" of wire pieces with an orientation of the pickup coil to the compensation loop?

Not clear to me.


What are differences between the 2 basses and their circuits?

[David King]

Presumably the transformer coils are shielded but I wonder if they are also oriented alike? If at all possible I'd superimpose the sensing and dummy loops as closely as possible.
Can you listen to the two noise signals so see what is being picked up and then move the instrument around the room and in different orientations. I wonder if your dual trace scope is latching onto the same frequency range for each trace. Hard to explain two totally unrelated signals when one would think 50Hz would predominate.

[bea]

In this case I would try to reverse the phase/polarity of the transformer.

Which didn't have any effect in the 2nd case. But a strong one in the 1st.

No relevance as the wavelength at e.g. 100Hz is 3000 km.

I'm aware of that. The wavelength is irrelevant at any frequency of interest. That's why i specualted on interference patterns.

Magnetic (interference) fields are polarized/directional and their orientation, phase and amplitude may vary strongly with position depending on source and distance of source. Far fields are easier to "buck" than near fields.

The main sources of noise is the higher frequency noise of the supplies of neon tubes or (if i'm in my rehearsal room) the field of the large toroidal transformer, not the 50 or 100 Hz hum.

What are differences between the 2 basses and their circuits?

Electrically none (one has a vol pot, the other just a load resistance, both have the same parallel capacitor which on bass 2 is switchable.

The pickup positions are the same. The location of the electrics cavity are different: Sharkfin has it in the upper horn, and Primavera in the usual spot below the strings. Which leads to a difference of the orientation of the compensation loops relative to the PUs: roughly 45° for Sharkfin and roughly 90° for the Primavera.

So it is not too surprising that the hum cancellation of the Sharkfin works better than in the Primavera. What IS surprising to me that the signals from the loop and the PU in the Primavera are completely unrelated. The compensation loop produces "just noise", while the PU signal has a dominant frequency spectrum.

And the CTs are unshielded as well.

[Helmholtz]

Which didn't have any effect in the 2nd case.

What is the second case? Second bass?

or (if i'm in my rehearsal room) the field of the large toroidal transformer, not the 50 or 100 Hz hum.

Why wouldn't the large toroidals emit mainly 50Hz hum? ( Phase control dimmers only produce harmonics of 50Hz.) Fluorescent (not neon) tubes may be operated by HF SMPSs, though and thus would produce noise that is not related to mains frequency.

[QUOTE]Which leads to a difference of the orientation of the compensation loops relative to the PUs: roughly 45° for Sharkfin and roughly 90° for the Primavera./QUOTE]

How this? Aren't the compensation loops in planes parallel to the PU coil planes?

The compensation loop produces "just noise", while the PU signal has a dominant frequency spectrum.

This indicates a wiring problem.

That's why i specualted on interference patterns.

Interference patterns require at least 2 coherent sources.


Do you have an LCR meter?

Not quite clear from the picture: Is the controls' cavity shielded against electrical interference?

[bea]

The Sharkfin is Bass 1. Primavera is Bass 2 and looks like that:



last question first: the cavities and the channels are unshielded. Still unshielded. Will be done later when i make the active circuits. Anyway a guitar can and must be "sufficiently" silent without shielded cavities.

Strings are not grounded - grounding strings becomes useless as soon as soon as nylon tapewounds are used.

Refections (steel in the concrete of the nearby walln ?) can cause coherent sources.

But anyway - i am testing pretty close to computer, and there are many switching power supplies nearby. but there are also 3 lamps which can be clearly and uniquely identified as sources of electromagnetic fields. And moving the guitar in these fields has a strong impact on the noise. With or without compensation.

I checked the wiring, and the schemes are identical and correct. Also the DC resistance of the CTs are identical. The length of the primary wire varies ba about 20%.

LCR-Meter: one of my Multimeters has L and R ranges. At least large capacites seem to be measured well.

Angle: the planes of the coils are parallel to eacch other exept of course the windings in the CT. But the axis of maximal length are rotated in these planes.

(the cancellation effect of the problem loop was best when the plane of the compensaton loop and the planes of the pickup coils were at an angle of 45°. Which should not happen.)

For ever future step it is necessary to remove the coil from Primavera. Which i will do now.

EDIT: Measuring L gives 0.7 H for the loop, measured at the seconday ot the CT.
And something around 1.1-1.2H for the pickup. So the CT system should work, and the specs should be close enough for at least some humbucking.