Challenging and exciting new design needs a guru !

[David Schwab]

Are you sure that the coil is in this orientation (opening parallel to the "body", that is, coil axis parallel to the body)?

Yes, it was a solenoid coil.

From the 1978 patent, you see that it's a long thin magnet in a thin tube mounted parallel underneath the strings.

Not obvious?

Imagine a single-string single-pole pickup.
Stretch it up several inches tall.
Now lay it on its side lengthwise under a string.

(sigh)

The magnet's entire length was covered in 10k-15k winds of #42 (most likely), but the magnet/coil ends, the last 1/2", produce the bulk of the signal because of the very polar magnetic field distribution.

It was easy to miss that because it's not what you were expecting. I wasn't either. It wasn't until I talked to Dan (salvarsan) that I realized the pickup was on its side!

[salvarsan]

You need to look at the Gittler patent, #4,079,652 -- easiest place to get a copy is Google Patents.

I manicured the Figure 10 pickup drawing slightly to emphasize the windings and magnets, a stack of 5 1/8" diameter by 3/4" long Alnico rods.

Item 75 is the bridge, 80 is a guitar string.


Figure 12 shows a skinny tube with flared ends which is the bobbin, also made from 304 stainless.
The flanges had to clear the tube's .25" ID.
The tube wall is .010" thick and makes for

• a .145" diameter bobbin core,
• a 3.73" core length, and
• a .247" diameter flange.

If you plug these figures into a coil estimator for a 1000Ω coil,
you get between 12k and 13k winds.

[Mike Sulzer]

Yes, it was a solenoid coil.

It was easy to miss that because it's not what you were expecting. I wasn't either. It wasn't until I talked to Dan (salvarsan) that I realized the pickup was on its side!

OK, I missed that. This would be easy to analyze by the variable reluctance method if the alnico were really high permeability, the strings were really thick and the gaps were smaller. But it is not so simple in reality. Too much flux leaks out along the magnets, and the strings do not do a good job of conducting the flux around.

So if you think of the magets as magnetizing the string at some angle (and the string probably modifies the angle somewhat), then if you mentally draw the loops of the field of the string, you can see the flux through the coil in the direction of the axis of the coil does indeed change some when the string vibrates.

This still does not seem like a very efficient pickup. I think it would be better if you used a rod magnet on each end of a steel rod (to better confine the flux), but I still doubt that it would do that well.

[salvarsan]

Here are some basic magnetic field plots for a 1/8" by 3/4" Alnico rod.
This one shows the polar distribution of the field strength.


This one shows field strengths for Alnico 5 and 8. Since the 1/8" diameter
is suboptimal for less coercive Alnico alloys, it made sense to model A8 and
see if the field strength improved.
It did.

Alnico 5, Alnico 8, and the field direction are plotted here.


The next question is, instead of a Gittler magnet,
i.e., 1/8" by 3.75" stack of alnico rods, what happens
if we use two A8 3/4" rods separated by a 2-1/4" length of soft steel?

[Dave Kerr]

Dave Kerr,
The pickups are oriented parallel to the strings, not pointing up but yes, as long as there are two coils to each string with the N (or S) polarities pointing at each other in the middle, it might just work...

I was thinking a large number of mini bobbins (small neo cores?), oriented conventionally and daisy-chained in series along the full length of the tube

[David King]

I really don't think we want to sense more than one or two narrow apertures if we can help it. A 3.75" wide aperture is going to cancel so much treble and midrange that it's going to sound like it's coming out the end of a very long pipe down into the earth. The low end will have absolutely no coherency at all.

[Rick Turner]

Who needs coherency when you look like that?

Piezos...

Check with Richard McLish at RMC. They'd work great on a Gittler.

Take the look of those old pickup tubes and turn them into tuning fork-like vibrating reeds tuned to flats and sharps...

Put springs in 'em and turn 'em into spring reverbs...

[salvarsan]

Rick,
It may be just a feeling, but I'm beginning to suspect that you think the Gittler
pickups are completely resistant to improvement, and eye-crossingly goofy, too.

[Rick Turner]

Who, me?

I think the guitars are amazing, though uncomfortable, but the pickups...well, there is a challenge to being given the packaging first and being told to make something work well in that visual form!

If I were a monied collector, I'd have one of the original Gittlers in my guitar collection. I think they are a truly brilliant minimalist design...worth of any Museum of Modern Art you might mention. In a way they are more like the Chapman Stick than any other guitar-like instrument. They were there before Steinberger, and of contemporary guitar designers, I can only think of Teuffel or Pagelli in Europe for equally brilliant modern design.

I'm actually a real fan of the Gittler design, but I'm just having some fun here re. the pickups, though I know from experience that piezos work just great on this kind of guitar. I just don't want to make them.

Get the ergonomics right and these are potentially really unique and wonderful guitars...for some folks. Niche market. Like Steinberger...

[LtKojak]

Get the ergonomics right and these are potentially really unique and wonderful guitars...for some folks. Niche market. Like Steinberger...

Or the Chapman Stick.

Didn't I say this before in this thread...?

[salvarsan]

In praise of goofiness

Who needs coherency when you look like that?

If you play a Gittler, it's not enough to look goofy,
you must also be able to sound goofy.

That means a synthesizer and a signal that is facilely sampled by MIDI, which means
6 pickups having minimal crosstalk.

let's assume that an improved hum-cancelling pickup is possible for the Gittler.

Because of crosstalk from conducted vibration in a stainless steel guitar,
a pickup potting requirement is absolute.

One design: a small two-blade geometry like the Roland and Gibson hex sensors.

The 2-blade pickup would be small and with a squared aspect ratio.
The square aspect that fits in a circle is .707 * ID, or .176" at the corners for a .25" bore.

That's an onerous size constraint.

If the 304 stainless steel (3/8" OD, 1/4" ID) was replaced with an alloy at least 4x stronger,
(which is easy enough to find, these days) then the tube wall thickness (1/16") could be cut in half.
NB, there are nickel-chromium and copper-chromium alloys that fit the bill, but you can go crazy and get superalloys, too.

A 1/32" wall thickness mean a 5/16" ID = .3125".
The hypothetical pickup is then .22" at the corners, less onerous, but tolerable.

The magnets would necessarily be neodymium since Alnico that small won't hold magnetization.

[bbsailor]

If you play a Gittler, it's not enough to look goofy,
you must also be able to sound goofy.

That means a synthesizer and a signal that is facilely sampled by MIDI, which means
6 pickups having minimal crosstalk.

let's assume that an improved hum-cancelling pickup is possible for the Gittler.

Because of crosstalk from conducted vibration in a stainless steel guitar,
a pickup potting requirement is absolute.

One design: a small two-blade geometry like the Roland and Gibson hex sensors.

The 2-blade pickup would be small and with a squared aspect ratio.
The square aspect that fits in a circle is .707 * ID, or .176" at the corners for a .25" bore.

That's an onerous size constraint.

If the 304 stainless steel (3/8" OD, 1/4" ID) was replaced with an alloy at least 4x stronger,
(which is easy enough to find, these days) then the tube wall thickness (1/16") could be cut in half.
NB, there are nickel-chromium and copper-chromium alloys that fit the bill, but you can go crazy and get superalloys, too.

A 1/32" wall thickness mean a 5/16" ID = .3125".
The hypothetical pickup is then .22" at the corners, less onerous, but tolerable.

The magnets would necessarily be neodymium since Alnico that small won't hold magnetization.

Dan and all

The Gittler is a radical guitar and requires a radical pickup design approach.

Think moving coil pickup in some of my earlier threads.

Think diametrically magnetized rod magnets. K&J Magnetics - Products You can use single or multiple magnets per string to obtain the best balance of harmonics from each string.

Think individual low to high Z transformers for each string for almost 100% isolation between adjacent strings. Impedance ratios and string loading could be optimized for the impedance and frequency range of each string along with favoring or attenuating various harmonic ranges.

Think pickup tube alloys that have minimal impact on magnetic transparancy.

Think active gain adjustment for each string.

Think low impedance ground return using the neck support rod. Then you could have a single loop from each string or the common ground return going through a 5,000 turn TA23A11 or TA23B11 from www.yuanxing.net or from the hot isolated end of the string being looped through from one to a few times to balance and mix the output from each string.

Think of measuring the current going through each string by using a Tektronics current probe.

I have discussed most of the above ideas on this forum before.

Maybe the time has come to more fully explore moving coil pickups.

Let the fun begin!

Joseph Rogowski

[big_teee]

I don't guess I get it!
Why do you keep saying it has to fit in a 1/4" tube?
I thought the whole idea was to come up with a new improved P/U.
I would throw the 1/4 out the window and comes up with something that works, then make a suitable cover to go with it. But then again I don't get the whole concept to start with.
Later,
Terry

[salvarsan]

Thanks, Joe.

I was hoping you'd join in since I wasn't entirely clear on the moving coil concept.
The current mode pickups are easy to prototype, too.

[David King]

So as I understand Joe's "moving coil" pickup the string and neck form the single turn coil which swings wildly over some fixed magnets? The signals are isolated because there can be no crosstalk if the bridge electrically isolates the hot end of each string. The hot end is fed into a 1:5000 turns transformer and the output goes on to a hexaphonic midi thingy like a Roland GR20? Cuwel