Project: keyhead changer on knife-edges

[richard37066]

Georg -

Observing the evolution of your creation and learning the variables is fascinating. Don't be shy about posting little tidbits as opposed to a lengthy explanation. Can't wait for the next installment.

Richard

[Georg]

Richard,

I have never understood the meaning of the word "shy" in this context
It is just a matter of finding time...


Right now I am pondering how to construct pull-rods with built-in stretch for overload, or - alternatively - construct spring-loaded bellcranks.
I want a stopper on each pull-rod inside the keyhead changer frame, as limit for maximum raise/lower pull. This means something has to give somewhere else along the pull-train, if/when a pedal/lever pulls further.

The idea is to allow for a light and moderately flexible undercarriage for my somewhat top-heavy PSG, with no adverse effect on tuning-stability. The method will also provide me with relative simple and perfectly tunable two-pitch raises/lowers to meet my JI tuning requirements.


Later - when finished designing the entire keyhead changer - I am going to study how to make the non-bearing body-covers act as tuned resonators for increased body-tone "on demand", and go from audible resonance all the way to perfectly quiet with the movement of a lever and/or push of my arm. Will add another sound-dimension to the humble PSG.

Oh, and I want to isolate the strings electrically at the bridge-end, so I can use the individual strings as PU-coils when playing with non-conducting bars ... just to mention one of the alternative ways I want to extract sound for amplification.

[Georg]

How parts are supposed to fit together...

Have updated the drawing a couple of times.

[richard37066]

Georg -

Really like the "Pacman" rollers. Any concern about any of them slowly migrating to a stop on one side or the other? Is that screw just to the left of the "nut" for individual height adjustment as opposed to grooving them for constant string height?

Richard

[Georg]

Richard,

I expect the Pacman rollers to start out by banging into a limit every time I change strings, and now and then in between. Doesn't cause real problems since they will self-adjust - strings or sliders will slip - till rollers are free between largest raise/lower.

I will probably groove the rollers - all identical - to increase friction and make strings stay centered, but I'll use the (left) screw for hight-adjustment.

Also, no matter how I tune, all my 5 PSGs are slightly "off" open - in my ears. The cause seem to be the bending-offset for strings with various thickness/core/winding/tension (which we have discussed before), which is why I always fine-tune my PSGs with the bar.
A "paper-width" or so out of alignment at the nut is enough to fix such offsets for one set of strings. So I have calculated in slight lengthwise counter-offsets of individual rollers with the right screws holding the steel-pieces with knife edges fixed to the neck's "backbone".

[Georg]

A rough sketch of the entire keyhead changer / neck construction.

Text and details are invisible on that downscaled image, so those interested can see the full-size sketch on my site.

All can see that I haven't bothered with proportions / measurements. I don't need exact measurements at this stage. Since I don't have a top-plate/soundboard or other conventional body-parts that take up space, I will try to make the finished PSG slightly on the thin side (less height).

[Georg]

Looking through the layer-cake from above...

...and a full-size sketch is available on my site.

Note that the keyhead will, as you can see even on the resized drawing above, look pretty "normal". It will of course be a bit wider and longer than most keyheads to allow for independently moving sliders / tuning gears and parallel strings.
- Have not decided whether to leave the changer scissors partly visible from above, or to let the sliders cover them entirely.
- Have decided to color all but the tuning gears, rollers and bridge dark

Note also the "same width along the entire neck", as I have the depth necessary to enforce the neck under the PU cavity.
- The intended neck-width will be pretty much equal to a 10 strings Alumitone PU. I will prepare the PU mount for about 25 millimeter (1 inch) quick repositioning lengthwise as a form of tone control.
- The PU mount will kind of float below the neck itself, so I can quickly adjust it to be seated towards the neck, towards the neck's "backbone", or floating in between. I have tested the response-differences and tone characteristics for these PU mount alternatives, and would not be without any of them in this construction.

The "sympathetic vibration adjustment" screw near the bridge may (or may not) make someone a bit curious.
- The first is that without a connection between neck and "backbone" there, the "backbone" serves no purpose. However, if the aluminum neck is secured too tightly to the steel "backbone" these two parts will try to warp each other as these materials expand/contract. Thus, the connection must be horizontally loose and vertically quite tight.
- Under string-tension the aluminum neck will bend upwards a fraction of a millimeter almost no matter how thick/strong the neck is, and despite the neck looking thick on my drawing it will intentionally be made thin enough to bend a little. By tightening the two screws holding it down to the "backbone", the neck can only bend so much.
- By tightening those two screws exactly - with a minute vertical gap, and add two screws that almost close the gap from below, the aluminum neck will vibrate in a very narrow gap when excited by vibrating strings. Adjusting to just the right gap-size will make the neck hammer against the "backbone" when vibrating, and understanding that I guess you all can imagine the effect I will try to tune that gap for.

[Bent]

Ok Georg, time for me to attempt an understanding of this. By nature, those flat drawings are tough for this brain to absorb. ...likely because I am not a drawing person by any stretch. More like a trial and error type of guy. So me saying this is not a reflection on your artistry, but on my own lack thereof.
Just a couple of things for now:
The yellow thing that is the slider..then you have the key attached to a raised part of that line. I take it that the raised part attached to the key slides along the long yellow line? If so, how? On a dovetail? On a square channel?
Then in the bottom pic where we see all the white round dots: Are those bell cranks seen from above?

[Georg]

Bent,

My drawings lack finer details mainly because an image-manipulation program like Gimp on a laptop is unsuited as design-tool ... even for rough sketches. I won't go into details about what have happened to all my other design-tools here on the farm while I was away, other than that it may cost me more than the price of a factory-new steel guitar to make it all operable again and doing so is not one of my priorities at the moment.

The yellow thing that is the slider..then you have the key attached to a raised part of that line. I take it that the raised part attached to the key slides along the long yellow line? If so, how? On a dovetail? On a square channel?

The raised yellow part with the key on, is a part of the long yellow slider on an about 7 degree angle.
The top of the changer scissor (light green) is hooked through a square hole in the end of the long slider. Tight fit and a real "hook", so slider can't slip off when at that angle. Knife-edge on slider will self-align on the knife-edge-seat on the 6 millimeter wide changer scissor top when string-tension is added.
Knife-edge and knife-edge-seat are steel, while slider itself is aluminum. Can't make that clear in my drawing at the scale I have used.
Have not drawn in guides necessary to keep the 10 sliders (for a 10-strings PSG) run friction-free horizontal along-side of each other, only the PacMan rollers that support them vertically. String-tension on tuning-pegs will pull sliders to one side, so need one horizontal roller for each slider in addition to the one shown.

Then in the bottom pic where we see all the white round dots: Are those bell cranks seen from above?

No, they are just indicating drilled-out holes in the horizontal bar of the "T" shaped steel "backbone" for the aluminum neck, to reduce weight. Pattern doesn't mean anything, as I will have to test for sonic properties with the actual profile mounted before deciding what can be taken out or not - it is all about sound.



When it comes to bell-cranks, axles, pedals, levers, etc. you will feel right at home, I think, as all this will be mounted in/onto an open frame with bracings. You can imagine that frame being like a Dekley-frame on legs but without neck and top-plate, with the entire mechanics accessible from above as well as below.

Difference is that my frame won't have to hold string-tension or prevent body-drop, so I will have it made as light as technically possible. The frame only have to be rigid enough to prevent leg-spreading, so I imagine using welded square aluminum "pipes" in the shape of a short, narrow, ladder laid flat, with 4 round pipes welded in the corners to insert legs into (my thanks to Mac for the idea of using pipes for leg-mount).

Everything you see in my drawings so far - neck with keyhead/changer - will be slid partly down into that ladder-like frame from the top, once all the pedaling-mechanics - bell-cranks, axles, pedals, levers, etc. - is mounted in the open frame.

Neck with keyhead/changer can be seen as one integrated piece, and will be secured to the frame with a few of the smallest vibration-dampening couplings for electric motors I can find. To make the connection stable for the horizontal pull of pull-rods, I envision the use of nylon-tipped screws - somewhat like those you use as pedal/lever stoppers - to clamp the changer-end horizontally (thus also vertically) to the end of the frame.


At this stage there will be no covers, so the instrument can be rodded, tuned and played-in with good access to all the mechanics from above alongside the neck and below, as well as from all sides. It will now be a complete and fully playable pedal steel guitar, looking like an aluminum/steel skeleton.


Front and back covers will be shaped, bent and braced to cover the top and sides of the frame/mechanics without touching anything, and be screwed onto the underside of the neck at a few points. These covers will then take up vibration from the strings, but very little noise from the inside mechanics because of the vibration-dampening couplings between neck and frame.

The covers will be something like an inch or so longer than the "skeleton", and will only partly cover the ends. At the keyhead end the nylon tuners must be accessible, and at the bridge-end it will be easier to hook up strings when there is no cover or end-plate.

I envision these covers to be shaped aluminum plate with a thick layer of plastic-paint, or with thin wood-veneer, on the visible surfaces. Will have to experiment a bit with these covers - both materials and their bracings, since they will have sonic properties and affect the instrument's overall sound - they add "body-tone" to an otherwise intentionally pretty "dead" construction.


Look plays "second fiddle" in my design, but as you know I have a few preferences. Will attempt to "un-polish" all visible metal parts and aim for a dark, matt, non-distracting overall appearance.

[Bent]

Thanks Georg,
Oh.. now a light is coming on...looking at the bottom drawing on your web site, magnified, shows 5 of those sliders on each side of the key head
I will re-read and try to digest. Down the line there will be more questions/requests for clarifications coming.