Project: keyhead changer on knife-edges

[Georg]

Have finally drawn up main parts for my keyhead changer on knife-edges.



Raise and lower made of steel plate, U-shaped, with raise inside lower. Knife-edges reinforced with shaped steel plates "welded" with heat-cured epoxy inside the U-shaped raise and lower parts.

Acts like a regular pair of scissors. For raise it is forcing the keyhead slide for each key away from the edge fixed to the neck-bearing, and for lower it lets the slide be pulled by string-tension.

Lower-return is (to be) achieved via a push-rod in the "lone" upper lower position, and I have several reasons for this solution.
1: no "reverse-effect" if/when lower-rod is pulled far.
2: I can use a long spring for constant return-force since the spring will be hooked up at the bottom of freewheeling bell-cranks 10-15 cm from the changer.
3: by pushing at the front-end of the changer-finger, instead of at the back-end, the changer-fingers will naturally line up straight and not tend to "wiggle". The fixed line-up of this push-rod between changer and bell-crank, will also will also steady the changer.
4: this arrangement gives me an extra lower since I can pull anywhere on the freewheeling bell-cranks. Much needed extra for my JI-tuned pedal combinations.
Only potential drawback for this lower-return variant is that knife-edge will lose contact if spring-push is adjusted too high.


Drawing does not show exact positions for knife edges and shapes of parts, as I have to test on a one-string prototype to get the +/- 4 millimeter slide I need before raise and lower lock into each other and neutralize further change - a design-feature.

Also have to make sure I can remove the raise from its place inside lower for maintenance, without risking "self-disassembly" while playing or replacing strings.

All knife edges will be made as sharp as a blunt knife, so I have to make sure I get the material and angles right for edges and seats.

That's all I have drawn so far ... the rest is in my head.

[richard37066]

Georg -

Nice!

Shades of Fields' PS210. Yer a man after my own heart.

Couldn't help but chuckle at the raise/lower tab arrangement. Essentially the same as on my drawings.

Work on the refinements and keep us posted. Forge ahead, my friend!

Richard

[Georg]

Richard, thanks.

Lots of refinement necessary, as can be seen on this "action-drawing"...



When I get all 3 knife edge points positioned and angled correctly, the around 7 degree slope for the sliders should stay pretty stable through full raise/lower travel.

The upwards pull will be high because of the vertical tuning-pegs I want on top of the sliders, which I have to angle the knife edges to take up. Don't know how high the upwards pull will be since I haven't designed sliders with tuners on yet.

[richard37066]

Georg -

I fully understand your predicament. I had arranged things such that the pull on the string is tangential to an arc drawn from the raise pivot point. The bugaboo is that the lower finger - when activated - lowers that raise pivot point thus changing the angle of the string (and the arc of the raise finger) as it's relaxed. What started out as a semblance of linearity along the arc went out of the window in a heck of a big hurry.

I'm still battling it. Should I come up with something with better geometry I'll pass it along to you for whatever it may be worth.

Oh - what tends to bother me is that Fields' original design seemed to work quite well. This begs the question as to whether I should really be so concerned about the aspect of linearity or to just let it fly as is. "As is" will work. My gut tells me that I'd like something a little better.

Richard

[Georg]

Richard,

Fields' design only has to counter a very minimal upwards force, and edges are angled for it. Also only 2 edges in his design since the string is hooked up to one half of the scissors.

I have a choice of reducing the upwards force by adding more parts, or utilize the full upwards force to stabilize the changer. I am going for the latter, and will probably end up with around 45 degree angled knife edges / seats, with around 80 degree seat openings, for optimal stability and movement.

[richard37066]

Georg -

Isn't your "upward force" a function of the angle of the "keyhead slider"? Isn't this analagous to the breakover angle of a string across a bridge? From calculations that I have made earlier - and posted on the forum - with a 7° breakover angle and string tension in the vicinity of 26-27 lbs, the upward force was in the neighborhood of 3-4 pounds. It would appear that, even with the inclusion of the "keyhead slider" arrangement, the attendant forces would be sufficient to hold everything in place while the upward force is sufficiently low so as not to attempt to pull things out of vertical alignment. And yes, shallower knife angles and their seats would help to mitigate this concern.

I may be speaking out of turn here since I do not have the full details of your design but, from measurements made by myself and others, a double raise or, even, a triple lower is readily accomplished with a displacement of less than 1/16" at the changer and given a 24" scale length. In spite of this, I still agonize over the "linearity" of things even though I'm not certain that it's even necessary.

Oh. I must thank you. As a result of your design, a small light bulb went on in this old head so I revisited my design and am coming much, much closer to linear pulls. I must admit that, in order to achieve greater accuracy, I've incorporated a small portion of your design into the newer configuration. Hope that you won't mind my pilfering a part of your idea. Will post the results with appropriate explanations when I'm satisfied with the thing.

Richard

[Georg]

Richard,

If you find something useful amongst my ideas, use it whichever way you like. That's what its released on the builders' forum for.


The 7 degree slope-angle in itself isn't causing problems with vertical pull-force. I want some upwards pull-force to keep the changer's knife edges tight.

It is the vertical tuning-pegs that will have to be raised some 1 1/2 - 2 inches above, on top of each slider, to get the sideways tuners accessible/tunable, that cause extra pull-force as the string-tension will lift the sliders' back-end where the changer is. The changer will take up and transfer this vertical pull to the keyhead-bed through the knife edges.

I can hold the sliders down with "packman rollers" at the back-end with no added friction, but that means there literally won't be any vertical pull-force left to pull the changer tight. Knife edges must be forced tight, as otherwise they must be extremely precise made (microns) and no wear will be acceptable. As long as there's "a little" force on them, quite a few microns imperfection or wear won't matter the slightest, but if the force is too high the pressure will make them too blunt too quickly and/or they will cut into the seat. It is a "balance" (in more than one way).

I will use hardened steel for knife edges and seats, but since I want them to "last for life" with hardly any lubing and no other maintenance, I want to get those forces under control during the design phase. While a couple of pounds is fine, it looks like I may get 2-3 times more on third string which may be a bit too much in the long run.


I can put the tuners the normal way - pegs inwards, which lowers the tuners and reduce the vertical pull-force considerably, but if so I have to split the changer in two halves - for 5 + 5 strings. This may not be such a bad idea as it will open up more space for the vertical "backbone" centered under neck and keyhead. My PSG won't have a regular "body", so that single "backbone" along with the neck/keyhead-bed will have to take up overall stress and contribute all "body-tone".

Before designing with the tuners the normal way, I have to make sure the pegs can't interfere with neighboring strings since the strings will follow the slope the tuners are mounted along. Getting the right tuners with short pegs and made for no more than 6 millimeter deep holes, is imperative for perfect line-up of horizontal pegs. Exact tuner-dimensions don't matter much for vertical pegs.

[richard37066]

Georg -

I admire your skill and tenacity. I'm really beginning to understand your concept but (with tongue in cheek) have you ever heard of a blivet? It's said to be 10 pounds of you-know-what in a 5 pound bag!

I might note that whenever I write such as is to follow, it is not intended to be critical or to exhibit any brilliance but, rather to simply throw some things up into the air so as to better understand.

A thought or two. First, would you define the phrase "reverse effect"? I'm not certain of its' meaning. Does this allude to the string tension being the primary or sole force in returning things to a "normal" state?

Secondly - knife edges are inherently self-centering. That is, any horizontal pressure component will force the edge and seat together into its' designed mating position. Since the "reverse return" rod is in such close proximity to the primary edge/seat it - as you have noted - will tend to counterract the forces of string tension. There's not much mechanical advantage with such a short lever hence - as you've noted, once again - there might be a considerable force exerted by the "return rod". If - and I say this advisedly - If the "reverse return" effect is as I've suspected above, then can it be countered, instead, in favor of a tab at the bottom of the lower finger and a spring placed in the "normal" position to a point to the left of the "neutral" position? Yes, I know - that's the way that "everybody" does it but it just might eliminate some hardware. Unencumbered string tension at one end and a typical return spring at the other should lock things into place.

If I've erroneously assumed something, then don't hesitate to set the record straight.

You certainly do some innovative thinking, my friend. I'm confident that the moments will be worked out such that your combination will operate effortlessly.

Please keep us all advised of any progress.

Respectfully,

Richard

[Bent]

OK, these days my head is full of conventional, boring AP changers and retrofitting two of them to 2 existing guitars with one steel residing in England. So my mind might be other places.
-I am not ashamed to admit my ignorance in saying that I don't understand much of Georg's changer idea, other than the fact that the contact points pivot on knife edges. It is really hard to derive anything further from the one dimensional drawings. I do hope more drawings are coming with views from all sides.
What is the "keyhead slider" thing? And what does it do?
-The raise/lower plates must have an axle of some sort, no? All I see now is the stop at the bottom and the knife edge pivots on top I can see where the string tension holds everything in place but what happens when the string is off?
-Where does the tuning key and string go?

My questions surely must show how dense I am at this concept...
Richard, you seem to understand these plans to the fullest and you are also good at explaining things. So maybe together with Georg, the two of you could attempt to enlighten this old brain?

[mac639]

Hi guys....

I get the idea sort of for this changer. Very clever. I'm like Bent though, can't understand this keyhead slider thing. Is it like a keyless string attachment part?. And along the top will be a sliding piece with a screw adjustment or something. If that's the case I can see this working quite well.....otherwise I'm still confused I guess
Mac