Arcs 'n' degrees

[Bent]

Richard wrote:
In attempting to achieve a semblance of "linearity", (though it's definitely NOT linear), it follows that the closer one can get to rotation along the tangent of the arc - about the 0° mark - the lesser the force required to move the crank. It is under this condition where the vertical moment becomes truly "miniscule" and is of no consequence. Having said that, I must reiterate that - in the simplest of circumstances - there are four "moments" to be considered in the train from the pedal to the changer. The first is at the pedal itself, the second at the lever (or arm) attached to the crossroad, the third at the bellcrank and the fourth at the changer finger. If everything is close to being "linear" then the sum of the moments is at the minimum to, say, raise the pitch of the string. If, however, things are really cockamamie then the sum of the moments can be considerable. Let me qualify the term "considerable" - it will most likely manifest itself solely in an increase of the required force necessary to activate the change. For some, an increase in, say, pedal pressure is a desired personal preference. For myself, I prefer things to have the resistance of greased owl-crap. If I could use brain waves to activate a pedal, then I'd be happy. For others, not so.

Needless to say, the required force can be altered (reduced) by changing the position of the pullrod at the bellcrank and/or the changer finger but, inevitably, at the expense of greater pedal travel. We do this all of the time when attempting to have things start and end at the same time.

Can anyone say "compromise"?
I have no doubt that you, Bent, make every effort to optimize everything on your instruments and is evidenced in the final product. Why not, then, optimize the pulls - the minimum moments - when setting up the instrument? If it's "as good as it gets" then no one has a quarrel.

Richard

As usual, this thick skull is getting less absorbent with time...
If we could just concentrate on the bell crank/rod movement for now. I realize that every joint and movements count. But let's just discuss this part ..less confusing for me that way.

I find it easier to visualize things if I think HUGE - way exaggerated. So say we have this 2 foot long bell crank that will be pulling on the rod from -45 deg to 90 deg to +45deg. At the start of the pull, the severe angle makes for a hard pull as well as a long stroke pull, in that some of the travel of the bell crank is devoted to making the rod swing in an arc (non-linear) BUT this non-linear pull gets gradually better as it approaches 90 deg.Then, as the pull continues over toward the +45 deg it gets progressively worse again. (lets call this YOUR way)

With "MY way" the pull starts off with the same disadvantageous position as your way, but it gets gradually better , to arrive at max pulling moment at the end of the pull.
Maybe we are saying the same thing and I am just not grasping things?
Convince me some more my friend.
YES, I try to optimize at every stage on the BenRom's. Where's the fun in doing anything less? That's why, when I read your stuff on arcs etc, I got on it right away... Only the best of arcs on the BenRom!

As an aside, but very much related I think...Is it best to have the rod holes in the bell crank lined up with the center line of the cross rod? Instead of staggering a bunch of holes more or less in line with the edges of the cross rod? (cross rod is .375 square) I am making new bell cranks to line the holes up with the center line... My friend Martin Weenick suggested I do this. Martin is a great thinker and builder....

[Georg]

Is it best to have the rod holes in the bell crank lined up with the center line of the cross rod?

For optimal "timing" I would think one line of holes is best, but you may have to produce two or more interchangeable bellcranks with holes slightly offset along that center-line to "fill the holes" for perfect "timing".
Seem to remember you told me that such interchangeable bellcranks can be found on Fulawka PSGs..?

I have contemplated having bellcrank axles where the center part with bellcranks can be adjusted off-center with the pivot points at both ends ... enough to bridge the distance between two holes and not necessarily same offset for two bellcranks on same axle. That would allow for perfect timing with fewer, in-line, holes in the bellcranks. More, and more complex, parts to make a stable axle, but hitting right every time with one type/variant of bellcranks becomes easier. Should also be easier to put in / take out such a 3-part axle in a populated undercarriage.

[Georg]

For myself, I prefer things to have the resistance of greased owl-crap. If I could use brain waves to activate a pedal, then I'd be happy.

I have thought of using brain waves, but have come to the conclusion that it will be better, and more practical, to use the position of the thumb-pick over a number of frets introduce extended changes on one or more low strings - via servos. Can probably cover +/- 5 frets (+/- 5 half-tones) with reasonable precision.

As for ease of pulls... I will attach pull-rods with springs to the fingers on my modded Dekley to take up a part of the string tension on some strings. That will ease both raises and lowers since the lower-return springs can be eased back a bit when string tension gets taken up elsewhere.

[Dave-M]

Bent and Richard: Forgive my butting into your discussions, but would this help?

In the extreme case of moving the bell crank over a 90 degree rotational path, Bent’s way would start at the top vertically, and never move if the crank pull was precisely 90 degrees from horizontal.

If slightly off toward the side of the pull it would require tremendous force at first, but diminish to minimum at the end of the travel. There would be a slight bit of extra force to go the few degrees of "off-centeredness" past minimum.

With Richard’s way you would start with mediocre force, pass though minimum, and wind up with a mediocre force.

It boils down to two "mediocres" being less than one "tremendous".

[richard37066]

Dave-M -

Bless you! A clear and succinct description!

Bent -

Once again, I believe you (and Martin Weenick) to be 100% correct. To my way of thinking, the holes in the bellcrank should, indeed, be in line with the crossrod. This affords the opportunity of applying pressure TANGENTIALLY to the arc through which it moves. However - as you and Georg have pointed out - and know - when it comes to a timing problem there just isn't enough room to stuff a bunch of holes in a line on that bellcrank thus having to consider resorting to a Blanton-type system.

When Georg visited me a couple of weeks ago, I showed him preliminary drawings of a bellcrank of my design which not only affords the fine adjustment of the Blanton but also the opportunity to pass an additional pullrod - or two - THROUGH the structure. In addition, the width of the bellcrank matches that of the changer fingers thus allowing two such cranks to be stacked side-by-side and inline with adjacent fingers. My, oh my, but it's elegant! But it has one huge damnable problem! It is composed of three pieces - two of which are quite small. Read: - LOTS of machining time, tapping small holes, etc.. For someone like myself who is interested solely in a one-off for my own pleasure then that is of little consideration. For another, like you, who is constructing machines for sale, the added work involved just might be very prohibitive. You'd have to jack up the price in order to cover your time. I could probably email that cursory drawing to you as an attachment. You'll no doubt, then, roll your eyes and declare that it would be out of the question - for YOUR purposes. But that's quite OK since I fully understand your need to keep expenses to a minimum - within reason.

I greatly admire much of Georg's thinking but, in doing so, he makes me mindful of the fact that - although simplicity is beauty - there are times when simplicity has to be sacrificed in favor of attaining a degree of precision. It's pretty much a forgone conclusion that a slightly greater degree of precision is required as regards the garden-variety bellcrank. Maybe - if the members beat this over the head enough - a reasonable solution will be arrived at which satisfies not only adjustability but, also, economics.

I have neither made a mathematical nor an experimental effort at quantifying the forces involved in this discussion regarding the bellcrank. I do, however, stand upon firm ground in my assertion that, when all forces are essentially tangential to the arc through which such "levers" move then the forces will be minimal. I do not discount the notion that, in the final analysis, minor misalignments do not amount to a hill o' beans. But begging the question, since the ability to make a myriad of adjustments are readily at hand, then why not just turn a screw or two and put things on the money? Can't ask for more than that!

Your integrity is above reproach. Do not EVER abandon it.

Richard

[Georg]

It's pretty much a forgone conclusion that a slightly greater degree of precision is required as regards the garden-variety bellcrank.

How much precision is "high enough precision" here? How much of a timing deviation between octave-tuned strings on same pedal/lever does it have to be to be "too much" - especially when playing music?
I think we have to quantify some common goals one way or another, regardless of whether we want, and can afford, to machine the parts for "absolute precision" or not.

My pragmatic view...
On my "modded monster" with extended E Major, I have 3 levers and one pedal that raise/lower 2 sets of 3 octave-tuned strings. With free-wheeling bellcranks as "gears" on the lowest-octave strings it's "a recipe for bad timing", probably up to 10 cents timing-deviation on some changes. In practice however the timing-deviation is barely audible at the absolute slowest changes - changing slower than one will ever do while playing music.

If anything, balancing splits across octave-strings may throw achieved timing-perfection off. I have found it impossible to make 3 splits kick in at the exact same point of travel. In practice though split-balancing only has to be "close enough" - easier to achieve with split-rods than with splits in changer IMO - for it to become inaudible while playing music.

For completeness...
Inaudible timing-deviation to me is "next to zero beating" along the entire travel, which one doesn't need a particularly good ear to hear if the sound-chain delivers true lows with all filters/effects off.

[Bent]

Richard,
Really we are beating this thing to death: Me anyway, with my exaggerated arcs and 2 foot bell cranks.
At the outset I think you mentioned 5 degree arc for a typical pull, no? That, really is as close to linear as needed right there. So whether or not we start a bit below tdc and swing beyond 90, or start a bit further below tdc and end up at 90, isn't gonna , like you say, amount to a hill of beans. We have to look at the final result, and at the things that can steal precious pull-length, like play in joints, play in the holes for the pull rods, abnormal amount of dead play needed at the rod/changer plate connection. In other words, the whole pulling train has to be taken into consideration - all it's components - I think we will end up thinking that the arcs and degrees that we have been batting at the last 2 days, are one of the least worrisome. We will end up with a good pull doing it both your way and my way.

I also want to thank Dave for his input and clear thinking

Richard, congrats on your Blanton type crank, my absolute favorite of bell cranks. What a great asset it would be to have Jerry on this forum with his forward thinking and just plain good smarts! Sadly he is not into computers...
I hope it works out for you. Yes I would be interested in seeing a clear drawing of it with all the exact measurements. I would even make you a prototype and send back to you, as time allows of course.

bcrnk.jpg (21.29 KiB) Viewed 2074 times

As for me, I haven't sat down to tackle my version of the Blanton, yet. In lieu of that, I have just made what we will call a linear bell crank with the 11 holes in a straight in line with the center line of the cross shaft. The holes in my raise and lower bars are spaced at .295". The holes in the crank are spaced at half that amount, making one hole in between for timing of pulls etc. This should be sufficient. I know - Blanton is better!
The material may seem thin and flimsy..the arm with the holes is only 3/8 wide and 3/16 thick. But it proves to be sufficient. I hooked my digital scales up to the hole farthest away from the shaft and applied 50lb pull (23 kg)..nothing bent, broke or deformed in any way. This crank is a few grams lighter than my other ones and that's good too.
I am almost 100% happy with it - now I just need to make up some thing or some idea where I can mass produce

[Bent]

It boils down to two "mediocres" being less than one "tremendous".

Or - the way I see it; two mediocres being the same as one tremendous, no?

[Dave-M]

Bent:
With the thin material, could you stamp and form your arm for mass production? The die could also pilot the holes for tapping. Possibly stainless steel would not fracture at the bends as much as aluminum, and could be thinner.
Could use nut and bolt instead of tapping.

[Bent]

Dave,I have toyed with that idea. It makes perfect sense. But the thing is, I would need to make a die or have one made. That's expensive, as well as the fact that I would need some sort of a press or brake. I have been thinking about having a frame made and use a hydraulic jack.
..or get one of those old things they called fly press But those are scarce and pricey as well.
To have a stamping shop do it would cost upwards of $10000 for the die and the first run.
I should try the jack thingie... hopefully I could make a die on the mill.
Nut and bolt I am afraid would be too cumbersome..not enough room. Tapping is ok though.