I have placed the Excel Tool used (spin data, calcs, graphs) here if you want to take a look and test it out.
Appologies - the above is the first clockwise (CW) spin.
The link for the first Anticlockwise (ACW) Spin is below - its this one that is discussed in the posts above !
http://www.zohosheet.com/public.do?fid=53420
If anyone wants a walk though the model and its use just let me know.
Regards
gkd
I will try printing these out for consideration over the next 2 weeks.
Mike.
gkd & Co
I studied your work and wrote a detailed reply while travelling and ever since I got home I can’t find it! Tomorrow I’ll check the rubbish bin just in case.
Very sorry.
I did find this amongst my notes…
-
We need to first “qualify” every wheel we play…
That is - we need to measure the degree of tilt if any
The Rate of Rotor deceleration
The Rate of Ball deceleration
Anything else? -
We need a standard test to qualify wheels
-
We need standard terminology
That is - Drop Point
Impact Point
Angle of the Pocket Frets
Decay v Decelaration
Bounce & Scatter
Offset
Anything else?
More later
Mike.
Mike,
I hope you find the notes. I agree the standarisation of terminology would be useful, sounds like a good job for a new post topic.
I’ve ordered the 8 DVD pack to do some improved analysis on an unlilted wheel. I’ll let you known how that goes.
Hey, gdk, that looks like serious video analysis! Interesting.
Can just make one immediate comment on the data observed right now (it is an acquired reflex of mine to take a critical look at potential data quality issues before spending too much effort on analysing it):
Using your spread sheet data, I calculate
Rotation speed = 1 / (a point in time of one passage minus the point in time of the previous passage)
Ball speed decay rate (deceleration) = difference in two consequtive rotation speeds.
I am surprised to see that the ball speed decay rate, or deceleration, first increases towards a peak, and then falls off towards the end.
Physically, deceleration should be the greates in the beginning as the ball has the highest speed. The breaking force caused by friction and air resistance should be at the highest as the ball has the highest speed. The decelerating forces should never increase, but fall constantly (not necessarily linearly, but certainly decreasing).
In the anti clock wise spread sheet data this mysterical peak phenomena is very clear. Deceleration is 0.09 r/ss (deceleration being measured in the unit roations per seconds squared) between passage 2 and 3. Then increasing steadily and substantially until reaching 0.16 r/ss between passage 6 and 7, after which it decreases to 0.10 on the following passage and ends at 0.08 at the 11th passage.
The clock wise spread sheet data shows a similar pattern. The deviation is at the beginning as the deceleration starting at 0.13 r/s*s falls to 0.08 two rotations later. Then it develops almost exactly as in the anti clock wise data, increasing to 0.16 during the next 4 passages, after which it first falls very quickly and then falls slower until ending at 0.08.
How come?
There can be no physical explanation. Tilt indeed makes the deceleration increase and decrease (sometimes maybe even turning inte an acceleration). But that happens within one rotaion without affecting the average deceleration from one rotation to another, as in these spread sheet data.
If it is a matter of measurement errors, then it seems to be serious. It is not only substantial (doubling the deceleration rate when it should actually be falling), but it also seems to be systematic.
I am curious to know if you could verify this strange phenomena with incrasing rate of ball speed decay, maybe by observing another couple of spins too. (Maybe the DO use magnets in the wheel to control they outcome!)
Hi Rollo,
Reguarding your comment: "I am surprised to see that the ball speed decay rate, or deceleration, first increases towards a peak, and then falls off towards the end. "
So would I be, the deay rate only ever oves in one direction in the graphs as far as I can see. I just checked the them and I can’t see the peak you are refereing too so let me know wheich link and which Tab on that link you are refering to and I’ll take a look.
I can not open that site, it opens then my explorer crashes.
Gkd you are pain in the… Too many questions and too much reading for me. lol.
Put it this way, the time used 4.2 for some rotors may be better 4.6.
If you are close to starting point your errors are smaller.
Interesting point that you indicate is,
“Does that mean we have no solution for strong tilt using EX, just visual buildup monitoring instead?”
When I first time come across tilted wheel I used e2 but with different calculation.
Simply there was no multiplication by 3. The time I reduced but I do not remember by how much. Also starting point was somewhere around 12 sec.
CW spin
Let’s say the wheel is slow and time is 2 sec.
We start on top of wheel (12 o’clock)
Start timer and 2 sec after the ball is at 6 o’clock, (1.5 rotations) under ball there is number zero.
Ball travels additional x rotations and hits diamond at zero.
Next spin we do same, if ball after particular time is again at 6 o’clock that spin is 100% identical. So we can expect same. If it is 7 that ball is slightly faster, it will hit slightly higher on diamond or it will go for additional revolution.
If we are at 9 that simply can be same spin as first described but we entered 1 rev earlier.
The ball at 9 will be again over zero. Because we are 1 rev earlier and zero still did not come to 6 o’clock and the distance difference that ball travels now is close to wheel change for 1 ball rotation.
Now we can play every spin where ball after elapsed time is at 6 or 9 ‘o clock. Even if it is at 3 o’clock result will be same. When ball indicates 4, 5, 7, 8 we may not play it depends how strong tilt is. Simply change of ball traveling distance adjusts wheel change for time difference at starting point.
Yes, I took the difference once too many. It is not the change in speed which fluctuates up and down, it is the change in the change of speed, or the change in speed decay rate, which sometimes increases, sometimes decreases. This is however much more sensitive to measurement errors.
Time Speed Change "Change in change of speed"
1,92
2,56 0,64
3,26 0,70 0,060
4,02 0,76 0,060 0,000
4,83 0,81 0,050 -0,010
5,72 0,89 0,080 0,030
6,69 0,97 0,083 0,003
7,78 1,09 0,114 0,031
9,10 1,32 0,233 0,119
10,65 1,55 0,233 0,000
12,44 1,79 0,234 0,001
14,56 2,12 0,333 0,099
17,12 2,56 0,440 0,107
20,30 3,18 0,620 0,180
From which wheel did you get that data.
Can you show one more spin so we can compare.
Sorry, I should’ve explained better that my post was a (late) response to gdk:s question 31 october.
I just took the spread sheet data gdk linked to earlier, about the anti clockwise ball data. That is the time and speed columns. Then I try to deduce the decceleration (speed decay rate) from that.
I think that the conclusion is that the ball behaves as expected and that speed decays at a fairly steadily increasing rate. The variations in “change in change in speed” up and down probably come from rounding errors. Part of the peak with 0.199 probably belong to the following 0.000 value, so to speak. This I misinterpreted earlier.