Weekly blog

From MariachiWiki

February 12, 2008 lesson

Plateau measure efficiency what things are real and what isn't. Experimental Errors=how well we measure is limited by how we measure. The truth is in quote because most of the time we don't know what the truth is. The professor spoke of accuracy, precision and random errors which means that the more measurement you make the more answers are going to be closer to the truth. Accuracy, precision and systematic errors have to do with the human errors present when an object that is used to measure something else was made.

-R=N/t then dR/R rate =number of times counted divided by time. If you take the square root of the above formula, then you can estimate the % of error possibility in a measurement of cosmice rays particles. 2/3 of the time the right answer is going to between + and - As man made it up in the balloon, the cosmic rays increased-which means not all of cosmic rays make it down to us. Today we are measuring how perpendicular the cosmic rays are. Increase # of detectors to eliminate coincidences

My group is composed of Gillian, Desiree and myself of course. We tested the detectors and 45 degree angles and 0 degree-which means in a vertical position, parallel to each other with about 40 to 60 cm of space in between them. We set detector # 2 at 5.51 and # 3 at 5.68

We first tested the detectors several times at each angle in 20 second increments and the counts varied from 47 to 126 coincidences. At that point in the class we concluded that there wasn't a consistency in the number of coincidences. So we decided to measure the counts in 200 second increments.

We concluded that even though the difference in coincidences in between each measure was bigger at 200 seconds than at 20 seconds, the percentage of differences in measure of the several 200 seconds increments is very small when you take into account how many coincidences it counted.

But even though the number of coincidences counted in the 20 seconds increments is small, when you take into account the percentage of the differences in between each measure, the percentage is greater than that of the 200 seconds increments.

check out the following link to see the graph| [[1]]

February 19, 2008 Last week we measured cosmic rays with the detectors at 45 degrees facing north; tonight the detectors are facing east. Gillian and I used some trigonometry to figure out the exact angle in which we placed the detectors each time we measured the cosmic rays. I'm learning how to plot a graph after taking measurements. Gillian was walking me through it: highlight what I want on the chart: the first line is highlighted; then I holding the control key down I highlight the second line; then I click on the chart picture.

Lecture notes Showers-some particles don't produce other particles, some decay. Index of a fraction: transparent particles produce a light boom. Professor Marx said that in order to detect more particles, we would need to cover a wider area with detectors.

electromagnetic rays=rays of light-bounces a radial signal.

Theory- a cosmic ray creates an ionization and causes lightning to strike

the higher Barimetric pressure higher density in atmosphere, but we don't know what produces these energies.

As particles speed up, they gain mass-more energy

In order to do the trigonometry, we had to figure two sides and of course we knew it was a right angle, so that gave us 90 degrees.

We figured out one side by measuring the bottom of one of the sides of the wooden stand that supports the two detectors. We used this side as the hippotenus-which remained the same throughout every measurement and testing that we did.

So the only the only change in measurement was the adjacent-which we measured with a ruler. Formula opposite/adjacet (inverse Sin)

this is how we got the angles for every testing of cosmic ray particles we did.
,br> We measured the count with lots of octagon angles see graph|[[2]]

February 26, 2008

Lecture In today's lecture we learned about Chi square-which is the ((average of all points together)-(data point))/error= Chi squared (X squared)

Chi squared calculates how close your measurements are to being correct, and you want it to be close to 1.

It is always possible to connect the dots

When the Chi squared is 0, that means it's too good and we haven't learn anything. When it is too large, there is something wrong.

error bars=range of angles

error= count/time

As a group we experimented with linear and polynomial lines for the graph. We used the measurements we took today and last week.

We discovered that the line doesn't fit so well at higher angles for 3rd order polynomial Y, and that for linear Y it doesn't fit.

The other discovery was that when measuring the cosmic ray particles,the count that is off, is the one that throw us off when measuring Chi squared.

MARCH 4, 2008

SYSTEMATIC ERRORS

MEASURING TWO DETECTORS WITH ANOTHER ONE IN THE MIDDLE-WHICH WASN'T CONNECTED(GROUP MEMBERS:HARRY,JOE AND KELLY) FLUX= ((COUNT RATE)D^2)/((A1)A2)) LARGE EFFICIENCY PROBLEM

: COUNT VALUE WAS TOO LARGE

EFFICIENCY WAS LOUSY:NO PATTERN OR SYSTEMATIC CONSISTENCY

USED DIFFERENT FORMS OF GEOMETRY WITH NO BETTER ANSWER OR SOLUTION THE RESULT IS THAT AFTER TRYING ALL OF THE ASSUMTIONS, DIDN'T FIND A BETTER WAY TO GET AT THE RIGHT ANSWER.

COSMIC RAY CHRIS HYPOTHESIS CR WILL INCREASE AS DETECTOR IS HIGH UP

BIG DIFFERENCES IN THE COUNT BECAUSE OF THE FOUR FLOORS OF CONCRETE ABOVE US COULDN'T FIGURE OUT HOW TO GET A NICE LINEAR LINE ON THE NORTH SIDE OF THE BUILDING FOR THE S LEVEL WAS THE LARGEST RESULT AFTER DOING CHISQ ^2-91 ELIMINATED THE S LEVEL AND THE CHISQ ^2 WAS 13 MAYBE SURROUNDINGS ARE AFFECTING THE DATA RESULTS. MAYBE NEXT TIME THEY WILL TAKE COSMIC CHRIS UP TO THE ROOF TO ELIMINATE SURROUNDINGS EXCEPT THE ATMOSPHERE.

LOOKING FOR A LINEAR RELATIONSHIP

HIGHER RATE COUNT AT THE BOTTOM AND A HIGHER RATE COUNT AT THE TOP

CHISQ^2=TAKE AN AMOUNT OF DATA AND TAKE THEIR AVERAGE AND COMPARE TO ANOTHER AVERAGE YOU CAN CHANGE THE CHISQ^2 BY CHANGING THE SLOPE

imagining you got really good data points. Suppose the counts are constant or keep going down. I think that in this case we have to look at the Sistematic Errors.

The longer the delay, the longer the wire and shorter pulse.

Delay Box Dispersion was also discussed

To avoid accidental counts use more expensive equipment

The idea is not to measure error but to find the best possible count measure.

At this point non of the groups arrived at one specific answer, but rather we all had theories and hypothesis as to how or why of the different measurements, the different Chi^2;why is the latter too big or not close enough to 1.

other unanswered questions Why is there a delay reaction or the absence of one? Why is there a higher or lower count on one side of the room? Does being in the basement have a lot to do with the amount of Cosmic Rays that come down to us and subsequently limit the count? Are we on the right track?

Through more experiments and research-perhaps-these questions will have an answer. See 1st group presentation coverted[3]] and powerpoint presentation

MARCH 25, 2008

I'm with a different group now-Greg, Patrick and Lena. WE ARE MEASURING TIME IN BETWEEN THE TWO COUNTS OF TWO DETECTORS. WE PLACED TWO DETECTORS PARALLEL TO EACH OTHER ON THE OCTAGON. THE THIRD DETECTOR IS OUTSIDE OF THE OCTAGON. WHAT WE ARE MEASURING IS HOW LONG OF A DELAY THERE IS IN BETWEEN ONE PARTICLE FROM A SHOWER, AFTER HITTING THE FIRST DETECTOR ON THE OCTAGON, THAT HITS THE SECOND DETECTOR, AND HOW MUCH LONGER IT TAKES FOR A SECOND PARTICLE, FROM THE SAME SHOWER, TO HIT THE THIRD DETECTOR.

WE ALSO TRIED TO DETECT EVERY TIME THERE WAS A TRIPPLE COINCIDENCE, BUT THAT DIDN'T SEEM TO WORK BECAUSE IT IS MORE LIKELY THAT A SEPERATE PARTICLE FROM THE SAME SHOWER WOULD HIT THE ISOLATED DETECTOR. <

AND IT IS MORE LIKELY THAT THE SAME PARTICLE THAT HITS DETECTOR THREE ON THE OCTAGON, WILL HIT DETECTOR-TWO WHICH IS ALSO ON THE OCTAGON

THE ns TELLS US THE DIFFERENCE IN POSITION BETWEEN THE 2 CURSORS

THE LONGER THE DISTANCE THAT THE PARTICLE HAS TO GO THROUGH TO PASS THROUGH THE DETECTOR, THE HIGHER THE ENERGY CHARGE THAT THE DETECTOR WILL RECEIVE FROM THAT PARTICLE. THE LONGER THE PARTICLE IS IN THE DETECTOR, THE MORE ENERGY THE DETECTOR WILL RECEIVE: THE PARTICLE HAS TO TRAVEL WITHIN THE DETECTOR. WHAT'S GOING ON IS THAT THE ATOMS ARE BEING HIT BY THE PARTICLES WITHIN THE DETECTOR.

THE LENGTH OF THE CABLE CAN AFFECT THE TIME DIFFERENCE: THE LONGER THE CABLE IS, THE LONGER IT WILL TAKE FOR THE TIME DIFFERENCE TO GET TO THE BOX.

WE TOOK SEVERAL MEASUREMENTS WITH THE OCTAGON AT DIFFERENT DEGREE ANGLES. AND WE ALSO MOVED DETECTOR ONE SEVERAL TIMES

[[4]]

April 1, 2008

wfdata2008325[1].xlsx

Click [[5]] and enjoy my second journey-with a second group in cosmic rays research.

We are testing-better yet-measuring the difference in time in which two particle from the same shower hit detector 2 ad 1. After two weeks of testing, we still couldn't figure out why on average two of the detectors were receiving two different coincidences at the same time.

during this lesson we determined that the error on the average should be smaller than the single measurement.

with more than 10 standard diviation the above pictures depict the positions of the detectors in the different measurements we took to determine the difference times of the particles hitting detector 2 and 1

a few things were determine by my group:

The error will always decrease
(t(1))-(t(2)) should be bigger because of the time delays
the time on the data we collected is 4 hours ahead
showers are heating in some area here
I considered that maybe it is not the way the shower down but rather the length that it takes the particle to go through the detectors.
either 3 different particles from the same shower are hitting the detectors: one particle hits one detector, but all three particles are from the same shower.
or 1 particle hits 2 detectors and another hits the third detector.
conclusion
it is more likely that there are three different particles with one particle hitting one detector, but all particles are from the same shower.

April 8, 2008

today we did presentations [[6]]

Desiree's group worked on the speed of the particles. Speed=d/t Based on previous data her group predicted that particles would move faster than the speed of light
minimum: perpendicular straight line from one detector to the other
Errors time between occurrence 1-tried to measure between distance 2-speed of signal between the cables
campared their measured speed to the published speed
Result: 14% error; their speed was slower >did not get very far >no major conclusions >a Histogram was done- time in 3ns >big errors because they are doing hard measurements

Tom and Karyn's group
Recap: stacked detectors 4 detectors 2 by 2 > there was no possible way that one particle would go through 4 detectors >by the end of 1st experiment, they ended up with more questions 'Question'
>if continue experiments would they get the same data and it did

switched detectors and moved them by about one tile >collected data for 10 minutes instead of 3 >error analysis improved
low count rate
'Data comparison'
>why is it so much higher when they are closer together
detectors were placed 9 tiles apart, and they got 3 counts in 10 minutes
Professor Marx pointed out that there's good and bad with cosmic rays:'Good' because they are free, and 'Bad' because they are a mixture of all experiment particles
there are lots of pions in original collision, but less when they come down to us
Muons, electrons and everything else can be seperated by putting lead in the building
Our group
what we are measuring is a mixture of showers in different directions-3fold coincidences.
Group 4-Tania, Harry and Joe
the more energy, the more light >they took 16 different coincidences and average the energy >'Top scentillator' >particle will spend some time in the vertical and then go to the other scintillator
How to reduce the spread? 'Second Investigation'
decided to take a lot of single counts >Histogram of Energy vs number of counts for 300 Events >data did seem to follow the expected curve >data seems consistent Third Investigation
Energy deposited corresponds to energy lost by a particle
Muon in all 5 detectors, not a good data >best regression coefficient is at .5 slope
Follow up project:
extend the work of Tom and Karyn with horizontal separation and 4 fold coincidence rates
5 counters-put 1 and 2 in the outside and compare the numbers
Vincent's group-solo
'Cosmic Chris Part II'
>moved it closer to the middle of the building >Picked 5 different spots


each reading was for 2 minutes
x-axis is the height y-axis is the rate
height increases as you get to the back of the graph
lower counts as cosmic Chris gets closer to the building
questions: why is the count lower in the math building?
the math building has two extra floors
Gillians group-solo
she worked on finding where all the data that is accessible to her came from. Where is it stored?
she did some testing and measurements of her own and compared her results to that of the different sites she found.
she discovered that she must be doing something right because her findings where very close to the findings of the sites she researched.

April 15, 2008

histogram



I'm in a new group with Desiree. We are basically measuring the time it takes a cosmic ray particle to hit a second detector after hitting the first one. so the time recorded is the time between dector A and C. When we record time in the computer, the later doesn't give us the difference in time: it gives us the specific time in which each detector got hit by the same particle. So we had to deduct time A from time C

Velocity factor(VF) of common coaxial cables
If you make the wire shorter, once the particle hit a detector, it will take less time for the signal to get to the box.
the longer the wire, the longer it will take for the signal to get to the box.

THE SOME ERRORS
the signal can only go through the wire 66% of the spped of light. So the error turned out to be about 3 ns

importance of gage and difference
insulating material, middle wire and outside conductor, not likely to be able to minimize errors.

Because of the errors found in Desiree's previous experiments with her previous group, Dr Marx told us to pick the lowest point and the highest point of the histogram and find their slope as you can see in the above diagram.

((Y2-Y1)/(X2-X1))= m
We used two RG58 cables: each one was plugged into a detector and the box as you can see in the diagram.

And the distance in between detector A and detector C was 32cm when they were closely together.

And when they were apart, the distance was about 193cm

The above histogram or frequency distribution, shows the number of times you get a frequency.

April 22, 2008 [[7]]

speed and histogram[[8]]graph movement-everytime a particle hits one detector and the other, it shows this type of graph. It detects a cosmic ray and reaches some threashold Right now we are trying to minimize errors. Sistematic errors = something is wrong with what you are using to make measurements. For example, if a tape measurer is missing two inches and we are not aware of it, each time we measure with this tape measurer, we'll have a sistematic error.
Random error-as I understand it-is human error.

We are still trying to measure the difference in time when one detector is hit by one particle and the second detector being hit by the same particle.
Dima has a system set up in the computer that enables the latter detect multiple hits in a short time.
So we measure about 1000 hits with the detectors being about 32cm apart, and the average difference in time was 8ns
to sort out through the large amount of data, we first deducted the time that the first detector was hit from the time that the second detector got hit-the first detector is called A, and the second is C
Next we had the computer change the time difference into nanoseconds.
then we had the computer sort out-by lowest time to highest-each time the difference in time was repeated.
from this point we created a histogram which you can see if you click on one of the graph's icons.

The point is not the length of time we run the system but rather how many counts we get.
We are measuring the extremes to get a better error
Prof. Marx suggested that we plot both sets of data in the same histogram: measurement when the detectors are 197cm apart and at 32cm apart.
He also suggested that we find the widest point half way up in the histogram, know the distance, and this will give us the error.

day-scope-2ch.vi(program) records everytime there's an event: can record multiple hits in a short perioed of time.

April 29, 2008

check out today's graph| [[9]]Today we finished the last of the experiments for the semester. We needed to take the mean. Error for the mean-in time measurements.

We've been looking for the error in the measurements taken. there is always room for error.

we assumed that the machine-the box-and the distance don't have errors

there is error in the cable because the cosmic rays move faster than the speed of light

the cable used-RG 58-only measures 66% of the speed of light. Therefore 34% of the speed of the cosmic ray is not measured

Furthermore there has to be some error in the time.

the formula with histogram will tell us where the error is

We took all information, made 2 histograms, found the error and ploted it.

We tried to figure out if the number is a decent number

We calculated the error with the 2 numbers

the difference in slope between 197cm and 32cm gives us the error

Error should be between .0524+(plus) or -(minus) .0072

the intercept is not of a particular interest to us

to calculate the velocity

cm = m ns = s (10^-9)ns/(s)

May 6, 2008

Final presentation[[10]]

May 12, 2008 This is my final blog for this course. When I watched the other students' first presentations this semester, I noticed that on their powerpoints they had designed the different setups they worked throughout their first experiments. And I wanted to learn how to do that, but I didn't know how. One student told me to go to animations. Karen told another student to go to paint. [[11]]

2- I also wanted to learn how to graph on my own
Most of the time someone else in the group was doing the graphing
And when I did do the graphing, it was brief and toward the end of the lesson.
Of course by then I was exhausted, with very little energy, and my memory was gone
So I couldn't remember how to do it the next time I tried.[[12]]So I learned that with a little information-even if it isn't yours-I can graph in different shapes and forms.

3-The other thing I wanted to learn is the 3D design that Vincent used throughout the semester on his presentations to illustrate the level of particles on different floor levels and sections of the Math and Physics buildings. [[13]] 4-How close is the velocity that Desiree and I measured to the speed of light?

Using the formula Dima had given me the error on the velocity is 1.74cm/ns

From this we can say that the velocity that we measured is 30.63cm/ns+/- 1.7cm/ns =3.06E8m/s+/- 0.17E8m/s (2.89-3.23m/s)

Well the speed of light is 3E8m/s and our findings are 2.89E8m/s and 3.23E8m/s
So the difference between the speed of light and our measurements is -.11E8m/s and + .23E8m/s
this tells us that our measurements are so close to the speed of light that the difference is practically undetectable and that the error is practically unnoticeable.

We might be on the right track.