Time Machine Experiment

last updated:22 Mar 2003

All informations in this page are published free and are intended for private/educational purposes and not for commercial applications .

Time Machine Experiment, 1MHz Clocks (06 Mar 03)

Basically, the last experiment I ran using the Timex clocks was inconclusive since the standard deviation (measured error) was larger than any measured difference in the clocks. I built these new one megahertz clocks to get a better measurement, (and hopefully measure a significant time shift above the statictical errors). I'll discuss the statictics more, later, but for now I want to talk a little about these little marvels, I have constructed. These clocks have 36 bits of precision, and can mark time down to 1 microsecond! The engine of this clock is a 1MHz crystal oscillator. During synchronizing the starting/stopping of the clocks, they are hooked together through a pair of wires with a power plug connector. There are 36 LED's on each clock to powered by 3x12bit counters. The top chip is a quad nand gate chip. This serves two purposes, 1 it provided a fast switch for the clock pulses to the counters, and 2 it provides an electronic no-bounce switch for the reference clock using an RS Flipflop circuit.

Here is the traveling clock. I had to cut the sides off the perf board to allow this one to fit in the the hole of the Van De Graff sphere. There are three switches on this clock. Top right switch has three positions, run, stop, and neither/sync to other clock. Bottom right switch resets the counters. Bottom left switch allows the LEDs to be turned off. This is essential if you want the battery to last more than few hours.

OK, This is the reference clock, it has 4 switches. Top two right hand switches are set/reset for the RS flipflop. These start and stop the clocks, when they are hooked together. Bottom right switch is the counter reset. Bottom left, is the LEDs on/off switch. You may ask, why did I use two swithces to start and stop the clocks? I did this because it was the only way I could get it to work, and give me reliable synchronized starts between the two clocks.

OK, Here is a close up of the crystal I used.

And here is a close up of the counter chips I used (part number 4040BPC) The quad nands are part number SN74HC00N.

Here is my results for testing the syncronization of the two clocks. I preformed this test by quickly starting and stopping the clocks (about 30 times). Since the clock are so fast, there was appreciable drift, so I took some longer measurements between the two clocks to measure the drift. I was then able to subtract this drift (sec/sec) from the quick start and stop measurments. The results (histogram plot above) show that the clocks can be synchronized to each other with an error of about 2 microseconds. Notice the statistical distribution is skewed. I'm not sure exacly why this is yet, but I could be due to error in the clock drift measurement which I had to subtract off. (Next I need to use these clock to make real time measurements..... )

Time Machine Exp. Up Date, 1MHz clock results (12 Apr 03)

Here are the results of the time machine experiment using the 1MHz clock. Previously I used a slower clock which introduced a large error in the measurements. I now believe that the error was caused by temperature drifts causing the crystal oscillator frequency to drift (since the temperature of my house does change substantially over a week). I went to the faster/more-precise 1Mhz clocks so I could control the termperature more closely (and other variable like the degradation of the Van De Graff belt.. etc.) These new clocks only have to run about 1/2 of an hour to get the same precision as the timex after a week. Above is the experimental setup, with my temperature monitoring device.

Here is a close up of the temperature monitor device I used (electronic thermo couple unit sold as a handy-dandy indoor/outdoor thermomitor)

Ok Here are my results. I took ten measurments, five with the VDG generator running at High Voltage, and five with the VDG generator running, but the voltage is shorted out (No voltage). The Measureuments were alternatly taken to avoid systematic error over time (I.E. not shorted, shorted, not shorted, shorted,..etc.). I presented the data in the chart in their respective groups to make the chart more readable. Next, I calculated the mean and standard deviation of the two measurements. The difference in the means was 6.26E-8 and the error was 6.69E-7 (an order of magnitude larger that any measured difference). The Standard Deviation (error if you will) was greater than the difference of the means. That 'means' this experiment is inconclusive... (and I need to build, even yet... a beter clock) But don't fear.. I have plans to build..My own version of an Atomic Clock.

I'll just say a few word about statistics. Above is the formula for ... you guessed it.. the standard devition. From my previous start/stop measurements.. I know I could accuratly start/stop the clocks within 2 microseconds. I also measure the constant drift and was able to compensate for that. What I can't account for is a random rate of drift. But that is the beauty of experiment, If I take several measurements I can 'measure' the amount of 'unknown error' there is in my experiment. (BTW.. I am serious about the atomic clock things... It a comming) Let me close by discussing one more item... The Aharonov-Bohm Effect. After studdying and reading the original paper (which incidently suggests an experiment using electric potentails... as well as the well known magnetic potentails) I have come to the conclusion that the A-B effect may actually predict the time shift effect I'm looking for... I.E. the A-B effect may actually be a manifestation of the Electrogravity-effect. (And yes I understand that two seperate wavefuntions must be spacialy seperated.. potentalized and then brough back together and interfeared... This is not what I'm referring to .... I am more interested in the nonlocality implication of the A-Bohm effect, and the effect it may have on two relative potentailized clock..(and more specifically the time rate of progression) remember that the (rate of) phase (change) of the quantum mechanical wavefunction is dependent on the local potentail... Also phase is not a measureable quantity... It is only meaningful when compared (interfeared) with another wavefunction. The same is true of two clocks.... a time shift is only meaningful when the two clocks are complared. I'll have more to say on this ..later

Page 1

Copyright Information

Unless otherwise noted, all materials at this site (including without limitation all text, html markup, graphics, and graphic elements) are copyrighted ), 2002 by Reginald L. Jaynes. The material available through this site may be freely used for attributed noncommercial educational purposes only. We ask that due credit and notification be given the author.  

All materials appearing on this website may not be reproduced, stored in any retrieval system, or used in any way for commercial purposes without the express prior written permission of the copyright holder.  

Disclaimer: The author assumes no liability for any incidental, consequential or other liability from the use of this information. All risks and damages, incidental or otherwise, arising from the use or misuse of the information contained herein are entirely the responsibility of the user. Although careful precaution has been taken in the preparation of this material, we assume no responsibility for omissions or errors.  

HOME

Send e-mail to: dr_jaynes@hotmail.com