OU blog

Personal Blogs

ExLibris

Deriving Gamma in Special Relativity

Visible to anyone in the world
Edited by Martin Thomas Humby, Thursday, 27 June 2024, 19:03

This is an addendum to some aide-mémoire notes I am making on Relativity. Perhaps somebody may find it useful. Nowadays aide-mémoire has become more and more of a requirement for me.

In Special Relativity (SR), gamma gamma (Greek lower case), aka the Lorentz Factor, signifies the multiplier required to obtain a dilated (increased) time interval between two co-located events in inertial frame A as observed from another inertial frame B where A and B are in relative motion. Time dilation increases with relative velocity so gamma can be written as a function gamma of v . Gamma is ubiquitous in SR calculations.

Observers in both frames will measure the same spacetime separation between events occurring in either frame but, to allow a simple time-only translation using just gamma , the events in A must be co-located: – in the same spatial location in the frame, that is.

Time effects and most other outcomes of SR are symmetrical. If A sees B’s clock running slow then B will see A’s clock slow. This must be so to maintain the relativity principle: the motion of any entity can only be defined in relation to another entity and consequently absolute motion or rest have no meaning. With all motion being relative, observations made from one frame must mirror those made from the other.

Similarly, light (electromagnetic radiation) is not subject to Galilean velocity addition and has the same absolute speed c for all observers moving or otherwise. This is a requirement if electromagnetic phenomena are to show the same results in all locations as indeed they do. Both these requirements need additional examination elsewhere.

The derivation

For various reasons I found it difficult to find a convincing derivation for gamma. There was symbol inconsistency between sources with the common use of get-outs such as ‘rearranging,’ ‘simplifying’ foxing my rudimentary algebra skills. Some internet sources are just plain wrong I think. It was amusing to see that various online ‘AI’ equation solvers produced results with little resemblance to Einstein’s. Below is a derivation that shows all algebraic steps explicitly.

Most derivations, apart from Einstein (1905 p7), feature light clocks and Einstein does appear to mention the elongated light path seen in a moving frame:

Figure 1 Light Clocks shows a stationary clock and snapshots of a moving clock with right triangle h, vt/2, ct/2.In fact the mirrors, shown vertically spaced, can have any orientation at 90º to the direction of travel.

The norm is to consider a full clock cycle where a light pulse leaving the bottom mirror is reflected off the top mirror and returns. The ratio between the length of half a cycle in the stationary clock and half in the moving clock is obviously the same but a possible justification for the full-cycle approach is to emphasize that the two events, leaving and returning, must be co-located. Following this option:

If the time taken for a full cycle of the stationary clock is cap t , the distance the light pulse travels at velocity c between the mirrors is:

h equals c times cap t divided by two left parenthesis one right parenthesis

Similarly, in the moving clock if t is the time taken for a full cycle, the increased distance the light travels along a single diagonal is c times t solidus two . The distance the clock has moved during half a cycle is v times t solidus two where v is the relative velocity of the moving frame. The spacing of the mirrors h remains unchanged. Applying Pythagoras’ theorem to the right triangle in Figure 1 we can equate light travel distance c times t solidus two to h and v times t solidus two :

left parenthesis c times t divided by two right parenthesis squared equals h squared plus left parenthesis v times t divided by two right parenthesis squared left parenthesis two right parenthesis

Combining equations (1) and (2)

left parenthesis c times t divided by two right parenthesis squared equals left parenthesis c times cap t divided by two right parenthesis squared plus left parenthesis v times t divided by two right parenthesis squared

expanding the brackets

c squared times t squared divided by four equals c squared times cap t squared divided by four plus v squared times t squared divided by four

then multiplying both sides by 4 and dividing by c squared

c squared times t squared equals c squared times cap t squared plus v squared times t squared

t squared equals cap t squared plus left parenthesis v squared solidus c squared right parenthesis times t squared

rearranging gives

negative cap t squared equals negative t squared plus t squared times left parenthesis v solidus c right parenthesis squared

multiplying both sides by -1

cap t squared equals t squared minus t squared times left parenthesis v solidus c right parenthesis squared

take out t squared

cap t squared equals t squared times left parenthesis one minus left parenthesis v solidus c right parenthesis squared right parenthesis

square root both sides

cap t equals t times left parenthesis one minus left parenthesis v solidus c right parenthesis squared right parenthesis super one solidus two   or   cap t equals t times Square root of one minus left parenthesis v solidus c right parenthesis squared may be preferred.

From this, the period of one (or n) clock cycles cap t of the local clock is equal to the period of the same number of clock cycles of the moving clock multiplied by left parenthesis one minus left parenthesis v solidus c right parenthesis squared right parenthesis super one solidus two . The longer a clock cycle the slower the clock will run so dilated intervals for this clock as seen by an external observer will involve the reciprocal

multirelation normal cap delta times t sub cap a equals one divided by Square root of one minus left parenthesis v divided by c right parenthesis squared times normal cap delta times t sub cap b identical to gamma times normal cap delta times t sub cap b

where  normal cap delta times t sub cap a and normal cap delta times t sub cap b are the intervals between events seen in their respective frames.

Examples

gamma times normal cap delta times t sub cap b gives intervals, as seen by an observer in frame A, between co-located events in B. If a spaceship B has a relative velocity of half the speed of light relative to space-station A then

equation sequence part 1 gamma equals part 2 left parenthesis one minus one divided by four right parenthesis super negative one divided by two equals part 3 1.1547

Say that after every minute that passes by shiptime B transmits a time signal – these transmissions constitute co-located events in B. Standing off at right angles to B’s course to minimize any Doppler effect, space-station A receives signals at 1.1547 minute intervals. So, for every 1.1547 minutes of time that passes in A only 1 minute has elapsed in B.

To obtain the reduced time period experienced in moving frame B we need the reciprocal of gamma:

normal cap delta times t sub cap b equals one divided by gamma times normal cap delta times t sub cap a

From A’s perspective it will take B a year to travel half a lightyear but for B the time elapsed is 1/1.1547 = 0.8660 years.

Acknowledgement

The above derivation is based on Gray (2022) expanded and annotated with basic examples. Any misunderstandings and errors are without doubt my own.

References

Einstein, A. (1905) On the Electrodynamics of Moving Bodies. (trans. Dr. Anna Beck, and consultant Professor Peter Havas), Princeton Digital Einstein Papers [online] https://dn790004.ca.archive.org/0/items/einstein-1905-relativity/Einstein_1905_relativity.pdf

Gray, N. (2022) A Student’s Guide to Special Relativity, Cambridge University Press

Permalink Add your comment
Share post