Premature optimization; that is, spending all of your time analytically modeling and/or simulating systems before they’re built, is bad. It takes far too much time and mental labor, and, without any experimental link to reality to keep models on track, that is very well (and often is), time just wasted!

For example, the hours you spend modeling an IGBT in SPICE, or a transformer in COMSOL, perhaps with data pulled from spec-sheets, is time gambled. I’ve found that in almost every case, it takes far less effort, and far less time, to just build a test rig for the device and see for example, what waveforms, or what flux-densities can be achieved. Often it is the case that such measurements are far different from what your models expect, and further often, one finds that there are many further variables that weren’t accounted for in your model!

If you have an idea in mind, use your ingenuity and available tools to “just build it”. With practice one soon finds that often your guesses will work! Perhaps your design is not optimized, but, once your system is stable, it becomes a far, far easier task to optimize it thereafter than it would have been initially (mathematically). Sometimes, you’ll even find that what was initially “proven” impossible by theoretical modeling is in fact, entirely feasible, and maybe even the right way to go.

I shall now provide a case study:

Some time ago, perhaps one year to date, I set out to design a 100kHz, 30kV transformer for flyback converter purposes. The system goals were simple:

1) Take 72VDC

2) Turn it into 20kV, at 100kHz, with a power throughput of 5kW.

Initially, I did what I could to model, and simulate some designs. In doing so, I eventually found after 5 days’ work that something to this effect should be physically possible, given ideal conditions:

I then built such a transformer, and lo and behold, it sucked something fierce. This was concerning, as there were several parasitic elements taken into account; flux leakage, nonlinear core losses, stray capacitances, etc, yet still, the thing just sucked. And so, it was thrown out, and I adopted a new philosophy.

Instead of returning to my model, I proceeded to go about it the way I’m most familiar: the intuitive way. I considered instead of differential equations, the physical relationships between wire spacing, flux density, core size among other parameters, and picked values I felt were best.

When choosing a core, instead of taking into account permeabilies and fields modeling, I looked instead, at published BH curves. That is, data collected by the manufacturer of cores in experiment. Graphs of core loss vs temperature, saturation flux density vs frequency, replaced my mathematica equations and scaling constants, and the “proper” physical core size was determined simply by ripping apart a switch-mode welder, and *seeing what has thus far worked*.

Once a suitable core was obtained, my flux density setpoint was determined not by maxwells’ equations in \(\mathbb{R}^3\), but rather, by a few turns of wire on the core, a car audio amplifier, and an oscilloscope. It took all of 5 minutes to determine that my chosen core saturates (ie, distorts my sinusoidal test signal) at about 40V per turn. Mathematically, that should have been 18-ish.

In an effort to keep stray capacitance low, guided of course, by the simple concept of “a capacitor is two parallel plates”, the secondary winding was redesigned to be physically large and of few turns. Considering of course, that every flux line passing through the solenoid would contribute to induced emf, making the coil physically large wasn’t as much of a problem as some texts made it out to be.

Leakage inductance was kept low by replacing primary turns of wire, with primary turns made from copper sheet. I decided this was OK after considering that most of the current in a high frequency conductor would be flowing on the surface, anyway, so there’s little advantage to using wire. (spice model that, I dare you!)

Several more “educated guesses” eventually led to, a transformer. One that worked much better than my simulated transformer, but probably wasn’t optimized. Making notes of some errata I then followed up the

design with another revision*

*Eg, I noticed some very high density flux escaping the cores’ mating surfaces via experiments with iron filings, so I cut my primary in two to leave space for these to escape unhindered. This prevent magnetically induced, circulating currents in my primary.

…which, when all was done, worked, with only 2 days of patience and effort. Hot damn!

That my reader, is how science & engineering is done: **by seeing what works**. Doing so experimentally is fast, time efficient, and fun! Most others unfortunately don’t see benefit to such activity; they are too caught up in the beauty and false-reassurance of mathematics to understand that, models *are only models*. They are not reality, and they take time to build. Time that’s often better spent,*inventing.*

This applies elsewhere. If for example, you wish to design a protein, don’t waste months simulating Schrodinger’s equation in Matlab. Instead, look at NMR data of proteins similar to the ones you wish to design, understand how they function, and tinker with them. Add new groups, see what happens. When designing an airplane, CAD some interesting designs and toss them in a wind tunnel. When designing a nuclear reactor, consider not intense mathematics, but rather, your alloys, their characteristics, and the tools you have to measure and machine them. I dare you to build a model that perfectly predicts metal creep under extreme neutron bombardment!

Science is not as concrete as many imagine it to be. When going about it, don’t get lost in your head –instead, follow in the footsteps of Faraday, Tesla, Edison, Hedy Lamarr, Curie, Rontgen, Jack Kilby, Compton and countless others who’ve changed the way we see the world.

*Just do it!*

Plan your experiments with a pen and paper, make them work, and if you stumble on something of value, model it afterwards. Often it’s not as expensive to do so as you might think, and it gets work done faster too.

]]>For those like me whom are without the opportunity of riding through these fields on the stallion of true mathematics, this is discouraging. So much so that as of late, I have become convinced that my classes and their closed-box curricula are not helping me all that much in accomplishing the goals I have set for myself.

I’m learning of course, how to solve many simplified, general-case problems. Problems others have fabricated to be solved correctly in only one way, for the ease of others hired to evaluate how well students solve such problems given limited time. And while that is great exercise, and gives some children the self-esteem and courage needed to peruse a career in their chosen field; that’s not the way such field work. The real world today is numerically simulated, usually, with the help of software.

To say I’ve learned nothing though in my stay thus far at RIT would be a lie in its most general form. I’ve had the fortunate opportunity to learn things about life many others haven’t yet seemed to notice; things too countless to tally.

I have learned that unfortunately the world is a sad place to live in. Most, if not two-thirds of all people are only out to cheat you. They’ll do so if it’s beneficial to them financially or socially, and whether they take on the title of car dealer, insurance agent, lawyer, businessman, scientist, doctor, artist or engineer is irrelevant -for many, dollars speak louder than friendship.

I’ve learned that people in power are often there solely because they wish to be. I’ve learned that police departments, and the systems of law we have in place to empower them are fundamentally broken, in that criminal records and jailtime do nothing for this society but socially force others back onto the same roads, that such punishment intends to keep untraveled.

I’ve learned to at all cost avoid political discussions. I’ve learned that many drug ‘abusers’ have some of the biggest hearts you’ll ever find in a person, and that saving your own money with the hope of in the future paying others is fruitless if you wish to pay them well.

I’ve learned that academia is in a state of despair; papers are published not for content but instead for numbers, and the other metadata attached. They are evaluated based on how complex and unintuitive their simple concepts have been twisted to be, and professors are hired and judged not on the importance of their work, but instead on their number of citations and the value of the grant dollars they bring in. It’s heartbreaking to see them, at one time young wide-eyed students themselves, forced to continually write unread proposals instead of moving forward the arts and sciences they so love.

What I feel is my most prudent lesson of all however, is that everyone needs help.

Whether it is help in something as simple as a math operation, or as complex as coaching someone off the brink of suicidal despair, help from others, is fundamental to solving problems. No one goes about large projects on their own and if they claim to, they are fools and pompous liars.

Forgoing such help in my past projects has brought me to a state of mind where hours of the day have become irrelevant, and it has brought me to a state where the only thing that ever matters, is “what needs to be done next?”. It has brought me to such a state of physical and mental torment that I have forgotten to eat for days, and on occasion forgotten to sleep as well. It’s now unclear even, how to recognize that I am tired.

Forgoing help, has brought me to a state where my only means of available relaxation and rest, were the forced escape of alcohol and cannabinoids. I’ve found myself socially into a world of work-abusers; a world where people –even ones so young as junior undergraduates, have become reliant on cocaine and amphetamines just to get done the work they have promised others. Work that could be done easily, if they would just ask for help.

This ends now.

It is exceedingly humbling to recognize that the advice offered by my first mentors at this university was correct. If I’m going to move forward with the projects I anticipate to complete, I need to be talking to people and not transistors. I am very fortunate to have spent nearly 15 years of my life doing the latter; in that, the experience I’ve thus far gained is one most haven’t the chance to have until age 30. However, it takes many hands to build an airplane, or in my case, what I wish to be a paradigm-shifting x-ray machine.

That said, the state of research science is still broken -broken enough, that I cannot expect such a project to move forward under the roof of my, or other universities in finite polynomial time. Instead I need to find a team, money, and some modern equivalent of Dave Packard’s Palo-alto garage to make this happen. I shall look for that then, once I have finished the hackerspace project I’ve promised to oversee and complete.

It’s time to learn how to ask for help, and how to coherently organize thoughts and people. It’s time to learn how to write, and how to intuitively understand others’ complex emotions, as well as the mass emotions of a crowd. It’s time to learn how to rip through published articles and extract their useful content, and it’s time to understand the mess that is law; both patent, and criminal. It’s time to do this while I’m young, impressionable and have relatively little to lose. In any case, a degree in these modern days of social connection and access to limitless information, is irrelevant to my desired career of, “inventor”.

∎

]]>Upon reading chapter four of my assigned physics textbook [*Modern Physics, Krane*], I grew both tired and annoyed with the generalizations, or “leaps of faith” which author continually made. I soon found it more useful instead, to spend time reading the papers upon which these principals have been derived. Astonishingly however, I failed to find a modern, usable English translation of Werner Heisenberg’s landmark paper! More unfortunately even, the closest I did come on the hunt for such a translation was the discovery of a broken-english, NASA OCR script from 1988 hosted on the web archive. That won’t do.

Thus utilizing a day’s time, Google translate, MathJax and my personal skills at reading broken-english datasheets, I below have provided a modern translation of W. Heisenberg’s paper. For convenience of the reader, I have replaced some original variables used in the paper to more represent those found in common texts today. New notations such as euclidean norms (i.e, \(|f(x)|\)) have been instated, as well.

Dr. Heisenberg’s various justifications alone make for an interesting (and perhaps, very useful!) read, but for those short on time I have prepared also, a “too long, didn’t read” summary immediately preceding.

If we are to derive a model that quantizes space, perhaps to cells with lengths some finite dimension \(h\), then we are left with in the space \(\mathbb{Q}^2\) for example, a 2-dimensional grid of possible positions. Objects in this grid then, may be given some arbitrarily-defined co-ordinate, \(q\).

q of course, is a function of \((x,y)\) inside \(\mathbb{Q}^2\). \(x\), and \(y\) may only be integer multiples of h, or specifically:

\(q = \left \{ \forall (x, y)*h\in\mathbb{Q}^2 \right \}\)(don’t be scared, I’m just having fun with LaTeX!)

Now, if \(q\) is a function of yet another quantized variable, \(t\), then \(q(x,y)\) may be broken into \(q(x(t),y(t))\).

Thus if it’s fair to say “\(q\) can move as time advances integer multiples of h”, then it is possible to define some distance \(q_x\), that \(q\) has moved in that elapsed time \(\Delta t\). We may thus define a 1-dimensional “velocity” \(v_x = \frac{\Delta q_x}{\Delta t}\).

\(q\) however, is not a continuous function in this space, as it may only take on discrete values, themselves integer multiples of \(h\). Therefore it is useless to define “the velocity at a point”. More generally, \(q\)‘s average velocity for any time interval, \(\Delta t\), smaller than \(h\), is not definable.

Restated, only values of \(q_x\), or \(v_x\), can satisfy the below statement;

If time advances as \((integers) * h\), then \(\Delta q_x \geq h\) if our definition of “velocity” is to make any sense.

By extension, momentum in this direction, which is defined as \(m v_x\) must satisfy \(p_x \geq h\), if \(m\) can be no smaller than \(h\) as well.

*Now consider the thought:*

What if we were to look at the object \(q\), with absolute precision? That is, \(q_x\) is exactly defined, and \(\Delta q_x = 0\).

Then, if \(v_x\) is a function of \(\Delta q_x\) then as \(\Delta q_x(t \rightarrow 0)\), or “the change in \(q_x\)” approaches zero, then the function \(v_x(\Delta q_x(t \rightarrow 0))\) becomes indeterminate. This relation works on the converse as well, such that the relation:

\(\Delta q_x * m \Delta v_x \geq h\) is justified!

In our 3 dimensional world \(\mathbb{Q}^3\), this equation becomes the familiar Heisenberg uncertainty principal:

\(\Delta q_x\;\Delta p_x \geq \frac{h}{2 \pi}\)The factor of \(2 \pi\) is a geometric normalization.

The origins of this relation’s elegance are plain to see: it is one derived from simple principals! Below, Heisenberg purports similar arguments exist for an energy-time relationship, and proves both relations are just as true for wave-functions as they are for discrete, “particle” functions. I’ll leave that lesson to be a test of your reading comprehension skills, however.

W. Heisenberg, a modern translation by Adam Munich

First, exact definitions are supplied in this paper for the terms: position, velocity, energy, etc. (of the electron, for instance), such that they are valid also in quantum mechanics; then we shall show that canonically conjugated variables can be determined simultaneously only with a characteristic uncertainty. This uncertainty is the intrinsic reason for the occurrence of statistical relations in quantum mechanics. Their mathematical formulation is made possible by the Dirac-Jordan theory. Beginning from the basic principles thus obtained, we shall show how macroscopic processes can be understood from the viewpoint of quantum mechanics. Several imaginary experiments are discussed to elucidate the theory.

We believe to understand a theory intuitively, if in all simple cases we can qualitatively imagine the theory’s experimental consequences and if we have simultaneously realized that the application of the theory excludes internal contradictions. For instance: we believe to understand Einstein’s concept of a finite three-dimensional space intuitively, because we can Imagine the experimental consequences of this concept without contradictions. Of course, these consequences contradict our customary intuitive space-time beliefs. But we can convince ourselves that the possibility of applying this customary view of space and time can not be deduced either from our laws of thinking, or from experience.

The intuitive interpretation of quantum mechanics is still full of internal contradictions, which become apparent in the battle of opinions on the theory of continuums and discontinuums, particles and waves. This alone tempts us to believe that an interpretation of quantum mechanics is not going to be possible in the customary terms of kinematic and mechanical concepts. Quantum theory, after, derives from the attempt to break with those customary concepts of kinematics and replace them with relations between concrete, experimentally derived values. Since this appears to have succeeded, the mathematical structure of quantum mechanics won’t require revision, on the other hand. By the same token, a revision of the space-time geometry for small spaces and times will also not be necessary, since by a choice of arbitrarily heavy masses the laws of quantum mechanics can be made to approach the classic laws as closely as desired, no matter how small the spaces and times. The fact that a revision of the kinematic and mechanic concepts is required seems to follow immediately from the basic equations of quantum mechanics.

Given a mass \(m\), it is readily understandable, in our customary understanding, to speak of the position and of the velocity of the center of gravity of that mass \(m\). But in quantum mechanics, a relation \( \mathbf{p\;q-q\;p} = \frac{h}{2 \pi} \) exists between mass, position and velocity. We thus have good reasons to suspect the uncritical application of the terms “position” and “velocity”. If we admit that for very small spaces and times discontinuities are somehow typical, then the failure of the concepts precisely of “position” and “velocity” become immediately plausible:

If, for instance, we imagine the one-dimensional motion of a mass point, then in a continuum theory it will be possible to trace the trajectory curve \(x(t)\) for the particle’s trajectory (or rather, that of its center of mass) *[see Fig. l, above]*, with the tangent vector to the curve indicating the velocity, in each case.

In a discontinuous theory, in contrast, instead of a smooth curve we shall have a series of points spaced at finite distances *[see Fig. 2, above]*. In this case it is obviously pointless to talk of the velocity at a certain position, since the velocity can be defined only by means of *two positions*, with two different velocities corresponding to each point.

The question thus arises whether it might not be possible, by means of a more precise analysis of those kinematic and mechanical concepts, to clear up the contradictions currently existing in an intuitive interpretation of quantum mechanics, to thus achieve an intuitive understanding of the relations of quantum mechanics.^{1}

^{1} This paper was written as a consequence of the efforts and wishes expressed clearly by other scientists much earlier, before quantum mechanics was developed. I particularly remember Bohr’s papers on the basic tenets of quantum theory (for instance, Z.f. Physik 13, 117 (1923)) and Einstein’s discussions on the relation between wave fields and light quanta. In more recent times, the problems here mentioned were discussed most clearly by W. Pauli, who also answered some of the questions that arise (“Quantentheorie, Handbuch d.Phys. ["Quantum theory", Handbook of Physics] Vol. XXIII). Quantum mechanics has changed little in the formulation Pauli gave to these problems. It is also a special pleasure for me here to thank Mr. W. Pauli for the stimulation I derived from our oral and written discussions, which have substantially contributed to this paper.

In order to be able to follow the quantum-mechanical behavior of any object, it is necessary to know the object’s mass and and the interactive forces with any fields or other objects. Only then is it possible to set up the Hamiltonian function for the quantum-mechanical system. I shall in general refer to non-relativistic quantum mechanics, since the laws of quantum-theory electrodynamics are not completely known yet.

If we want to clearly understand what is meant by the word “position of the object” – for instance, an electron relative to a given reference system, then we must indicate the definite experiments by means of which we intend to determine the “position of the electron.” Otherwise the word is meaningless. In principle, there is no shortage of experiments that permit a determination of the “position of the electron” to any desired precision, even. For instance: illuminate the electron and look at it under the microscope. The highest precision attainable here in the determination of the position is substantially determined by the wavelength of the light used.

But let us imagine however a gamma-ray microscope and by means of it, determine the position as precisely as desired. But in this determination a secondary circumstance becomes essential: the Compton effect. Any observation of the scattered light coming from the electron (into the eye, onto a photographic plate, into a photocell) presupposes a photoelectric effect. That is, this event may also be interpreted as a photon striking the electron, there being reflected or diffracted to then deflected once again by the microscope’s lens, finally triggering the photoelectric effect.

At the instant of the determination of its position – in example, the instant at which the photon is diffracted by the electron, the electron discontinuously changes its impulse. That change will be more pronounced the smaller the wavelength of the light used, i.e, the more precise the position determination is to be. In the instant at which the electron’s position is known, therefore, its impulse can become known only to the nearest order of magnitude corresponding to that discontinuous change. That is, the more precisely the position is determined, the more imprecisely the impulse will be known, and vice-versa. This provides us with a direct, intuitive clarification of the relation \( \mathbf{p\;q-q\;p} = \frac{h}{2 \pi} \).

Let \(\Delta q\) be the precision to which the value of \(q\) is known (\(\Delta q\) is approximately the average error of \(q_1\)), or here, the wavelength of the light used to “see” the object. Let \(\Delta p\) be the precision to which the value of \(p\), the object’s momentum, can be determined, or in this case, the discontinuous change in \(p\) during the Compton effect. According to the basic equations of the Compton effect, the relation between \(\Delta q\) and \(\Delta p\) is then:

How that relation (1) above stands in a direct mathematical connection with the commutation relation \(\mathbf{pq-qp} = \frac{\hbar}{2 \pi}\) shall be shown below. Here we shall point out that equation (1) is the precise expression for the fact that we once sought to describe by dividing the phase space into cells of size \(h\).

Other experiments can also be performed to determine the electron’s position, such as impact tests. A very precise determination of the position requires impacts with very fast particles, since for slow electrons the diffraction phenomena, which according to Einstein are a consequence of the de Broglie waves (see for instance the Ramsay effect), preclude a precise determination of the position. Thus, once again for a precise position measurement the electron’s impulse changes discontinuously and a simple estimate of the precision with the equations of the de Broglie waves once again leads to equation (1).

This discussion seems to define the concept of position of the electron clearly enough and we only need to add a word about the “size” of the electron. If two very fast particles strike the electron sequentially in the very brief time interval \(\Delta t\), then the two positions of the electron defined by these two particles lie very close together, separated by a distance \(\Delta l\). From the laws observed for alpha-particles we conclude that \(\Delta l\) can be reduced to a magnitude of the order of \(10^{-12}\) cm, provided \(\Delta t\) is sufficiently small and the particles selected are sufficiently fast. That is the meaning, when we say that the electron is a particle whose radius is not greater than \(10^{-12}\) cm.

Let us move on to the concept of the “path of the electron.” By path or trajectory we mean a series of points in space (in a given reference system) that the electron adopts as successive “positions.” Since we already know what “position at a certain time” means, there are no new difficulties here. It is still readily understood that the often used expression, for instance, “the 1s orbit of the electron in the hydrogen atom” makes no sense, from out point of view. Because in order to measure this 1s orbit, we would have to illuminate the atom with light such that its wavelength is considerably shorter than its “size”, approximately \(10^{-8}\) cm.

But one light quantum of this kind of light would be sufficient to completely throw the electron out of its “orbit” (for which reason never more than a single point of this “path” could be defined, in space) and hence the word “path” is not very sensible or meaningful, here. This can be easily derived from the experimental possibilities, even without any knowledge of the new theories.

In contrast, the imaginary position measurements can be performed for many atoms in a 1s state. (Atoms in a given “stationary” state, for instance, can in principle be isolated by the Stern-Gerlach experiment) Thus, for a given state, for instance 1s, of an atom, a probability function must exist for the electron’s positions, such that *it* corresponds on the average, to the classical trajectory over all phases, and that can be established by measurements to any desired precision.

According to Born^{2} this function is given by \(|\psi _{1s}(x)|^2\), where \(\psi _{1s}(x)\) is the Schroedinger wave function corresponding to the state 1s. I want to join Dirac and Jordan, in view of subsequent generalizations, in saying: the probability is given by \(|S(1s, q)|^2\), where \(S(1s, q)\) is that column of the transformation matrix \(S\) from \(E\) to \(q\), which corresponds to \(E = E_{1s}\) ( where \(E\) = Energy)

^{2} The statistical meaning of the de Broglie waves was first formulated by A. Einstein [Sitzungsber.d.preuss.Akad.d.Wiss. 1925, p3]. This statistical element then plays a significant role for M. Born, W. Heisenberg and P. Jordan, ["Quantum mechanics 11." [Z.f.Phys. 35, 557 (1926)], especially chapter 4, 53, and P. Jordan” TZ. .Phys, 37, 376

(1926)]; it is analyzed mathematically in a fundamental paper by M. Born [Z.f.Phys. 38, 803 (1926)] and used for the interpretation of the collision phenomena. The foundation for using the probability theorem from the transformation theory for matrices can be found in: [W. Heisenberg (Z. f.Phys. AO, 501 (1926)], P. Jordan [libid. 40, 661 (1926)], W. Pauli. [in Z.f.Phys. 41, 81 (1927)] P. Dirac [Proc. Roy, Soc.(A) U2f 621 (1926)], P. Jordan [Z.f.Phys. 40, 809 (1926)]. The statistical side of quantum mechanics in general is discussed by P. Jordan [Naturwiss. 15, 105 (1927)] and M. Born [Naturwiss. 15, 238 (1927)].

In the fact that in quantum theory for a given state, for instance 1s – only the probability function for the electron position can be given, we may see a characteristic statistical feature of quantum theory, as do Born and Jordan, quite in contrast to the classical theory. On the other hand, if we want to we can say with Dirac that the statistics came in via our experiments. Because also in classical theory only the probability of a certain electron position could be given, as long as we do not know the atom’s phases. Rather, the difference between classical and quantum mechanics consists in this: classically, we can always assume the phases to have been determined in a previous experiment. But in reality this is impossible, because every experiment to determine the phase would either destroy or modify the atom. In a definite stationary “state” of the atom, the phases are indeterminate in principle, which we may consider a direct clarification of the

known equations:

Where \(\mathbf{J}\) is the action variable and \(\mathbf{\omega}\) is the angular variable.

The word “velocity” of an object is easily defined by measurement if it is a force-free motion. For instance, the object can be illuminated with red light and then the particle’s velocity can be determined by the Doppler effect of the scattered light. The determination of the velocity will be the more precise, the longer the wavelength of the light used is, since then the particle’s velocity change per incident photon due to the Compton effect will be the smaller. The position determination becomes correspondingly uncertain, as required by equation (1). If the velocity of the electron in an atom is to be measured at a certain instant, we should have to make the nuclear charge and the forces due to the other electrons disappear, at that instant, so that the motion may proceed force free, after that instant, to then perform the determination described above.

As was the case earlier, we once again can convince ourselves that a function \(p(t)\) for a certain state of the atom – say, 1s – can not be defined. In contrast, there again will be a function for the probability of \(p\); this state, which according to Dirac and Jordan will have the value \(|S(1s, \mathbf{p})|^2\). Again, \(S(1S, p)\) means the column of the transformation matrix \(S\) of \(E\) into \(p\) that corresponds to \(E=E_{1S}\).

Finally, let us point out the experiments that allow the measurement of the energy or the value of the action variables \(\mathbf{J}\). Such experiments are particularly important since only with their aid will we be able to define what we mean, when we talk about the discontinuous change of the energy of \(\mathbf{J}\). The Franck-Hertz collision experiments permit the tracing back of the energy measurements on atoms to the energy measurements of electrons moving in a straight line, because of the validity of the energy theorem in the quantum theory.

In principle, this measurement can be made as precise as desired if we forego the simultaneous determination of the electron position, i.e., of the phase (see above, the determination of p), corresponding to the relation \( \mathbf{Et-tE} = \frac{h}{2 \pi i} \). The Stern-Gerlach experiment permits the determination of the magnetic or an average electric moment of the atom, i.e., the measurement of magnitudes that depend only the action variables \(\mathbf{J}\). The phases remain undetermined in principle. If it is not sensible to talk of the frequency of a light wave at a given instant, it is not possible either to speak of the energy of an atom at a particular instant.

In the Stern-Gerlach experiment this corresponds to the situation that the precision of the energy measurement will be the smaller, the shorter the time interval during which the atom is under the influence of the deflecting force. Because an upper limit for the deflecting force is given by the fact that the potential energy of that deflecting force inside the beam of rays can vary only by quantities that are considerably smaller than the energy differences of the stationary states, if a determination of the stationary states’ energy is to be possible. If \(\Delta E\) is the quantity of energy that satisfies that condition, (\(\Delta E\) at the same time is a measure of the precision of that energy measurement), then \(\frac{\Delta E}{d}\) is the maximum value for the deflecting force, if d is the width of the ray beam (measurable by means of the width of the slit used).

The angular deflection of the atom beam is then \(\frac{\Delta E \Delta t}{d \Delta p}\) where \(\Delta t\) is the period of time during which the atoms are under the effect of the deflecting force, \(\Delta p\) is the impulse of the atoms in the direction of the beam. This deflection must be at least of the same order of magnitude as the natural beam broadening caused by diffraction in the slit, in order for a measurement to be possible. The angular deflection due to diffraction is approximately \(\frac{\lambda}{d}\), where \(\lambda\) is the de Broglie wavelength, i.e…

\(\frac{\lambda}{d} \sim \frac{\Delta E \Delta t}{d \Delta p}\) and since \(\lambda = \frac{h}{p}\)

This equation corresponds to equation (1) and it shows that a precise energy determination can be attained only through a corresponding uncertainty in the time.

We would like to summarize the results of the previous section and generalize them in this statement: all concepts used in classical theory to describe a mechanical system can also be defined exactly for atomic processes, in analogy to the classic concepts. But purely from experimentation, the experiments that serve for such definitions carry an inherent uncertainty, if we expect from them the simultaneous determination of two canonically conjugated variables. The degree of this uncertainty is given by equation (1), widened to include any canonically conjugated variables. It is reasonable then, to compare the quantum theory with the special theory of relativity.

According to the theory of relativity, the term “simultaneous” can only be defined by experiments in which the propagation velocity of light plays an essential role. If there were a “sharper” definition of simultaneity – for instance, signals that propagate infinitely rapidly – then the theory of relativity would be impossible. But since such signals do not exist – because the velocity of light already appears in the definition of simultaneity – room is available for the postulate of a constant velocity of light and therefore this postulate is not contradicted by the appropriate use of the terms “position, velocity, time”.

The situation is similar in regard to the definition of the concepts “electron position” and “velocity”, in quantum theory. All the experiments we could use to define these terms necessarily contain the uncertainty expressed by equation (1), even though they permit an exact definition of the individual concepts \(\Delta p\) and \(\Delta q\). If experiments existed that allowed a “more precise” definition of \(\Delta p\) and \(\Delta q\) than that corresponding to equation (1), then quantum theory would be impossible.

This uncertainty – which is fixed by equation (1)

Now provides the space for the relations that find terse expression in the commutation relations of quantum mechanics. This equation becomes possible without having to change the physical meaning of the differences, \(\Delta p\) and \(\Delta q\).

For those physical phenomena for which a quantum theory formulation is still unknown (for instance, electrodynamics), equation (1) represents a demand that may be helpful in finding the new laws. For quantum mechanics, equation (1) can be derived from the Dirac-Jordan formulation, by means of a minor generalization. If for a certain value of an arbitrary parameter \(\eta\) we can determine the position \(q\) of the electron at \(q’\) with a precision \(\Delta q\), then we can express this fact by means of a probability density \(\psi(\eta, q)\) that will be noticeably different from zero only in an area of approximate dimension \(\Delta q\) around \(q’\). We can thus say, more specifically:

We thus have for the probability amplitude corresponding to \(p\):

In agreement with Jordan, we can say:

In that case, according to (4), \(S(\eta,p)\) will be noticeably different from zero only for values of \(p\) for which \(\frac{2 \pi (p – p’) \Delta q}{h}\) is not substantially larger than 1. More especially, in the case of (3) we shall have:

Thus, assumption (3) for \(S(\eta,p)\) corresponds to the experimental fact that the value \(p’\) of \(p\) and the value \(q’\) of \(q\) were measured [with the precision restriction (6)].

The purely mathematical characteristic of the Dirac-Jordan formulation of quantum mechanics is that the relations between \(\mathbf{p, q}, E\), etc., can be written as equations between very general matrices, such that any variable indicated by quantum theory appears as the diagonal matrix. The feasibility of such a notation seem reasonable if we visualize the matrices as tensors (for instance, moments of inertia) in multidimensional spaces, among which mathematical relations exist. The axes of the coordinate system in which these mathematical relations are expressed can always be placed along the main axis of one of these tensors. It is after all always possible to characterize the mathematical relation between two tensors \(\mathbf{A}\) and \(\mathbf{B}\) by means of transformation formulae that will convert a system of coordinates oriented along the main axis of \(\mathbf{A}\), into one oriented along the main axis of \(\mathbf{B}\). The latter formulation corresponds to Schroedinger’s theory.

In contrast, Dirac’s notation of the q-numbers must be considered the truly “invariant” formulation of quantum mechanics, independent of all coordinate systems. If we wanted to derive physical results from that mathematical model, then we must assign numerical values to the quantum mechanics variables, i.e., the matrices (or “tensors” in multidimensional space). This is to be understood as meaning that in that multidimensional space a certain direction is arbitrarily chosen (that is, established by the kind of experiment performed), and then the “value” of the matrix is asked for (for instance, the value of the moment of inertia, in that picture), in the direction chosen. This question has unequivocal meaning only if the direction chosen coincides with one of the matrix’s main axes: in that case there will be an exact answer to the question. If the direction chosen deviates but little from one of the matrix’s main directions, we can still talk with a certain imprecision, given by the relative inclination, with a certain probable error of the “value” of the matrix in the direction chosen.

We can thus state: it is possible to assign a number to every quantum theory variable, or matrix, which provides its “value”, with a certain probable error. The probable error depends on the system of coordinates. For each quantum mechanics variable there exists one system of coordinates for which the probable error vanishes for that variable. Thus, a given experiment can never provide precise information on all quantum mechanics variables: rather, it divides the physical variables into “known” and “unknown” (or: more or less precisely known variables), in a manner characteristic for that experiment. The results of two experiments can be derived precisely from each other only when the two experiments divide the physical variables in the same manner into “known” and “unknown” (i,e., if the tensors in that multidimensional space already used for visualization are viewed from the same direction, in both experiments.) If two experiments cause two different distributions into “known” and “unknown” variables, then the relation of the results of those experiments can be given appropriately only statistically.

Let us perform an imaginary experiment, to more precisely discuss these statistical relations. We shall start by sending a Stern-Gerlach beam of atoms through a field \(F_1\) that is so heterogeneous in the beam direction, that it causes noticeably numerous transitions due to a “shaking effect”. The atom beam is then allowed to run unimpeded, but then a second field shall begin, \(F_2\), as heterogeneous as \(F_1\). We shall assume that it is possible to measure the number of atoms in the different stationary states, between \(F_1\) and \(F_2\) and also beyond \(F_2\) by means of an eventually applied magnetic field. Let us assume the atoms’ radiative forces to be zero.

If we know that an atom was in the energy state \(E_n\) before passing through \(F_1\), then we can express this experimental fact by assigning a wave function to the atom – for instance, in \(\mathbb{p}\)-space – with a certain energy \(E_n\) – and the undetermined phase \(\phi_n\), and an initial phase \(\phi_0\):

Let us assume that here the \(\phi_n\) are arbitrarily fixed, such that the \(c_{n\;m}\) is unequivocally determined by \(F_1\), The matrix \(c_{n\;m}\) transforms the energy value before passing through \(F_1\) to that after passing through \(F_1\). If behind \(F_1\) we perform a determination of the stationary states – for instance, by means of an heterogeneous magnetic field – then we shall find, with a probability of \(c_{n\;m}\overline{c_{n\;m}}\) that the atom has passed from the state n to the state m. If we determine experimentally that the atom has actually acquired the state m, then in the subsequent calculations we shall have to assign it the function \(\sum_{m}c_{n\;m}S_{m}\) with an indeterminate phase, instead of the function \(S_m\).

Through the experimental determination “state m” we select, from among the different possibilities \(c_{n\;m}\), a certain m and simultaneously destroy, as we shall explain below, whatever remained of phase relations in the variables \(c_{n m}\). When the beam passes through \(F_2\), we repeat the same procedure used for \(F_1\). Let \(d_{n\;m}\) be the coefficients of the transformation matrix that convert the energies before \(F_2\) to those after \(F_2\). If no determination of the state is performed between \(F_1\) and \(F_2\), then the eigenfunction is transformed according to the following pattern:

Let \(\sum_{m}c_{n\;m}d_{m\;l} = e_{n\;l}\). If the stationary state of the atom is determined, after \(F_2\), we shall find the state “l” with a probability of \(e_{nl}\overline{e_{n\;l}}\). If in contrast, we determined “state m” between \(F_1\) and \(F_2\), then the probability for “l” behind \(F_2\) is given by \(d_{n\;l}\overline{d_{n\;l}}\). Repeating the entire experiment several times, (determining the state, each time, between \(F_1\) and \(F_2\)) we shall then observe the state “l”, behind \(F_2\) with the relative frequency \(\sum_{m}c_{n m}\overline{c_{n\;m}}d_{m\;l}\overline{d_{m\;l}}\). This expression does not agree with \(E_{n\;l}\overline{E_{n\;l}}\).

For this reason Jordan mentions an “interference of the probabilities”. I, for one, would not agree with this, because the two experiments leading to \(E_{n\;l}\overline{E_{n\;l}}\) or \(Z_{n\;l}\), respectively, are really physically different. In one case the atom suffers no disturbance between \(F_1\) and \(F_2\). In the other it is disturbed by the equipment that makes the determination of the stationary states possible. The consequence of this equipment is that the “phase” of the atom changes by quantities that are uncontrollable in principle, just as the impulse was changed in the determination of the electron’s position (of. §1). The magnetic field for the determination of the state between \(F_1\) and \(F_2\) will change the eigenvalues \(E\) and during the observation of the atom beam (I am thinking of something like a Wilson track) the atoms will be slowed down in different degrees, statistically, and in an uncontrollable manner. As a consequence, the final transformation matrix \(E_{n\;l}\) (from the energy values before \(F_1\) to those after leaving \(F_2\)) is no longer given by \(\sum_{m}c_{n\;m}d_{m\;l}\), and instead each term of the sum will have, in addition, an unknown phase factor. Hence, all we can expect is for the average value of \(E_{n\;l}\overline{E_{n\;l}}\) over all eventual phase changes, to be equal to \(Z_{n\;l}\). A simple calculation shows this to be the case.

Thus following certain statistical rules, we can draw conclusions, based on one experiment, regarding the results possible for another. The other experiment selects, by itself and from among all the possibilities, one particular one, thus limiting the possibilities for all subsequent experiments. This interpretation of the equation for the transformation matrix \(S\), or Schrodinger’s wave equation, is possible only because the sum of all solutions is also a solution. Here we can see the deeper meaning of the linearity of Schrodinger’s equations and hence they can be understood only as waves in the phase space; for it is same reason we would consider any attempt to replace these equations – for instance, in the relativistic case (for several electrons) – by non-linear equations is doomed to failure.

I believe the analyses performed in the preceding sections of the terms “electron position”, “velocity”, “energy”, et cetera, have sufficiently clarified the concepts of quantum theory kinematics and mechanics, so that an intuitive understanding of the macroscopic processes must also be possible, from the point of view of quantum mechanics. The transition from micro to macro mechanics has already been dealt with by Schrodinger, but I do not believe that Schrodinger’s considerations address the essence of the problem, for the following reasons: according to Schrodinger, in highly excited states a sum of the eigenvibrations will yield a not overly large wave packet, that in its turn, under periodic changes of its size, performs the periodic motions of the classical “electron”. The following objections can be raised here:

If the wave packet had such properties as described here, then the radiation emitted by the atom could be developed into a Fourier series in which the frequencies of the harmonic vibrations are integer multiples of the fundamental frequency. Instead, the frequencies of the spectral lines emitted by the atom are never integer multiples of a fundamental frequency, according to quantum mechanics – with the exception of the special case of the harmonic oscillator. Thus Schrodinger’s consideration is applicable only to the harmonic oscillator considered by him, while in all other cases in the course of time the wave packet spreads over all space surrounding the atom. The higher the atom’s excitation state, the slower will be the scattering of the wave packet. But it will occur, if one waits long enough.

The argument used above for the radiation emitted by an atom can be used, for the time being, against all attempts of a direct transition from quantum to classical mechanics, for high quantum numbers. For this reason, it used to be attempted to circumvent that argument by pointing to the natural beam width of the stationary states; certainly improperly, since in the first place this way out is already blocked for the hydrogen atom, because of insufficient radiation at higher states; in the second place, the transition from quantum to classical mechanics must be understandable without borrowing from electrodynamics. Bohr has repeatedly pointed out these known difficulties, in the past, that make a direct connection between quantum and classical theory difficult. If we explained them here again in such detail, it is because apparently they have been forgotten.

I believe the genesis of the classical “orbit” can be precisely formulated thus: the “orbit” only comes into being by our observing it. Let us assume an atom in its thousandth excitation state. The dimensions of the orbit are relatively large here, already, so that it is sufficient, in the sense of §1, to determine the electron’s position with a light of relatively long wavelength. If the determination of the electron’s position is not to be too uncertain, then one consequence of Compton recoil will be that after the collision, the atom will be in some state between, say, the 950th and the 1050th. At the same time, the electron’s impulse can be derived – to a precision given by equation (1) – from the Doppler effect. The experimental fact so obtained can be characterized by means of a wave packet – or better, probability packet – in \(\mathbb{q}\)-space, by a variable given by the wavelength of the light used, essentially composed of eigenfunctions between the 950th and the 1050th eigenfunction, and through the corresponding packet in \(\mathbb{p}\)-space. After a certain time, a new position determination is performed to the same precision.

According to §2, its result can be expressed only statistically; possible positions are all those within the now already spread wave packet, with a calculable probability. This would in no way be different in classical theory, since in classical theory the result of the second position could also be given only statistically, due to the uncertainty in the first determination; in addition, the system’s orbits would also spread in classical theory similarly to the wave packet. However, the laws of statistics themselves are different in quantum mechanics and classical theory. The second position determination “\(q\)” selected from among all those possible, thus limiting the possibilities for all subsequent determinations. After the second position determination, the results for later measurements can be calculated only by again assigning to the electron a “smaller” wave packet of dimension \(\lambda\) (the wavelength of the light used for the observation).

Thus, each position determination reduces the wave packet again to its original dimension \(\lambda\). The “values” of the variables \(p\) and \(q\) are known to a certain precision, during all experiments. Since within these limits of precision the values of \(p\) and \(q\) follow the classical equations of motion, we can conclude, directly from the laws of quantum mechanics:

But as we mentioned, the orbit can only be calculated statistically from the initial conditions, which we may consider a consequence uncertainty existing in principle, in the initial conditions. The laws of statistics are different for quantum mechanics and classical theory. Under certain conditions, this can lead to gross macroscopic differences between classical and quantum theory. Before discussing an example of this, I want to show by means of a simple mechanical system – the force-free motion of a mass point – how the transition to the classical theory discussed above is to be formulated mathematically. The equations of motion, for one-dimensional motion, are:

Since time can be treated as a parameter (as a “c-number”) if there are no external, time-dependent forces, then the solution to this equation is:

Where \(p_0\) and \(q_0\) represent impulse and position at time \(t=0\) . At time \(t_0\) [see equations (3) to (6), let \(q_0=q’\) be measured with precision \(\Delta q\), \(p_0=p’\) with precision \(\Delta p\). If from the “values” of \(p_0\) and \(q_0\) we are to derive the “value” of \(q\) at time t, then according to Dirac and Jordan we must find that transformation function that transforms all matrices in which \(q_0\) appears as a diagonal matrix, into matrices in which \(q\) appears as the diagonal matrix.

In the matrix pattern in which \(p_0\) appears as the diagonal matrix, \(q_0\) can be replaced by the operator \(\frac{h}{2 \pi i} \frac{\partial}{\partial q_0}\). According to Dirac [i.e. equation (11)] we then have for the transformation amplitude sought, \(S(q_0,q)\), the differential equation:

Thus \(S\overline{S}\) is independent of \(q_0\), i.e, if at time \(t=0\), \(q_0\) is known exactly, then at any time \(t>0\) all values of \(q\) are equally likely, i.e., the probability that \(q_0\) lies within a finite range, is generally zero. This is quite clear, intuitively, because the exact determination of \(q_0\) leads to an infinitely large Compton recoil. The same would of course be true of any mechanical system.

However, if at time \(t=0\), \(q_0\) is known only to a precision \(\Delta q\) and \(p_0\) to precision \(\Delta p\), then by equation (3):

If we introduce the abbreviation

Then the exponent in (14) becomes

From which follows

Thus, at time \(t\) the electron is at position \(\frac{t}{m}p’ + q’\), to a precision \(\Delta q\sqrt{1+\beta^2}\). The “wave packet” or better, the “probability packet” has become larger by a factor of \(\sqrt{1+\beta^2}\) according to (15), \(\beta\) is proportional to the time \(t\), inversely proportional to the mass – this is immediately plausible – and is inversely proportional to \(\Delta q^2\). Too great a precision in \(q_0\) has a greater uncertainty in \(p_0\) as a consequence and hence also leads to an increased uncertainty in \(q\).

The parameter \(\eta\) which we introduced above for formal reasons, could be eliminated in all equations, here, since it does not enter in the calculations.

As an example that the difference between the classical laws of statistics and those from quantum theory can lead to gross macroscopic differences in the results from both theories, under certain conditions, shall be briefly discussed for the reflection of an electron by a grating. If the lattice constant is of the order of magnitude of the de Broglie wave length of the electron, then the reflection will occur in certain discrete directions in space, as does the light at a grating. Here, classical theory yields macroscopically something grossly different. And yet, we can not find a contradiction against classical theory in the orbit of a single electron.

We could do it, if somehow we could direct the electron to a certain location on a grating line and there establish that the reflection did not occur classically. But if we want to determine the electron’s position so precisely that we could say at which location on a grating line it would impact, then the electron would acquire such a velocity, due to this determination, that the de Broglie wavelength of the electron would be reduced to the point that in this approximation, the electron would be actually reflected in the direction prescribed by classical theory, without contradicting the laws of quantum theory.

According to the intuitive interpretation of quantum theory attempted here, the points in time at which transitions – the “quantum jumps” – occur should be experimentally determinable in a concrete manner, such as energies of stationary states, for instance. The precision to which such a point in time can be determined is given by equation (2) as \(\frac{h}{\delta E}\), if \(\delta E\) is the change in energy accompanying the transition. We are thinking of an experiment such as the following: Let an atom, in state 2 at time \(t=0\) returns to its normal state 1 by emitting a photon. We could then assign to the atom, in analogy to equation (7) the eigenfunction:

if we assume that the radiation damping will express itself in the eigenfunction by means of a factor of the form \(e^{-\alpha t}\) (the true dependence may not be that simple). Let us send this atom through an heterogeneous magnetic field, to measure its energy, as is customary in the Stern-Gerlach experiment, except that the heterogeneous field shall follow the atom beam for a good portion of the path. The corresponding acceleration could be measured by dividing the entire path followed by the atom beam in the magnetic field, into small partial paths, at the end of each of which we measure the beam’s deflection. Depending on the atom beam’s velocity, the division into partial paths will correspond, for the atom, also to division into partial time intervals \(\Delta T\). According to §1 , equation (2), to the interval \(\Delta T\) corresponds a precision in the energy of \(\frac{h}{\Delta t}\). The probability of measuring a certain energy can be directly derived from \(S(p,E)\) and is hence calculated in the interval from \(n\Delta t\) to \((n+1)\Delta t\) by means of:

We conceive of the experiment above entirely in the sense of the old interpretation of quantum theory, as explained by Planck, Einstein and Bohr, when we speak of a discontinuous change of energy. Since such an experiment can be performed, in principle, agreement as to its results must be possible.

In Bohr’s basic postulate of the quantum theory, the energy of an atom, as well as the values of the action variables \(\mathbf{J}\), has the privilege over other items to be determined (such as the position of the electron, etc.) that its numerical value can always be given. This privileged position held by energy over other quantum mechanics magnitudes is owed strictly to the circumstance that in a closed system, it represents an integral of the equation of motion (for the energy matrix we have \(E\) = constant). In contrast, in open systems the energy has no preference over other quantum mechanics variables.

In particular, it will be possible to conceive of experiments, in which the atom’s phases \(\phi\) are precisely measurable and for which then the energy will remain, in principle, undetermined, corresponding to a relation \(\mathbf{J\phi-\phi J}=\frac{h}{2 \pi i}\) or \(\Delta J \Delta \phi \sim h\). Such an experiment is provided by resonance fluorescence, for instance.

If an atom is irradiated with an eigenfrequency of say, \(\nu_{1\;2}=\frac{E_2 – E_1}{h}\), then the atom will vibrate in phase with the external radiation, in which case in principle it is senseless to ask, in which state – \(E_1\) or \(E_2\) – the atom is vibrating. The phase relation between atom and external radiation can be determined for instance by means of the phase relations among many atoms (Woods experiment).

If one does not want to use experiments involving radiation, the phase relation can also be measured by performing precise position measurements in the sense of §1 for the electron, at different times relative to the phase of the light used for illumination (for many atoms). To each atom we could then assign a “wave function” such as:

Here \(c_2\) depends on the intensity and on \(\beta\) the phase of the illuminating light. Thus, the probability of a certain position \(q\) is:

\left ( |\psi_2||\psi_1|e^{-\frac{2 \pi i}{h} [(E_2 - E_1)t+\beta]}\;+\; |\psi_2||\psi_1|e^{+\frac{2 \pi i}{h} [(E_2 - E_1)t+\beta]} \right )\)

The periodic term in (20) can be experimentally separated from the non-periodic one, since the position determination can be performed at different phases of the illuminating light.

In a known imaginary experiment proposed by Bohr, the atoms of a Stern-Gerlach atom beam are initially excited to resonance fluorescence, at a certain location, by means of light irradiation. After a certain length, the atoms pass through an heterogeneous magnetic field; the radiation emitted by the atoms can be observed over the entire length of their path, before and behind the magnetic field.

Before the atoms enter the magnetic field, they exhibit normal resonance fluorescence, i.e, in analogy to the dispersion theory, we must assume that all atoms emit in phase with the incident, spherical light waves. At first, this latter interpretation stands in conflict with what a rough application of the light quanta theory or the basic rules of quantum theory indicate: from it one would conclude that that only a few atoms would be raised to an “upper state” by the absorption of a light quantum and hence, that all of the resonance radiation would come from intensively radiating excited centers. Thus it used to be tempting to say: the concept of light quanta can be called upon here only for the energy impulse balance; “in reality” all atoms radiate in lower states as a weak and coherent spherical wave.

Once the atoms have passed through the magnetic fields there can hardly be any doubt left that the atom beam has split into two beams, of which one corresponds to atoms in the higher state and the other, to atoms in the lower state. If the atoms in the lower state were radiating, this would be a gross infringement of the energy theorem, because all of the excitation energy is contained in the fraction with the higher state.

Rather, there can be no doubt that behind the magnetic field, only the atom beam with the upper states is emitting light – and non-coherent light, at that – from the few intensively radiating atoms in the upper state. As Bohr showed, this imaginary experiment makes particularly clear how careful we must be with the application of the concept “stationary state”.

From the conception of the quantum theory developed here, it is easy to discuss Bohr’s experiment without any difficulty. In the outer radiation field the phases of the atoms are determined and hence there is no sense in talking of the energy of the atom. Even after the atom has left the radiation field we can not say that it is in a certain stationary state, if we are asking for coherence characteristics of the radiation.

But while experiments can be performed to test in which state the atom is; the result of this experiment can only be given statistically. Such an experiment is actually performed by the heterogeneous magnetic field. Behind the magnetic field, the energies of the atoms are determined and hence their phases are undetermined. The radiation is incoherent and emitted only by atoms in the upper state. The magnetic field determined the energies and hence destroys the phase relations. Bohr’s imaginary experiment provides a beautiful clarification of the fact that the energy of the atom is also, “in reality, not a number, but a matrix.” The law of conservation applies to the matrix energy and hence also to the value of the energy, as precisely as it is measured, in each case.

Analytically, the cancellation of the phase relations can be followed approximately thus: letting \(Q\) be the coordinates of the atom’s center of mass; we can then assign to the atom (instead of (19)) the eigenfunction:

The eigenfunction (21), however, will change in the magnetic field in a calculable manner, and because of the differing deflection of the atoms in the upper and the lower state, will have become, behind the magnetic field,

\(S_1(Q,t)\) and \(S_2(Q,t)\) will be functions in \(\mathbb{Q}\) differing from zero only in a small area surrounding the point. But this point is different for \(S_1\) and for \(S_2\), hence \(S_1S_2\) is zero everywhere. Hence, the probability of a relative amplitude \(q\) and a definite value \(Q\) is:

The periodic term in (20) has disappeared and with it, the possibility of measuring a phase relation. The result of the statistical position determination will always be the same, regardless of the phase of the incident light for which it was determined. We may assume that experiments with radiation whose theory has not yet been fully elaborated will yield the same results regarding the phase relations of atoms to the incident light.

Finally, let us examine the relation between equation (2), \(\Delta E \Delta t \sim h\) and a problem complex discussed by Ehrenfest and two other researchers by means of Bohr’s correspondence principle in two important papers. Ehrenfest and Tolman speak of “weak quantization” when a quantified periodic motion is subdivided, [by quantum jumps or other disturbances, into time intervals \(t\) that can not be considered long in relation to the system’s period. Supposedly, in this case there are not only the exact energy values from quantum theory, but also – with a lower a a-priori probability that can be qualitatively indicated – energy values that do not differ too much from the quantum theory-based values.

In quantum mechanics, such a behavior is to be interpreted as follows: since the energy is really changed, due to other disturbances or to quantum jumps, each energy measurement has to be performed in the interval between two disturbances if it is to be unequivocal. This provides an upper limit to \(\Delta t\) in the sense of §1. Thus the energy value \(E_0\) of a quantified state is also measured only with a precision \(\Delta E \sim \frac{h}{\Delta t}\). Here, the question whether the system “really” adopts energy values \(E\) that differ from \(E_0\) with the correspondingly smaller statistical weight – or whether their experimental determination is due only to the uncertainty of the measurement, is pointless, in principle. If \(\Delta t\) is smaller than the system’s period, then there is no longer any sense in talking of discrete stationary states or discrete energy values.

In a similar context, Ehrenfest and Breit point out the following paradox: let us imagine a wheel – for instance, in the shape of a gear wheel – fitted with a mechanism that after \(f\) revolutions just reverses the direction of rotation. Let us further assume that the gear wheel acts on a rack that can be linearly displaced between two blocks. After the specified number of revolutions, the blocks force the rack, and hence the wheel, to reverse direction. The true period \(T\) of the system is long in relation to the period \(t\) of the wheel; the discrete energy steps are correspondingly dense, and denser, the greater \(T\) is. Since from the point of view of a consistent quantum theory all stationary states have the same statistical weight, for a sufficiently large \(T\) practically all energy values will occur with the same frequency – in contrast to what we would expect for the rotating gearwheel. Initially, this paradox becomes even sharper when we consider our points of view.

In order to establish whether the system will adopt the discrete energy values corresponding to a pure gearwheel singly or with special \(f\) frequency, or whether it will adopt all possible values (i.e., values corresponding to the small energy steps \(\frac{h}{T}\)) with the same probability, a time \(\Delta t\), is sufficient, which is small in relation to \(T\) (\(T >> \Delta t\)). That is, although the large period for such measurements never becomes effective, it apparently manifests itself in that all possible energy values can occur.

We believe that such experiments for the determination of the system’s total energy would actually yield all possible energy values with the same probability; and this is not due to the large period T, but to the linearly displaceable rack. Even if the system should find itself in a state whose energy corresponds to the wheel quantification, by means of external forces acting on the rack it can be easily taken to states that do not correspond to the gearwheel quantification. The coupled system gearwheel-rack simply has periodicity characteristics that are different from those of the gearwheel. The solution of the paradox rather lies in the following: if we wanted to measure the energy of the gearwheel alone, then we shall first; have to dissolve the coupling between gearwheel and rack.

In classical theory, for a sufficiently small mass of the rack the dissolution of the coupling could occur without energy changes and therefore there the energy of the total system could be equated to that of the gearwheel (for a small rack mass). In quantum mechanics, the interaction energy between rack and wheel is at least of the same order of magnitude, as one of the gearwheel’s energy steps (even for a small rack mass, a high zero-point energy remains for the elastic interaction between the wheel and rack) Once the coupling is dissolved, the rack and the gearwheel individually adopt their quantum theory energy it values. Thus, to the extent that we can measure the energy values of the gearwheel alone, we will always find the values prescribed by quantum theory, with the precision allowed by the experiment. Even a minuscule rack mass will rob energy from the coupled system, and thus the the measured energy will different from that of the gearwheel alone. The energy of the coupled system can adopt all possible values (those allowed by time-qantification) with the same probability.

Perhaps the statement that the velocity in the X-direction “in reality” is not a number, but a diagonal term in a matrix is no more unintuitive and abstract than the determination, that the electric field intensity “in reality” is the time portion of an antisymmetrical tensor of the space-time world. The expression “in reality” is just as much or as little justified here as it is for any other description of natural phenomena in mathematical terms. As soon as we admit that all quantum theory variables “in reality” are matrices, the quantitative laws follow without difficulty.

If one assumes that the interpretation of quantum mechanics attempted here is valid at least in its essential points, then we may be allowed to discuss its main consequences, in a few words. We have not assumed that quantum theory – in contrast to classical theory – is essentially a statistical theory, in the sense that starting from exact data we can only draw statistical conclusions. Among others, the known experiments by Geiger and Bothe speak against such an assumption. Rather, in all cases in which relations exist between variables, in classical theory, that can really be measured precisely, the corresponding exact relations exist also in quantum theory (impulse and energy theorems). But in the rigorous formulation of the law of causality – “If we know the present precisely, we can calculate the future” – it is not the conclusion that is faulty, but the premise. We simply can not know the present in principle in all its parameters. Therefore all perception is a selection from a totality of possibilities and a limitation of what is possible in the future. Since the statistical nature of quantum theory is so closely to the uncertainty in all observations or perceptions, one could be tempted to conclude that behind the observed, statistical world a “real” world is hidden, in which the law of causality is applicable. We want to state explicitly that we believe such speculations to be both fruitless and pointless. The only task of physics is to describe the relation between observations. The true situation could rather be described better by the following; because all experiments are subject to the laws of quantum mechanics and hence to equation (1), it follows that quantum mechanics once and for all establishes the invalidity of the law of causality.

After closing this paper, new investigations by Bohr have led to viewpoints that allow a considerable broadening and refining of the analysis of quantum mechanics relations attempted here. In this context, Bohr called my attention to the fact that I had overlooked some essential points in some discussions of this work. Above all, the uncertainty in the observation is not due exclusively to the existence of discontinuities, but is directly related to the requirement of doing justice simultaneously to the different experiences expressed by corpuscular theory on the one hand, and by wave theory on the other. For instance, in the use of an imaginary X-ray microscope, the divergence of the ray beam must be taking into account. The first consequence of this is that in the observation of the electron’s position, the direction of the Compton recoil will only be known with some uncertainty, which will then lead to relation (1). It is furthermore not sufficiently stressed that rigorously, the simple theory of the Compton effect can be applied only to free electrons. As professor Bohr made very clear, the care necessary in the application of the uncertainty relationship is essential above all in a general discussion of the transition from micro to macro-mechanics. Finally, the considerations on resonance fluorescence are not entirely correct, because the relation between the phase of the light and that of the motion of the electrons is not as simple as assumed here. I am greatly indebted to professor Bohr for being permitted to know and discuss during their gestation those new investigations by Bohr, mentioned above, dealing with the conceptual structure of quantum theory, and to be published soon.

]]>For the past week or so I’ve been configuring my STM32F37x series microcontroller. Specifically one from the chip-tray STmicro custom-fabbed and hand-delivered to me (all I wanted was fedex ground!). This adventure has been one with fortune, and failure.

Delightfully this microcontroller was engineered proper, unlike the TMF320′s which gave me so much woe last month. That is to say, it responded to my initial st-link JTAG requests immediately and without issue, and hasn’t given any connection troubles thereafter. “Bricking it” does not appear to be a possibility given that all fuse bits are write-protected during its user-flash sequence, as well as its bootloader and other *really important things.* TI, please take a lesson from ST and write protect your non-program pages!

ST unfortunately does not maintain an IDE for their microcontrollers. Instead users are forced to use one of four commercial IDEs, the most notable of which are MDK-ARM (keil) and IAR workbench. I assume that this course of action is what allows [legally] for them to provide *so many* freeware peripheral drivers (many are in fact, developed by those who make these IDEs), but still, it’s a bit awkward having a 32k code limit on the freeware versions of such workbenches.

That’s where the fun appears to end, however. ST needs a librarian. Really. Or a new webmaster. After much struggle and somewhat-of an implicit “go to hell, plebeian” response from IAR support, I settled on Keil/MDK-ARM. While it’s UI leaves a lot to be desired, overall I’m quite impressed with its capabilities. Its compiler (not GCC!) compresses code like a champ, it has built-in core config files for nearly every ARM available, and its debugger is nothing short of amazing. I can snoop on local variables, even!

Although the company has many examples, datasheets and programming guides for their microcontrollers, it took communication with an engineer to find any of it. As one would have it, the collection of STM32F37x peripheral drivers was buried in a folder, inside a folder, inside a zip file, inside a lonely webpage, itself linked to only by one non-emphasized link in the “design resources” tab on the STM32F373RB’s product page. I kid you not.

It gets better though; as it turns out, while much documentation exists on the microcontroller’s busses, core, peripherals and their registers, *there’s absolutely nothing* about their C drivers. That is, nothing but the little /* @brief statements */ contained within them.

While inconvenient, this wasn’t a showstopper. Fortunately many of these helper functions directly correlate with the configuration registers documented in their big book, and every peripheral has several C examples to give the designer an idea of how to use them. That said, some functions do not work unless others are called beforehand, and it’s often a mystery as to what their proper order is.

Some tips, for my reader:

• *Enable your clocks!* Every peripheral has an RCC_blah clock command which must be run before the device will work, and some others also have a PWR_blah command that must be set before even that. Nothing throws an error if these aren’t turned on, and quite frankly it took me 4 hours to debug timers because of it!

• *Disable your watchdog*. There’s an unconfigured watchdog timer enabled by default in the option bytes, that will repeatedly kill your program unless it’s turned off. This took another 4 hours to debug.

• Use this tool to write your sys config file (for the stm32f37x only). The microcontroller won’t throw any errors if the sysclock is misconfigured, and again, this was a real bitch to debug.

• Be forewarned; despite what some of ST’s configuration programs (cubeMX, microXplorer) might imply, that *you cannot *arbitrarily define analog pins as differential pairs. Rather, the pins must be labeled in the form SDADCx_AIN#P, SDADCx_AIN#N, where # is some channel number. I’m going to need to fab new boards because of this. ಠ_ಠ

• This powerpoint has been the most useful piece of documentation I’ve found yet.

Some 30 hours after first contact I eventually did get things configured. As of now I have clocks running, timers counting and ΔΣADCs sampling. There’s still an ADC, USART peripheral, DMA and interrupt table to configure before I can *actually* start crunching vectors in \(\mathbb{R}^3\), and given the pace of current progress I expect that to take at least another week. Oh well.

However, I did run into an unexpected problem last night around 2. Evidently, *marketing engineers lie.*

As one would have it, all of TI/BB’s micro-tiny “rail to rail” amplifiers are in fact, not “rail to rail”. Rather, the ones I chose to buffer my current shunts (OPA244) bottom out at 100mV, and are nonlinear for a good 50mV thereafter! This means that my nice getcurrent_float() function bottoms out at 0.9A, and doesn’t start properly working until we pass 2.5A or so through the phase in question.

Good job Burr-Brown, you get a medal. Also included is one angry customer, who now has to redesign their board to include a negative-rail charge pump that should not been needed.

Oh well, such goes “progress”. Despite their idiosyncrasies I’m happy with the STM32 product line, and will do my best to master the machine.

]]>WE IGNORE THE DEPRESSING IMPORTANT THINGS AND BUILD STUFF ᕕ( ᐛ )ᕗ

So by forgoing sleep, how many things can we construct in one week? As one would have it, a good deal of things.

Ignore the wire-nut e-stop.

There’s really not a whole lot to see here. It’s little else but a laser-cut box, a contactor, a breaker, some buttons, LEDs and a phasing switch for operation on Y-only, or Y,U,V hookups. Yes wood is flammable, but quite frankly, there’s not much that can go wrong in a box like this.

If something does go wrong, I’ll be right there to act, anyway.

This one is a little bit more interesting. Sans one missing knob, it’s an optical (fibre) pulse generator for bigcoil. Dubbed, ‘the turboencabulator’, she’s capable of generating up to 300us pulses at either timed intervals, or to the tune of 4 channels’ worth of midi music [4-polyphony]. The four possible super-positioned notes, each run from a dedicated interrupt timer, are FIFO scheduled (that is, scheduled on a first-come, first-serve basis).

The turboenecabulator is powered from a lithium battery and may either enumerate as a USB device, or accept a legitimate DIN-MIDI signal from whatever source you so choose. There are still some bugs, but hey, *what do you want from one days’ work?*

It’s very inconvenient to split-screen my tablet while doing math homework. After all, we only have 1080 lines to play with here!

As a solution, I put together a USB-powered, second display for my tablet. It is an iPad Retina-display™®© (specifically, the Lp097qx1-spa2 from LG), powered from USB, and fed video through a specially-shortened displayport cable. Conveniently, the LG glass supports embedded displayport as its video signal input, so, this project was little else than a bit of level-shifting, and constant-current LED driving.

Sadly, I cannot run this panel at above 50% brightness or we’ll blow the polyfuse in my tablet’s USB port. But my, that glass is purdy nonetheless.

Advanced circuits managed to get my PCBs in by Friday, which of course means they’re populated by 3AM Saturday morning! Overall there didn’t appear to be many design errors with this run, and in fact I’m quite happy with how Brushbuster IV turned out.

We’re still waiting for my ST-Link JTAG programmer.

*UNFORTUNATELY,* Linear technology seems to have overstated the reliability of their LTC6802-1 DACs, that is, their “we can float them at 45V and use current to daisy-chain the SPI lines” promise.

I don’t know why I believed that; the first few moments of real pack loading made short order of the the ICs, and the $60 pads they were connected to. Dang.

Luckily, it wasn’t too hard to redesign the BMS to use the addressable LTC6801-2 ICs, and *real* optoisolators for their communication. I hope then, to order new boards sometime this weekend.

Bigcoil version two finally got tested -the good news is that she worked. The bad news is, we:

• Melted the capacitor-discharge relay.

• Replaced that relay, then melted it again.

• Broke the $2 AliExpress e-stop button.

• Tripped some GFIs, a couple times*

• Eventually, caught one of my sketchy ebay switches on fire, and by extension, exploded my gate drivers.

There are still issues that need to be solved :-3

* This one is interesting actually. In bigcoil, there is no, non-catastrophic method for GFI-tripping differential currents to form. It’s concerning then, why the GFIs actually tripped! It’s even more concerning, that try as we might, we were unable to trip the AFCI breaker on the circuit bigcoil was tested on! Cool protection products, Leviton.

Well that was fun.

Now it’s time for five exams. (•_•’)

]]>Modern physics, differential equations, vector calculus, AC circuit analysis and another physics course, for those curious.

It’s not that I don’t understand their concepts: allow me the use of a normed vector space, abelian under addition and most anything is possible. It just takes me a long time to do math -almost twice as long when I don’t have colored markers to keep track of things. Memorization, without binary relation to anything else, takes weeks.

In fact, I’m probably dyslexic. I’m also going blind in my right eye. More problems, to add to my list I suppose, which includes a chemically burned lung.

I spend all of my free waking hours, from 6:30 to 1, doing homework. After it’s finished, there’s no time to check over it as another class’s problem set isn’t yet done. I do that, then some more work, I hand it in, and it comes back covered in red ink.

“Make corrections”, some say. I would, if a DIY differential equation solver wasn’t due tomorrow. Or another 5 hour problem set.

I find it hard to believe that most people understand the scope of the work they ask others to do. It’s work I haven’t the time to do effectively, work that teaches me little, and brings about only feelings of discouragement and dismay, upon its return.

And all for what? A $16/hr job offer from another organization, where it’s likely that more people will tell me I’m wrong? To appease professors, so hell-bent in the ways they’ve been taught that even the idea of “vector division”, possible in 2D, and 4D spaces, or negative refractive indices is horrific in their eyes; even if it’s a valid solution to the problem? To please people afraid to read between the lines.

There is more to the world than most ever see.

It’s hard to find desire to keep myself on this path. Most I suppose, take solace in the fact that “a degree means money in the future”. They bandage their psyches with relationships, alcohol and fraternity life. How jejune.

I am tired of kids and their bullshit ‘life problems’. My car’s engine has seized, I can’t afford housing and quite frankly, I am disgusted with the ungrateful attitude most members of my age group promote.

The fact of the matter is, *I don’t care about this crap any more.* This is not how I learn. There is no pedagogical reason why 25 problems a week, per class is a good idea. I have no time to -study- ; to sit down, conceptualize and put into effect, the new techniques I’ve set out to learn. Memorization in preparation for testing, is not learning. Abusing alkaloids in an effort to stay afloat in the work they assign and maintain my measly 3.2, is not learning. Clearly however, it’s been proven time and time again, that such a system produce wonderful engineers.

Caffeine is a horrible drug. Alcohol is a horrible drug. It boggles my mind why these ones are uniquely legal.

I don’t forget the things I learn, ever. Just give me some time to learn them, please.

This life is unsustainable. I have no time to peruse the things I enjoy; the things that keep me sane (the segue project, for example).

I need to go out and do what I’m good at; I need to devote my time, to building something great. Governor Cuomo, expect a rather interesting entrant in your 43north competition.

“Little hellions kids feeling rebellious, embarrassed, their parents still listen to Elvis, they start feeling like prisoners, helpless.”

▯

]]>(or at least I try to convince myself it does)

Well, in it’s simplest case, it is the following:

3 coils, configured in either a wye, or delta arrangement, each having an inductance, and a mutual inductance with the other coils in the network. Surrounding these coils is an arrangement of magnets; rare-earth, usually, which create a static field upon which the stator coils interact. In a plebeian sense, it’s the ‘job’ of the stator, to generate a rotating dipole vector that the rotor will do its best to ‘follow’.

Now, because the total flux in the system is constant, it’s possible to write a kirchoff voltage relation for the stator:

Where ** VR** is the voltage drop due to ohmic losses in a coil,

Where ** VR** becomes a [resistance matrix] * [current vector],

This can be broken up further:

And, assuming that all of our stator windings are reasonably equivalent, and that our windings are magnetically distributed 120 degrees apart from each-other, we may simplify to a somewhat-nicer looking equation:

~Where~

R = the average stator winding resistance

L = the average stator inductance

M = the average stator mutual inductance

I dot = the time derivative of current

Vmax = the maximum back-emf that will be generated

Theta_e = the electrical angle which describes the current back-emf, and also, the currents I(theta_e) and Idot(theta_e). For aesthetic reasons, I do not write I or V as functions of theta_e, but they indeed are!

The above equation also assumes that back-EMF is generated sinusoidally. That is, when you spin the motor and look at the voltage across a stator coil, you’ll see a sine wave. Due to reasons I won’t delve into here, this **is not true for every motor*.*

With these substitutions made, it’s a bit more convenient now to smash everything back into as few matrices as possible:

Admittedly though, this equation still doesn’t do us much good. It requires explicit values for every current in the circuit, and who even knows what the value of “M” is?

What if however, we assume our currents to be balanced? That is, there is some relation Ia + Ib + Ic = a constant, which in theory, would let us simplify our matrices a bit.

Could this be the case? Well… certainly not if the motor is grounded! Just refer to the left figure, and you’ll soon see why.

( For those bad at where’s waldo: If a ground exists, Ia and Ib and Ic can all leak out through it, independent of each other! )

A simple solution to this problem is to *not* ground our motor. In doing so, we end up with the relation Ia + Ib + Ic = 0, which lets us make the following simplifications:

Which then becomes the rather trivial relation:

Look at that, our inductance matrix is now isomorphic to an inductance scalar. Hot damn!

There are still some more unknowns to kill, however. Specifically, what exactly is *Vmax*?

Well, Vmax, by our prior definition, is the maximum electromotive force generated by the rotor’s magnets flying past the stator’s coils. As such, per one of maxwell’s relations it’s a linear function of the rate of change of magnetic flux in the stator coil.

With this argument in mind and the constraint that no magnetic components are changing in physical size, Vmax must then only be a function of flux linkage and angular velocity of the motor!

And with that…

Look at that: a model which contains quantities that are either all constants, or physical elements we can directly measure. Nice!

But, this matrix representation is still not all that useful to us. I say this because it’s impossible to fully represent the model in this form, with conventional engineer’s tools like phasors. So, we need some sort of transform to take us from this balanced, 3-phase representation into something of the form e^ix, or restated, from a 3-tuple representation into a 2-tuple representation.

Thankfully, a lady by the name of Edith Clarke figured out how to do this in the 1960′s.

Yay, we just gave one component in our current/voltage tuples the boot!

Edith’s transform can be thought of as a geometric transformation: three vectors, rotating synchronously in space, are projected onto the complex plane. Assuming these vectors are “at theta=zero”, that is, oriented in such a way that vector A is co-linear with the real (alpha) axis, then we are left with a situation in where the two other vectors are pointing the other direction, and are offset 60 degrees from the real axis, or correspondingly, 30 degrees from the imaginary (beta) axis.

Now what is the real part of a 30-60-90 triangle on the complex plane? sin(60) = sqrt(3)/2.

What is the imaginary part? cos(60) = 1/2.

That in mind, take a look at the transformation matrix, and it all should make a lot more sense :-).

Now, we still have that ugly back-emf vector, **|cos(x), sin(x)|**. What can we do about that?

Well, in vector space, a 2-tuple is for all intents and purposes isomorphic to a complex number, which is also a 2-tuple. (read: “ordered pair”). So, with the convenient relation:

**e^ix = cos(x) + i sin(x)**

We can say that:

**|cos(x), sin(x)| for our purposes, is equal to e^ix**

*Ain’t that somethin? *

In hat notation, our equation is now beautifully simple!

Where **V_hat**, or any other similar vectors are equivalent to **( V_max e^i theta ).**

One might ask though, how is this two-dimensional equation useful in our 3-phase motor? What good does it really do us?

To answer that question, it does us good, because instead of three phase shifted cosines, we now only need to keep track of one sine, and one cosine component. This is a much easier proposition for DSPs to handle, and, because we used a linear transformation to get this equation, it’s possible to use an equally trivial transformation to bring us out of the complex plane, and back into three-phase space. Specifically like so:

**I_phase_a = 3/2 Re[I_hat]**

**I_phase_b = 3/2 ( -1/2 * Re[I_hat] + √3/2 * Im[I_hat] )**

**I_phase_c = 3/2 ( -1/2 * Re[I_hat] – √3/2 * Im[I_hat] )**

Where I_phase_n is a current, itself linearly related to the PWM value you’d feed to some leg of a mosfet bridge.

But that’s enough of this, let’s toss everything into mathematica to prove I’m not giving you crap!

Bam, would you look at that. Those curves sure look like a brushless motor to me.

With the proper constants chosen they should represent any motor, even a Segue motor!

It is often the case that something you discover during a project, completely invalidates all prior work on the project. Sometimes this may happen more than once.

After modeling my motors we soon needed to fill in the constants, flux linkage in particular. How does one go about finding that?

Well, you use lots of scotch tape, a power drill, aluminum foil, two LEDs and an oscilloscope.

SCIENCE.

Too much of it to handle.

Anyway… in this case, the maths above soon revealed that my 44V of lithium power was not going to be enough for Segue. That is, the back-emf generated by my motor will equal 44V, when Segue reaches a linear speed of 21 km/h. …w*hich is *not* 40 km/h!*

So ladies, gentlemen and children of the internet, I present Segue v5.

After getting fed-up with de-, then re-soldering hundreds of components every time a board was revised, I decided it was in my best interest to go back to the “multi-pcb-with-ribbon-everywhere” solution. That is, Segue will have 5 independent boards:

Cellsniffer– A 24 cell (88V) lithium-ion pack monitor and balancer.

S.T.S. Whitetail– A board full of buck converters, and an 88V 4A boost converter.

Brushbuster(s) 3– A new motor controller board, equipped with super-badass 8mohm 150V mosfets and a C2000 DSP

Seguebrain– A board with a bluetooth radio, an IMU and a cortex-M4, among other things.

“Lolomgwaffles it’s a penis” screams the internet. Yeah, get over it.

Now, as much as I would love to say I’ve gotten things working; IE, having a nice stochastically-spun motor available to show you, I don’t.

Why?

Because, programming a C2000 piccolo is evidently quite a pain in the ass. That is, code composer studio for some reason likes to hang during flash erases, which just so happens to brick my DSP.

I’ve yet to figure out why this is happening, but to be honest, I’m a bit fed-up with these ICs already. I mean really, *WHO STORES LOCKOUT PASSWORDS IN PROGRAM FLASH*. Is is really that much more expensive to have a 64 byte bank of EEPROM memory, explicitly for storing such configuration data? Evidently so!

Assuming all other TMS320 products do the same, remind me to *never* use a TI DSP inside an aircraft, or some other similar mission-critical system. I say so, because it would be really, really bad if you locked-up such an IC, buried deep inside an apollo-computer construction actuator control box and/or something of equal “this is going to be really hard to repair” stature.

“Oh hay guys, our JTAG flash bricked ur airplane”

ಠ_ಠ

]]>

As you may surmise from my website, I build things. I’ve been doing so since the age of 6.

Through the years elapsed, I’ve designed too many things to tally. Everything from particle accelerators and x-ray machines, to induction heaters and flashlamp-pumped lasers. I’ve designed robots, computers, and even thermal identification / body tracking systems.

I have a bad habit of not documenting most of what I design, but what I do happen to record eventually finds its way onto this website.

I’m currently enrolled in the physics program at the Rochester Institute of Technology, where I’m continuing my experiments and private research. Lately, this has been in the domain of robotics and cutting-edge power electronics.

I’m a thrice-failed entrepreneur, and founder of the RIT makerspace.

Please enjoy my varied musings,

-Adam

]]>Let me elaborate.

I credit most of my design ability to time spent in multiple disciplines. Drawing, sculpture, mechanics and photography among others, have all had nothing but positive impact on my ability visualize and design systems. To extent that for those without such experience, I fear for their abilities to design similar systems.

Restated, it is immeasurably more likely that an electrical engineer who has fixed cars in the past, will know what to expect from a motor they wish to integrate in their product. It’s fair to argue that an industrial designer who has been exposed to power electronics, knows what’s practical to fit inside such a thing as a powered cell phone case, and may even attempt such a project. It’s fair to argue that even musicians who have studied physics, are more likely to have less reflection and echo in the things they record.

It’s possible to propose many further scenarios, but what I state is, by forcing students through highly tailored paths during their undergraduate careers, we set them up for failure when it’s time for them to innovate. How can a mechanical engineer be expected to design a new powered scooter for kickstarter, if they know nothing about electronics? Similarly, how can a physicist be expected to create a new particle detector, if all they have been exposed to is mathematics? Especially so, when neither party knows where to begin in such studies!

My philosophy is a simple one; do your best to gain a plebeian knowledge about everything there is to know, past and present. When a problem arises ask upon this resourceful memory of creative solutions, and if something looks promising, go study it in further detail. With the tools we now have available as a human race, we’re beyond the time where an engineer must be able to recite from memory, all there is to know about control theory, or where a photographer must know the granularity of every film available.

I’d much rather have engineers who know the physics of camera obscura, and photographers who know how to build high-speed triggers from a microcontroller.

That is, I’d much rather have a world where people put forth effort to become jacks of all trades, if they choose to master in one.

]]>For the past year or so I’ve been working on and off on Segue; the world’s first DIY, brushless motor segway-robot-thing. My end goal is to design a Segue kit of sorts, to offer to all the ambitious nerds who would like to have their own DIY, aluminum-frame self-balancing transporter (for in theory, $1700).

This is how that’s thus far happened.

Someone stole my bike on a cold friday night. This did not sit well with me, so I did the only logical thing I could do at the time: I built a segway.

Fortunately I was able to find on campus;

- An old power scooter
- A gyroscope
- An accelerometer
- An arduino
- A piece of plywood
- An aluminum tube
- A machine shop with a broken door lock

Monday morning, I had this:

Oddly enough, my evil plans worked. This doomsday device worked well enough to ride around campus for several weeks, and it probably sealed my reputation as a nerd at this university. Awesome!

But, “Segue” had several shortcomings:

- The motors were too slow to travel at speeds greater than 4mph, which made me sad.
- The motors’ gearboxes chattered… a lot. Often, this led to some interesting, positive gain oscillations!
- Lead acid batteries + winter = bad batteries in a week.

…among other things.

So I set out to build a new Segue, one that wouldn’t suck too much to commute around campus with. It was to have;

- Brushless motors
- Lithium batteries
- A metal frame with a “bad-ass” factor > 1.0
- Really, really good system control.

This turned out to be harder than first expected.

First, I needed motors. Specifically brushless hub motors with a single, thick stator shaft. Where oh where, can one expect to find such a thing?

As one would expect, only in china. One month of negotiations and a $500 paypal transfer later, DHL showed up with my new toys!

Soon thereafter the next question was “what should Segue 2 look like?”, followed soon by, “how would I be able to manufacture more, if others want one?”. The answer to both of these questions, was immediately obvious:

–> Water jet cut industrial aluminum (⌐■_■)

A week of yelling at autodesk inventor, and about $300 at Klein steel and Nifty-Bar inc provided me with something worthwhile.

This, specifically! A sturdy, waterjet-cut frame held together with interleaving aluminum tongues, and friction through an interesting bolt-nut arrangement. Though that might sound worrying, this proved to be quite strong indeed!

The idea behind this frame, is to require nothing but a wrench and a screwdriver to assemble it. This is so because tig welding for assembly of a kit isn’t quite my idea of a weekend project, nor do I expect it to be anyone else’s.

Now, it was electronics time. It’s fair to argue that it’s still electronics time, but let’s not go there just yet.

The question to ask again, was “what does a Segue need?”, soon followed by “how can I do that, without spending $2,000 on mouser?”.

A Segue has batteries; batteries need a battery management system. A Segue needs a computer, it needs an inertial sensor and it needs power supplies for such. A Segue needs two motor controllers, and three FET half-bridges on each of those. We found that a Segue needs at minimum, $100 worth of electronic components –which isn’t too bad.

I won’t go into much detail on the early prototypes. It should be rather obvious, what I’ve done there from this next photo, so here it is:

I will point out though, the following decisions:

- The main computer is a Teensy 3.0, which itself is a cortex-M4 running Arduino. I chose to use this because I want Segue 2 to be easily hackable: c purists, go home.
- I chose to use a finite-state machine as my motor controller. This was a bad decision. I probably should have foreseen so, but alas, I assumed it might work anyway. Nope.

That is not how we Segue around these parts.

Board two combined everything into PCB. The state machines were dropped in favor of a nice BLDC motor controller IC, bugs in the battery management system and SPI bus were fixed, the power supplies were made a bit more reliable and layout was generally improved. The mosfets no longer suffered from random explosions.

I seem to have lost the photos I had of this board, but once again, we had motor control issues. Namely I learned from this design, that commercial motor control ICs suck something fierce!

I learned also, the importance of proper inrush current limiting!

Board three dumped the motor control IC in favor of an AVR32. Inrush current limiting was added, along with a proper fuse, and a bluetooth radio, once I realized I could use a smart watch to steer the Segue!

This is the board where I learned that AVR32 is a mess beyond comprehension, where somehow Atmel took the harvard architecture and broke it.

There are 9 ways to multiply two numbers in that instruction set. THIS IS NOT USEFUL. **(╯°□°）╯︵ ┻━┻**

AVR32 was soon abandoned in favor of TI’s C2000 Piccolo series DSP’s. They are in fact real RISC machines; and I can understand their command set which is just lovely when I’m trying to use them.

Final exams haven’t given me time to play with these just yet, unfortunately…

The main goals of the project as of now are to design a reliable, self-balancing segue-bot with:

- A frame supporting two 2kW brushless hub motors, both providing in theory, a top speed of 25mph.
- A 500Wh lithium ion battery pack, with associated charging, and battery management systems.
- A bayesian inertial measurement engine, to provide accurate ground normal, velocity and acceleration measurements for the rest of the system.
- On board, electronically switched power converters, to provide for things a 5V, 15V and 3V rail from the 51V battery stack.
- Bluetooth connectivity, for hands-free steering with a smart-watch.
- Suitable power electronics, for heat-free switching of the brushless motors’ stators.
- A dead-reckoning, bayesian-corrected motor simulation engine, to allow for “0 speed”, constant torque brushless motor driving. Gimballing, if you will.

Of these goals five have thus been accomplished. The frame is built, the battery management system is done and does not explode, the power converters are working reliably from the 51V they’re given, the bluetooth works, and the MOSFET buffers work and don’t require heatsinking. That took the whole of last summer!

What’s left to do at this point is nothing but math.

Brushless motors are tricky devils. In order to spin one’s rotor, you must provide with its three stator coils, a rotating magnetic field. That in itself isn’t too hard:

Stator n = A cos(0 - ωt + Φ)

Where each Φ is offset either 60 or 120 degrees from the last coil’s, depending on the design of your motor. Advancing t would spin the motor in one direction, regressing t moves it the other direction. ω is simply your angular frequency of electrical rotation, which itself is related to the motor’s physical rotation by some constant.

Simple, right? Well… not quite. Physics likes to screw things up.

A is the maximum amplitude of your wave function; the case of DSP control, this would be the maximum PWM duty cycle of your mosfets, and thus the maximum current given to the stator winding in question. There, lies one problem: voltage and current aren’t necessarily aligned in the motor. Why?

Think about a loop of wire moving in a non-uniform B field.n As flux lines cut this loop, an emf will be generated in it as per the maxwell-faraday equation:

Notice something though; in no way, does current play a role in this guy! That is to say, the emf generated in the loop of wire, moving in a B field, is *independent* of the current in the wire.

Of course if we want to push back on the B field as we do with a motor; one needs a current in the loop. This is so, because of the Lorentz relationship:

Where (charge * velocity) = (Coulombs / Second) = current. Note here, that voltage plays no role! Ruh roh.

As non-intuitive as it may seem, a voltage generated, and a current generating a force, in a wire with no resistance, are two entirely different phenomena. Ain’t that something?

But here’s the kicker: see that little (-) sign in front of d phi / dt in the maxwell-faraday equation? That’s the little bit of hell right there which makes this job hard. It’s of opposite sign, of the voltage we need to apply to our real, restive wire, to make current flow in the direction we need. Clarifying:

Spinning the motor generates a voltage of opposite sign as the one you are applying to spin it. This is velocity dependent.

So that means A in our equation, is also a function of ω. Not only that, but it is a function also of the motor’s stator resistance and B field strength, where these themselves are even functions of temperature! Delightful.

There is no general solution to A, as it will be different for every motor. But, it is typically an empirically derived function of ω, which will do its best to scale I such that we can properly offset the back emf of the motor, such that the system becomes linear.

Got that? Good. There are more wrenches to toss in!

Magnets are not of infinite strength, and motors are not perfect. Thus, it’s possible to force them in positions no one likes; positions where the magnets we’re trying to move are no longer aligned with the field of our stator. It’s possible to lose track of them, for all intents and purposes.

To avoid this, ω needs to be correctly adjusted, such that if the motor is decidedly slowing down and we’re unable to do anything about it, that electrical RPM’s stay synced with the mechanical ones [w_m]. If they don’t, then the game’s over.

So now, we need to make ω a function of ω_m. Ok, but what is ω_m?

Therein lies big problem #2. I don’t know.

Mechanical RPM is difficult to measure at low speed. This is so, because nearly every velocity sensor in existence, relies solely on either:

- Discrete reference marks passing by some point, (hall effect sensors, rotary encoders, etc)
- Non-discrete references, such as the voltage generated in a wire, as a magnet rotates over it.

As one would have it, both fall apart at low speed. Discrete references don’t pass by fast enough to properly infer some (delta reference) / (delta t) value, and non-discrete measurements such as the one described, become signals too small to measure. s a bonus, how the hell am I to define what ω_m is at zero speed?

In reality we need a rotor position measurement, and a really precise one to boot. That’s expensive!

Expensive doesn’t work on a DIY segue-bot kit. It’s the same reason why I’m not just using a small motor and a planetary gearbox to begin with.

I have hall effect sensors; 6 of them. I have a motor whose flux linkage, stator resistance, and rotor field strength I can measure, and whose stator currents I have PWM control of, and can measure with kelvin sensing. I have Maxwell’s equations, Wikipedia articles describing Bayesian statistics, and 6 weeks of winter break to figure all that math out.

Let’s do this:

With a known flux linkage, moment of inertia, B field strength and stator resistance, I can run a simulation of my motor inside the DSP. From this, I can know its momentum, and guess how it should respond to a given current. Based on its current angular velocity, I can then scale this current value to compensate for phase lead / lag. Then, based on some a priori measurement, give the mosfets some (ideally correct) PWM values, and increase/decrease that duty cycle based on the phase’s current sense resistor.

The angular velocity will be found through 1), what I’m modeling the motor to do and 2), some correction, based on information to from the hall effect sensors. At zero speeds, I’ll just “gimbal” the motor and hope for the best. In theory, its application (balancing the robot) won’t torque the motor to such an extreme, that I will skip commutation steps.

So in essence, I need to cram:

Hall effect sensor information

Current information

A priori knowledge

A desired speed — Into —> [ Magic filter ] —-> [ PWM value register ]

A desired torque

Maxwell’s Equations

Ohm’s Law

All in some way that makes the DSP not catch fire.

Hoboy.

]]>