This article is in four sections.
Emisson-Absorption-Scattering (EAS) Sub-Quantum Physics
EAS Nuclear Glue
EAS Neutron Beta Decay
EAS Mass Excess and Mass Defect
EAS Mass Excess and Mass Defect
Installed sometime prior to 4 Mar 2001. Latest update 19 Feb 2017.
Serious reconstruction underway. Please drive Carefully.
This page started as an explanation for the  mass excess of a neutron over the sum the masses of its ponderable decay products, i.e., a proton and an electron. It is being modified to provide an EAS rationale for nuclear mass excess and mass defect.
On atomic and nuclear scales we deduce mass electromagnetically. That is, if we want to know the masses of specific nuclei or of electrons, we subject them to known electrical and/or magnetic fields and by observing their trajectories or their changes in velocity we calculate or deduce their masses. We do not actually measure ponderable amounts of "mass."
Since neutrons are electrically neutral we "measure" the masses of differing isotopes of an atom, then deduce neutron masses based on the differences and energy input/output considerations.
The EAS model of sub-quantum physics holds that the visibility of a given nucleus or ion, with respect to its electrodynamic environment is inversely proportional to what we call its mass. The model also holds that the superposition principle (for calculating electrical forces in many-body systems) does not strictly apply. In the opinion of the author, superposition theory is a powerful tool for first order effects, but in a many-body scenario it has a tinge of action-without-reaction. It says that electrical charges can be affected by an electrical field without having any effect on the field.
In the EAS model, like charged particles are opaque to one another. For example, if a positive chargelet, which could have hit a given proton, doesn't reach it because it was scattered by an intermediate proton, then a degree of electrodynamic shielding would have taken place. The opaqueness of like charged bodies for one another thus produces a reduction in the effect of the external environment on systems of charges.
If a given nucleus or ion is less visible than superposition theory predicts, then its response to electrodynamic influences (electrical fields for example) will be reduced. It would behave as though it were more massive than expected. This would be a case of mass excess.
If  a nucleus  is more visible than superposition theory would have it, then it will respond more readily to its electrodynamic environment than expected, and thus would appear to be less massive than expected. This would be a case of mass defect.
(The problem here is that I can see no way for a nucleus  to be
more visible to the environment than superposition allows. This
leads to the possibility that mass deficit, as just defined,
may not exist.) I have just come to realize that my definition of mass excess
and deficit are not consistent with conventional usage regarding nuclide masses.
Essentially, I was setting the proton mass to be 1.00000 a.m.u. instead of
carbon 12 being 12.00000. -
Inside complex nuclei the nucleons compete for attention from the outside universe. If that environment, carbon 12, for example, is our starting point, then the nucleons in less complex nuclides would tend be closer to free access to external chargelet fluxes. They would be more electrically responsive to the external environment and therefore would behave as though they were be less massive than those back at our starting point. The general tendency would be increasing mass defect with decreasing Z (or decreasing mass excess with increasing Z. Nucleon traffic patterns (including traffic jams) inside nuclei should affect total nuclide visibility to the external world. [Added 19 Feb 2017.]
Lets talk about measuring the masses three different particles:
a: An electron
We inject each of these objects at a known low energy at one end of a linear accelerating cavity and let the electrical field impressed across the cavity do its work. Each object's exit speed will mainly be a function of its charge, its' "mass" and the voltage applied.
In the EAS model the electrical accelerating influence of an  electrical field is brought about by collisions of negative or positive chargelets with the test charge. In the acceleration cavity there will be a net flux of negative chargelets traveling from the negative electrode to the positive electrode and likewise a net flux of positive chargelets traveling from positive to negative.
In this model protons and electrons are semi-opaque to the passage of positive and negative chargelets, which are the neutrino-like (or virtual photon-like force carrying particles. . . .
Deuterium nucleus acceleration
The "push-of-war" will  be won by the two protons.
The following paragraphs will have to be reworked regarding how much shielding unlike charges provide to one another.
If we focus on just the neutron part of the deuterium nucleus (by "neutron" I mean the proton that happens to have the electron at the moment) the electron will be out of sight of the accelerator's net negative chargelet flux roughly half the time, because it's shielded by the proton. (It is assumed that the electron is of smaller radius than the proton, This "blockage" is equivalent to having the electron's electrical response to the external field reduced to near zero half the time. The electron will thus experience half the expected linear acceleration from the applied flux. This is like having a proton and a "half time" electron for acceleration purposes. If the electron is being acted on half the time by the impressed electrical field then it will accelerate half as much.
Again, people generally "infer" mass from the observed accelerations imparted to objects when we use known (electrical, magnetic, or gravitational) accelerating influences. If we get reduced accelerations for an object, and all else that we think we know remains unchanged, we interpret it as increased mass. (Our currently accepted paradigm says that charge is constant and that no shielding takes place. Superposition reigns.) In the case being discussed, the "nuclear electron" thus is observed to behave as though it were twice as massive as a free electron.
I did not do my homework before writing the following paragraph. The mass of a deuteron is less than the combined masses of two protons and an electron, or of a proton plus a neutron. Mass-defect at work! See Philosophy paragraphs above for where this is going to go. [This cautionary note added 8 September 2005.]
The electron will provide some blockage of positive chargelets that would ordinarily collide with the protons, so the "collective" proton responses to the applied electrical field will be less than that expected based on the non-shielding aspect of the superposition theorem. (The protons themselves should fractionaly shield one another as well. The deuteron (two protons plus nuclear electron) will "measurably" seem more massive than expected. The amount of this "increase" will be a proportional to the ratio of the cross-sectional area of the electron to that of the protons. The trick will be to refine the hypothesis so as to make the total apparent increased mass, electron and protons, come out to be equivalent to 1.51 electron masses.
The electrical blockage (or electrical shielding) discussed here may be applied to more complex nuclei. The EAS model may provide an avenue for evaluating the peculiarities of nuclide mass relations.
Perhaps this Thayer Watkins study can bail me out on this mass deficit problem (or make it worse).
A Possible Error in the Mass of a Neutron and Its Implication for the Binding Energies of Nuclides - [Added 28 Sep 2016.]
Atomic Mass Adjustment - G.Audi, A.H.Wapstra and C.Thibault - Nuclear Physics A729 p. 337-676, 22 Dec 2003.