Hmm...based on the above comments, maybe a 'refresher course' on Kettering design is appropriate here?
The beginning rule: The spark Duration must match the burn rate of the fuel at the compression used. A longer spark that this is a waste of [production engineer's] money (larger, heavier coils are then needed) and a shorter spark can both cause incomplete combustion and enhance ping by letting the unburned fuel coat the cylinder surfaces, making subsequent burn cycles momentarily too rich. This is also often perceived as a backfire in engines like the SOHC4 where a waste spark also occurs while both valves are open.
Honda apparently determined that (like their previous engines) a spark duration around 1.2-1.4mS was about right for the cam timing and RPM range desired: they had TEC design a coil for them with 2 output leads, very similar to the uber-successful CA72/77 design, but smaller and less expensive. The same condensors (0.24uF) were to be used: in order to generate the desired 7500 volt spark at 1.3mS (nominal) with 2.5 amps of current, the Q of the circuit was thus set. Interestingly, the Q of the early (pre-750F) coils is slightly higher than the CA77, probably due to newer winding equipment and the smaller, molded, bodies. As only about 4000 volts is needed to fire these engines, the 7500 volt value is adequate with plenty of margin, allowing solid spark over a wide RPM range (like idle thru 10k RPM) with no "tricky" parts required.
So...once the Q is set, the coils' winding inductance is set according to the size of the magnetic core used, and this determines the number of (wiring) turns needed. To reach the desired charge rate, the wire size is adjusted to make the final (primary) resistance come out to the value needed to consume the power available: this is about 2.6 amps/4.6 ohms on the early 750. This primary resistance controls both the charge and discharge rate of the coil, which is set to meet the desired RPM range (10 to 167 Hz for these revs). This part is simple enough.
The condensor: in the 20th century, almost all automotive condensors were made the same: 0.22-0.26uF, 400v, waxed paper or mylar windings - this design came from the (first successful) 1920s Ford system that the entire world's automotive industry adopted. Only BMW (cars) used a different value, and those where (in today's money) about $200 each, so Honda went with the 0.24uF condensor for economy (no-brainer, there...). This then sets the spark gap in the sparkplugs, to generate a kickback frequency via the high-voltage windings where this value of condensor can be used to set the resonance of the whole LRC circuit. This in turn sets the number of windings allowed/required between the primary and secondary layers in the coil while still achieving the desired spark voltage: and this, in further turn, sets the size of the wire used in the outer windings, and how thick the enamel insulation of this wire must be to make the efficiency high. If the wire is too small, the Q goes up too high and the coil puts out poorly at high and low RPM: if the wire (or insulation) is too big, the Q falls low and the output is weak all across the RPM range. ALL coils suffer this "fight" in their design: in the era of the Kettering Ignition, the only consistent "setting" was the 0.24uF condensor value, so most coils were made to function with it, one way or another.
Now comes the "tuning" part of it: in the CB500 (and all subsequent smaller SOHC Fours) Honda desired more spark voltage at the upper RPM, because the volumetric efficiency of the CB500 engine fell into the above-5000 RPM range. They needed more spark energy "up there" than before (compared to all their earlier engine designs), and the 500/550 also suffers from a tendency to "back-burn' up the intake tract during overlap, due to the long intake runners they used to flatten out the torque curve. So, a weaker spark at low RPM became desirable, and the Q was raised slightly by reducing the number of turns and the wire size in the secondary windings: starting in the 750F0 model, with its higher-RPM cam, they also used those same coils. These are noted by their lower primary resistance of 4.4 ohms (due to larger primary wire being used) and lower secondary winding resistance as well.
Higher Q means higher spark voltage (when running in the resonant RPM range), but it also means weaker starting-speed voltage and shorter discharge time (duration). To improve upon this, Honda changed the resistance in the sparkplug caps. In the early 750 the caps were 7500 ohms: in the CB500 and CB550K1 they were also this value. In 1974 Honda switched all of the Fours to 10,000 ohm plug caps: this made the 750K4/5/6 start better than it otherwise would have, but left it a little shy of spark above 9000 RPM. This was one of the 'detuning' items (along with 3 others) that prevented those bikes from reaching 100 MPH reliably, where the earlier bikes easily passed the Ton. These caps also burned out faster than the earlier ones, because they heated more and because their internal construction was not bonded, but was a carbon "button" captured between 2 copper contacts that burned quickly.
Enter the Dyna hi-output coils: riders not knowing about the plug-cap issue swapped in the 30Kv (3 ohm) coils with (by then) worn-out 15,000 ohm plug caps, and voila'! they got better spark (which they would have gotten less expensively by buying new plug caps...) and the bike(s) ran better, so long as they could feed the extra current to those coils (i.e., only the 750 was successful at this) on a regular basis. BUT...those of us 'in the know' could tell whenever one of these bikes rode by: the AM radio, television, and CB Radios went nuts because the coils were discharging so fast that they made good radio jamming systems.
They generated so much EMI and complaints from the general public that Congress (DOT) passed new laws against the Japanese motorcycle industry (effective circa 1980) to force them to "quiet" the motorcycle ignition systems. This led to the (less-than reliable, not fully tested) early CDI systems found on Hondas, Suzy's and Kawis of those days. The technology caught up with it by about 1988, but in those in-between years the bikes suffer reliability issues with the spark systems.
By now, the non-engineers out there have glazed-over eyes, so I'll quit...
