Hi TwoTired
I did not expect that generating efficiency would be linear with increasing rpm. But I find it surprising the alternator output would not keep increasing past 5,000 rpm.
DC generators have a more linear-like power output. Alternators, not so much. Particularly the SOHC 4s without slip rings, as the rotor magnet power is induced rather than directly energized by a winding for each specific pole piece. The alternator charging output pulses (six of them) Are sinusoidal. They rise from about .7 V to peak value in a rising then falling arc. As the rotary speed changes, the distance between the pulses become narrower, so more pulses in a given time frame occur to increase output power. At idle, the pulses are a farther distance apart, so there is less time when the actual voltage output is above 12V than when the engine is reved up.
The induced magnet strength of the rotor gets weaker at some given RPM (I expect 5000), above which point there is not enough time in the magnetic field of the coil to attain maximum pole piece strength. I suspect this is why Honda evolved their charging systems to Slip rings for the 650 versions. And permanent magnet types in other models.
The CB550F parts book shows the alternator parts with 323 numbers, CB500. I wondered if there was a later set of parts.
I don't think so. I've not encountered any.
As to using a fixed voltage to make calculations. I suppose you could do a table with different voltages, but why. The only current draw I question is the turn signals. Assuming there is sufficient current to get the flasher to flash, remind me to fix the ground on the left front turn signal, this current will only be drawn part of the time. I expect the amount of time out of the cycle that current is flowing varies to much to come up with a percentage.
The stock signal flasher starts on and won't interrupt current flow until enough power is drawn to bend the bimetalic strip in the flasher. When the voltage gets low, the lamps come on but don't blink, because the bimetallic strip never gets enough heat. So, the duty cycle of the signal lamps can be 100% when the battery is low. If you put in a flasher of a different type, (besides adding another unknown variable to the calculation), you can reduce the signaling drain to whatever your flasher provides. 50%??? But if you still have the turn beeper, you still have to add that current during the off time of the signals. (whatever that is)
My question really was does the 550 alternator produce more power that a CB400F1 alternator. As I have measured about 160 watts output from the alternator of my CB400F1. The answer is does not look like it. I have a 1977 CB550 K3 that I have not heard run and should hear run. I will check it's alternator output one of these days and compare it with the alternator output of my CB400F1.
I have not measure the 400F output. But, given the electrical resistance parameters, and the relative size of the power producing components, I would expect the 400F to be slightly weaker on average across production samples than the 500/550. It would not surprise me to learn that some 400F alternators made more power than some 550 alternators. But, most, I expect, would not.
The spec numbers in the books are not absolute. It should be a minimum spec. In order for vendor samples to meet spec and still have a production tolerance, most will work slightly better than spec. Say it is determined that a given recipe for making alternators has a 10% variability. Then you must alter the design to average 106% higher than spec. Some alternator samples will then provide 116% of spec, and some will only provide 101% of spec. But, none of your manufactured components should land in the scrap heap by design. Only if the build recipe is altered.
Another alternator output question. If say I have an alternator that produces 150 watts at 5,000 rpm. Will the alternator only put out that amount of power at a certain load or different loads? Say 12Volt X 12.5 amps or 15Volts X 10Amps.
Watts law is I x E = P. The voltage and amps measurement are directly relational. But, I must caution you that these numbers are delivery measurements and are highly effected by demand. To attain the max power output you must have enough circuit resistance to allow for max delivery. Further, if you place too much circuit resistance on the Alternator, the sinusoidal peaks cannot reach full excursion before being depleted. This will appear to look like an an out of spec unit if not tested correctly. Like with a connected battery as a buffer to the load resistance.
After I logged off and walked out the door I realized that there would be about a 3 Volt difference in the field voltage. Clearly that is going to reduce the power output of the alternator. I think that some voltage below the voltage were the voltage regulator cuts in say 14 Volts would be my standard for testing alternator output.
Absolutely correct! The field voltage has direct influence on alternator output capacity. The Honda shop manual shows 14.5 V as the full battery voltage cutoff. But, guess what? Another complication is that to prevent overcharging the battery, the regulator purposely reduces the field voltage, thus reducing alternator output. Generally this means the alternator cannot supply the bikes total needs and the battery begins to deplete. At some point the regulator notices and turns the Field back to full until the battery again reaches 14.5v where the cycle repeats. The frequency of this switching depends on vehicle resistive load, physical plate size of the battery, and the reduced power output magnitude of the alternator. (more calculations!!!). The trap, here is the test equipment chosen for monitoring. Analog meters deflect at the average DC level. So, it probably won't see a 14.5V peak. It will see something less, even though there is an AC component riding on the DC level. Digital meters may not be much help either, for they not only average, but they sample in very small time increments, that are unlikely to coincide with oscillation rate of the variable regulator cycling. Digital meters can also have a display update rate than can differ from the sample rate AND the Regulator cycle frequency. This all leads to what is called aliasing, where the displayed measurements are not actually a true representation of circuit activity. And it is something to be alert for when checking DC circuits that have AC elements introduced.
I think a lot of people would be surprised if they looked at the charging system component connections with an analog oscilloscope.
Hope this helps your investigation.
Cheers,
