I managed to complete a small portion of the testing I planned for the modified alternator. Unfortunately testing came to an abrupt halt when I tried for 3000rpm and a magnet came loose
Here is the test rig I am using:
<BR<p>
All results are the combination of all three phases rectified to produce DC. RPM is determined by measuring the frequency of one phase using a Fluke Industrial Scopemeter. As there are 12 poles on the rotor, 1 revolution produces 6 complete cycles - so I can simply multiply the frequency in Hz by a factor of 10 to get the rpm. In most cases I have tested up to 2000rpm.
As I my meter will only go up to 15A, I went all out and obtained a high precision, high current shunt:
As you may have picked up from the above image, my shunt consists of a 3.5" nail... Testing up to 15A in conjunction with the meter enabled me to calibrate the 'nail' - remarkably consistant and linear (up to the limit of the meter anyway). I have interpolated from these results to determine higher currents.
I have only managed to get results for the alt connected in series 'Y', along with one run in parallel 'Y'. I did not get to test the alt in a delta configuration.
As you can see from the results, the use of the field coil in conjunction with the permanent magnets produces significant changes to the power output. The one run I was able to complete using 24V on the field coil initially maxed at 250rpm because the drive belt started slipping. After retensioning the belt, scuffing up the drive wheel and applying spray on adhesive, I managed to get the alt to spin to 1000rpm. For the next round of testing I will use two belts. This test indicates that 24V on the field coil gives approx 50% more power over magnets alone. -24V on the coil causes output to fall by approx 75%.
During testing I ran the alt to 2000rpm many times with no trouble. The one time I got cocky and went for 3000rpm, a magnet let go and put an end to my fun. I have not stripped the alt down yet to investigate the damage, but from what I can see, it doesn't look too bad. It will however be next weekend before I can repair it and hopefully complete my testing. I plan to test the alt wired in delta, series/parallel, -24/0/+24 volts on the field, and also on some varying loads (including a 24V battery)
