2 PMAs tested today.

[Jerry]

I set up a test bench in the shop today. This was to make a compairison of Hughs 4 ft PMA and the Garbogen. Both 12 volt machines.

Hughs is wired as per his instructions (star).

Test equiptment.

Fluke trure RMS meter measures PMA DC output.

Analog 10 amp meter measures PMA output.

HF cheap meter measures prime drive drive voltage.

HF clampon AC amp meter measures prime driver drive amperage.

Inferred digital tachometer.

5 KW 120 volt varyac, supplies 0 to 130 VAC.

DC power supply, using 2 each 50 amp X 400 volt fullwave bridge rectifiersand 10 each 6,600 UF/450 volt elactrolyc caps for 66,000 UF of power filtering and storage.

Load 4 each 6 volt golf cart batteries at 220 AH = 440 AH at 12 volts.

20 fuse at the battery for saftey. This test is limited to 10 amps.

Here are a couple pictures of the 2 PMAs.

I wrote down a lot of #s today. I think however I'll just show the RPM diferances and power requiered to make amperage at specific RPMs.

     JK TAS Jerry

[Jerry]

My computer just froze up. I was afrade of loosing everthing I had just posted. So I posted what I had trying not to loose that much.

Here are the #s I was about to post. I'll call Hughs 4ft HU, I'll call the Garbogen

GG. I'll show the requiered drive power in watts for each as DPW. I'll list the #s from .5 amp to 10 amps and RPM for each amp #.

                HU                                GG

 .5 amp 307 rpm  DPW 86.07                347 rpm  DPW 70

  1 amp 327 rpm  DPW 127.44               371 rpm  DPW 90.27

  2 amp 350 rpm  DPW 184.8                410 rpm  DPW 135.68

  3 amp 374 rpm  DPW 234.74               433 rpm  DPW 163.02

  4 amp 404 rpm  DPW 289.12               433 rpm  DPW 195.65

  5 amp 422 rpm  DPW 311.06               485 rpm  DPW 214.17

  6 amp 454 rpm  DPW 351.48               508 rpm  DPW 236.16

  7 amp 480 rpm  DPW 369.90               533 rpm  DPW 303.6

  8 amp 505 rpm  DPW 378.72               555 rpm  DPW 318.98

  9 amp 535 rpm  DPW 422.12               582 rpm  DPW 367.50

 10 amp 582 rpm  DPW 532.80               621 rpm  DPW 468.28

It apears the Garbogen requiers around 50 more RPM to make the same power as Hughs PMA.

It apears Hughs PMA requiers more driving power to make the same amperage.

I would think Hughs PMA would like a larger more powerfull blade and the Garbogen would like a smaller faster blade.

I will be testing these 2 PMAs in my wind tunnel both with 4 ft blades. That may be a while off yet  but I wanted to see there diferances at the same amperage and compaire requiered drive power and rpm on the bench befor the wind hits the fan.

PS. the drive motor is one of my 2 HP Magnatek 180 volt DC units.

                         JK TAS Jerry    

                                           

[oztules]

Jerry,

I can find no comfort in your measured figures. It makes no sense to me that an iron filled alt would out perform the axial flux in the efficiency stakes.

I somehow suspect this may reflect the impedance match between your driving motor and your participants requirements rather than the true power in / power out comparisons.

Then again, I could be wrong??.... However, I think it would be more useful to use the variac etc, and ignore the input figures completely (don't even measure them), and instead incorporate a pony brake/or what ever to measure the input torque on each unit.

Putting this another way, I think the power torque/curve of the drive motor is causing the results to be of little value. We need actual input power to the alts..... rather than calculated input power figures (real torque measurements, not reflections of the drive motor/variac/torque/rpm imperfections), then we can sensibly consider matching the alt to a prop.

Forgive me if I have misread the test, just the way I see it from here.

..............oztules

[Flux]

I agree that the only true way to measure power is the dynamometer method using torque and speed.

Using the drive motor input power is not going to give particularly accurate results but should serve as a reasonable comparison.

I would like to see the dc input requirements for the drive motor, terminal volts and armature current. I am not happy about using the rectifier input ac current.

With the dc figures to the drive motor I would also like to see a no load run on the motor with no load on the alternator.

The power into the axial machine at .5A  will be little different from 12W so we have 70W going lost in the drive system and it becomes tricky when the drive motor losses are far higher than the thing you are trying to measure.

There is another way to get at this and it may be possible for Jerry to do this and keep the convenience of the belt drive without having to dynamometer mount the drive motor or test alternators.

A simple prony brake on the motor shaft ( blocks of wood clamped round) or a rope brake over the pulley with a pair of spring balances would measure motor torque.

We need motor armature current dc amps and torque at each of the test speeds.

We then have a calibration of armature current against torque and for each load we get power from the armature current and speed using the current/torque graph. Motor loss is eliminated ( we have to live with the belt problems but it may be small).

One other comment

Test equiptment.

Fluke trure RMS meter measures PMA DC output.

Analog 10 amp meter measures PMA output.

HF cheap meter measures prime drive drive voltage.

HF clampon AC amp meter measures prime driver drive amperage.

The power into the battery will be battery volts a battery current and these need measuring with dc mean instruments.

I am not sure where the Fluke figures here, if measuring the battery volts it will be ok but we don't want the rms current into the battery. I have no idea what an rms fluke measures on its dc ranges I have never had reason to check but a common multimeter that measures dc mean would be a better choice.

I have made my comment about measuring drive motor current rather than rectifier input current.

I admire Jerry for doing this stuff and we may have to deduce the best we can from some of his methods.

If he can calibrate the drive motor armature current against torque it would be a lot more accurate. Failing that if we have the true dc motor input power, a no load power/speed run and an accurate value of armature resistance we may be able to get input power near enough for direct comparisons as the speed difference of the 2 alternators ins not great.

Flux

[Jerry]

Thanks for the input guys.

I can measure armature amperasge on the drive motor. Or even replace that motor with a PM 2.5 HP tred mill motor.

Not quite understanding why if the drive motor requiers more power why thats not reflected in the load its under? It must also have something to do with the rpm at at specific load also?

The best test will the wind tunel.

Flux I do have the no load #s for both machines. This will include open volts.

      HU                                         GG

 280 rpm    13 V                           368 rpm  12.73 V

 316 rpm    14.8 V                         395 rpm  13.74 V

 333 rpm    15.8 V                         423 rpm  14.74 V

 358 rpm    16.8 V                         450 rpm  15.74 V

 398 rpm    18.8 V                         477 rpm  16.76 V

 418 rpm    19.8 V                         504 rpm  17.74 V

 438 rpm    20.8 V                         532 rpm  18.73 V

 460 rpm    21.8 V                         560 rpm  19.73 V

 479 rpm    22.8 V                         589 rpm  20.77 V

 499 rpm    23.8 V                         617 rpm  21.79 V

 521 rpm    24.8 V                         643 rpm  22.77 V

 542 rpm    25.8 V                         671 rpm  23.79 V

 565 rpm    26.8 V                         698 rpm  24.77 V

 584 rpm    27.8 V                         726 rpm  25.78 V

 606 rpm    28.8 V                         753 rpm  26.74 V

 625 rpm    29.8 V                         780 rpm  27.74 V

 646 rpm    30.8 V                         808 rpm  28.79 V

   Sorry didn't record drive power at these rpms. I'll do more testing with drive motor armature amps/volts both loaded and no load.

                            JK TAS Jerry

 

[tanner0441]

Hi

Just a quick comment. To measure power (aproximately) we mounted a petrol engine on a swinging mount and biased it against a spring, power out was related to torque reaction deflection.

Would it not be possible to do the same thing with the drive motor, calibrate the spring (dyno or proney brake) then measure the deflection when the alternator is loaded.  The reading would be related to load, not the electrical parameters of the motor, perhaps a pot could be used to display the reading on a meter.

Bian

[hamitduk]

is it a Diff Pully? size Diff?

Hami

[oztules]

Jerry,

Looking at the side casing of the motor I think you have field coils as well as the armature to consider. (the two bolts on the side at 90 degree angles would appear to hold field winding poles)

Consider this. If we have X ohms as the nett resistance of the armature and the field, the watts lost will be an Isq R function.... so losses in this part will be a square function... not linear. I think you can see the problems starting about now.

So as we load the same motor differently for the same power (different rpm and torque for the same power output), we will have different quotients of voltage x different amps to get the same power out. But the one with the higher amps at a lower voltage (likely the one with same power at lower rpms) will suffer much higher power losses in the copper circuit, and brush losses as a squared difference in the current.

This is before we start to look at the effects of higher magnetising currents due to the higher amps... and thats before we start to see what happens with the iron at higher currents. In the worst case, we could saturate the cores, and pour in more power for no appreciable power output increase.

I had just this problem recently, where I was driving a 10hp induction motor into the grid with a 4kw dc motor (just trying to run the batteries down a bit). The DC motor was rated with 4kw@2880 rpm, the induction motor was 4 pole 50hz..... 1500rpm. As I loaded up the induction motor to get more power into the grid, the current went up dramatically on the dc motor as it really needed a pulley (near 2:1) to get decent efficiency out of it. It was trying to pull it's 4kw at half the rated rpm..... things get unreasonably hot fairly fast.... obviously poor efficiency as we pushed the power up.....was fine at 1kw though, barley warmed up at all, at 4kw you could cook your toast on it. So efficiency was not linear to say the least.

At least that's how I think it works

It will be only by measuring with a spring and rpm, will you get believable figures which can't be debated.

I admire your hard work Jerry. a real goer.

..........oztules

[Flux]

Yes you are right, to use a wound field motor you need to supply the field from a separate rectifier at constant volts. The power measurements must be armature voltage and current.

I didn't look closely at the motor but got suspicious when Jerry said he could change it for a treadmill motor ( why change one pm motor for another).

Running shunt motors with armature and field in parallel is not good over any real speed range and for power out measurements you must not include the field losses.

Flux