Brook Hansen 1415 AC Motor Conversion

[adaml]

Dear All,

I am hoping that I might be able to seek a bit of assistance from those what know?!

I have been working on a Brook Hansen 5hp conversion, with much support from others, thanks G-, Sparweb.

I have managed to get the rotor cut down and 20no N50(alledgedly) mags installed, 4 pole.

My next stumbling block is blade sizing, I usually opt for the trial and error approach as I am afraid I really struggle with anything maths related - having read the likes of Jerry and Zubbly it would appear I am going to be looking at carving some fairly sizeable blades?  I can foresee stall as being a potential problem

Name Plate:

Motor:

I have tried to gather some data in the hopes it might help make a more educated guess than my normal methods (please excuse my terminology if it is not correct):

Open Voltage           

VOC           

RPM    V       

200.     13       
     
250.     16.4       
     
300.     20.5       
     
350.     23.3       
     
400.     27.1       
     
450.     30.1       
     
500.     34.1       
     
     

Into 12v battery           

RPM    V    A (meter set 200m)    A (meter set 10A)

200.     12.9    0.3                       0
     
250.     14.1    6.4                       0.34
     
300.     17.1    15.5                       1.17
     
350.     17.7    23.5                       2.29
     
400.     18.8    36                       3.33
     
450.     19.9    48                       4.51
     
500.     23.2    61                       6.42
     
650.                                10.73
     
     

I have not touched the internal wiring of the coils.  Six wires out, removed jumper and connection block only.

I am trying this one as an experiment and my goal is to use it to charge a 12v battery bank.  I was given two of these motors so this is really a learning curve for me - just hope I have not bitten off more than I can chew!!

I hope this info will suffice but please do say if more is required.

Many thanks in anticipation all.

Adam.

[Flux]

From your figures it looks as though it may suit a prop of about 6ft diameter. The cut in is fairly low so a slow prop about tsr5 may suit.

If you are in a low wind area the cut in would suit nearer 8ft but the alternator will not load an 8ft prop in high winds, it will become fast and noisy so you will have to furl early. 600 rpm and above is starting to get noisy at 8ft so you probably will not see much above 15A unless you want to live with something frantic. The efficiency looks to be very low so it may in fact match the prop better than indicated from the figures. Unless you can measure input power it is difficult to decide what prop it will match.

Don't even consider anything above 8ft.

Flux

[SparWeb]

Hi Adam!

You're in the club.  

The way you set up the test may have biased the results against you.  What battery did you use?  A single car battery will "cook" when it's the test load - as evidenced by the 23 volts as you charged it in your 500 RPM test.  A larger bank of batteries that can absorb the charge without high voltage will probably show a higher current and less voltage rise.

You also didn't mention how you drove the motor conversion at different speeds.  The speed of a drill press, for example, will vary as you load it, so it probably won't turn at the speed shown on the belt-ratio label.  On the other hand, a 10 HP lathe or milling machine won't even notice a load like that, and the labeled speed can be taken literally.

Flux's recommendations for the prop are your best next step.  If you really do want to conduct more tests to pin down the size more closely, I can give you some pointers.  I've done such tests several times, now.  At this point, I'd suggest making a set of PVC prop blades, mounting them and then getting it all in the air.  Either you discover acceptable performance, or it's not satisfactory, and changes to PVC are easy and inexpensive.

Got any more photos?

[Flux]

I have to confess that I was tired the other day when I looked at your figures.

I didn't see that you had got up to 23v into a 12v battery. Those figures are probably not a great deal of use and I certainly can't make any sense of those taken on the 2oomA range.

Your 650 rpm figure into 12v will be nearer 20A so the thing will probably match an 8ft prop much better than I originally thought from my assumption that the volts were held at 12.

As Sparweb said you need more accurate test results to deduce much from first principles but 8ft still seems a good size to start with.

Flux

[adaml]

Dear Flux, SparWeb,

Thank you for the valuable comments.

I have taken some new readings tonight into a single 12V 45a/hr car battery but discharged to 12V, the previous readings were taken with it at 13.1V.

Results:

RPM    V    Amp (10A setting)

200.     12.73    0.03
     
250.     12.99    1.37
     
300.     13.4    3.4
     
350.     13.9    4.75
     
400.     14.6    7.6
     
450.     14.8    8.56
     
500.     15    11.73
     
600.     15.9    14.98
     
     

Rotor with Neo's attached:

Test set up for set of 6ft blades:

I know the gap between rotor and stator could ideally be less.  I had to hand cut a fair proportion of the rotor down and wanted to ensure I removed enough material to be able to fit the magnets in rather than have to start cutting again with the hacksaw!  Also perhaps the magnets could have been closer together?

Pole shoes have been suggested and have read a bit having googled the board.  The problem would be in the fabrication of these and dis-assembling the motor.  Also this is a "prototype" and would be happy with 10A anything more a bonus at this stage.  I have an identical motor I could work on when I have learnt more and realised my mistakes!

I hope the above helps, please do let me know if there is anything further I need to add.

[adaml]

Steven,

Forgot to state how I drove the motor - unscientific - used my hand drill set to different speeds.  Hooked an rpm meter up to the motor and used that to check rpm was within reason and correlated as closely as possible to the drill.  Only method I have at present.

Regards,

Adam.

[Flux]

The new figures look more reasonable.

It still indicates blades in the 6 to 8ft region, perhaps a compromise at 2m ( 7ft ) may be a good starting point.

With the magnets that way round you have a lot of gap at the centre even if you can keep the edges close. This will restrict your top end power so large blades are not a good idea even though the 200rpm cut in could take something up to 9ft.

Pole shoes will almost certainly increase the top end power out but they may murder your start up and low wind performance. At very best they will need to be laminated and they will need to be skewed or staggered to break the heavy cog that you will get if you fit them without suitable precautions. Unless you want a lot of work and are prepared to experiment I would avoid them.

I would be tempted to try blades of 7ft diameter with tsr about 6. If you look at Dan's 10ft blades and reduce them to scale to 7ft, keeping the tip and mid point angles the same i think it will work well enough.

I don't have enough experience of pvc blades to comment, Once you have built as many wooden ones as I have I just can't see any simplicity in playing with bits of pipe to get an inferior result. If you do go for pvc then you may have a lot more experimenting to do. It may pay to start at 8ft and cut them down if needed but don't go for too coarse an angle or they will be too slow.

Flux

[ghurd]

What if a second, much smaller, magnet were placed on top of each existing magnet?

Like pole shoes, without the labor of laminating.

Something to reduce the gap at the center of the magnets without getting too tight on tolerances.

G-

[SparWeb]

Hi again Adam,

You already have a set of blades - I suggested PVC only because I assumed you didn't have anything to start with.  The solid wood will be a good start for experimentation, and later you can make something fancier and more durable, if you so choose.

I ran the numbers into an Excel spreadsheet to see the graphs (I'm a visual person) and the lines are straight and reasonable.  In fact I don't see the "top end" peaking out, making me wonder if there is a lot more power available at even higher speeds.  My experience, with 2 motor conversion now, shows a definite flattening of the power curve at high speed - ie. power goes up as speed goes up, but only to a certain point, then you don't get much more power for higher and higher speeds.

Since you haven't reached such a speed in your tests, then you could benefit from a slightly smaller than normal prop (so the tip speeds don't get high and noisy).  Starting with the 6-foot blades you have, I see a close-enough match that you should just go ahead and put them together, and see what happens.

If you want to try more experiments, you could try wiring it in Delta, to see what happens.  Your cut-in speed will be higher, but the power out will increase more rapidly.  Plot the two graphs of data and see how they compare.  You can always leave it in star like that, and try delta later.

So, do you know where you'll mount it yet?

[Flux]

I think this would be far better than trying pole shoes, just depends on finding suitable magnets.

Flux

[Flux]

I hadn't spotted that he has a 6ft wooden prop. That should work quite well and should get to 800 rpm without too much noise.

For very high winds the delta connection may be better but all round use I expect star will be better. If you get reasonable results you could try star/jerry connection later to get some of the advantages of both.

The lower flux from the wide air gap may help start up and the 6ft prop may work well. You probably won't get the speed up to the point where it shows much reactance limiting but if in a high wind area you will need some form of furling to keep the speed and noise down.

Flux

[ghurd]

Simply trying to bridge the international time gap...

It is currently wired in Jerry.  Individually rectified phases.  Outputs are paralleled.

Looks like Star cut in would be down around 120 RPM.

Seems pretty slow?

G-

[SparWeb]

Connected Jerry?  The photo above shows otherwise.  If so, then Star will definitely be too slow, and Delta will have cut-in speed and power curves comparable to the Jerry, so no point repeating the test.

Hmmm.... has anybody done comparison tests of motor conversions right up to the reactance-limited range, then compared jerry/delta/star for how pronounced the effect is?  Nevery tried Jerry myself because I use a 3-phase bridge instead of individual rectifiers.

Getting a craving for more lathe-mounted run-up tests, now.

[ghurd]

"removed jumper and connection block".  Pretty sure it is in Jerry now.

Reactance limiting on the big stuff?  I didn't figure you guys would have much problem there.

Now that you mention it, some of the small ones do peak out a lot faster than they should in Star from resistance alone.  Can get a lot more amps with Jerry, usually well over double and at a lower speed to boot.

If the note papers ever reappear, I will try to remember to post the numbers out of one of those tiny ECMs.  Drill press is only a 5 speed, so it will take some guessing.

Regular Delta never makes as much as Jerry.  I don't even bother testing in Delta any more.

G-

[adaml]

Hi All,

Many thanks for the comments.

As a test this evening did put it up in the air.  8mph wind average, 6A seen out in a gust.  This is with the set of 6ft blades.  

As per G-'s comments individually rectified, only change I have made to the wiring from factor is to remove the "jumper" from the block shown in the top picture.

I will try to get more data when I have more time and the light is not beating me.

Test rig:

Thanks again all, input very much appreciated.