Quote from: konehead on August 30, 2011, 12:01:43 AM
Hi all

messed around with magnets behind cores - got it up to 2020rpm now - it was 1620 yesterday. there seems to be no "logic" to those magnets behind cores...I had one magnet for each core before - it looked "right" at least  and that was the 1620rpm...but turns out this particular motor wants only 3 cores of the 5 cores on top plate to  have backing magnets, and on bottom plate its just two magnets....there are doubled up magnets (thicker) on top and just one magnet per core on bottom too...adding a magnet on a core that "doesnt want one" slows it very noticeably so I spent an hour  or so replacing and playing with the magnet possibilties unitl I found what the motor likes...
I made a identical motor too - now this one likes 4 (not 3) doubled-up magnets on top plate, and then one single magnet too so all 5 cores are "covered" on top-plate with this motor...bottom-plate likes only 3 magnets on cores down there and also just single magnet per core too like the other motor - they both seem to like "weaker" magnets on bottom plate ...again, adding a magnet to any core that "doesnt like it" slows motor pretty bad...very interesting that even identical-as-you-can-get made motor/gens will want different magnet arrays behind the cores -

Kone, if you transferred or swapped the magnets from one set up over to the other, you would be able to determine if it is the magnets themselves, or whether seemingly identical motors want different magnet arrangements.

Regards...

Quote from: Cap-Z-ro on August 30, 2011, 09:21:07 AM

Kone, if you transferred or swapped the magnets from one set up over to the other, you would be able to determine if it is the magnets themselves, or whether seemingly identical motors want different magnet arrangements.

Regards...

I agree 100%.  As we have seen in other research areas, "identical" magnets are far from that.  Even if purchased in the same lot.

Bill

Quote from: chalamadad on August 30, 2011, 05:06:36 AM
Here is a tutorial about how to make custom reed switches: http://www.youtube.com/watch?v=TC8-o1jm5Zo

Rebuild one. Works like a charm!!! Far better than standard reed switches.

Except that ....are any two cores with coils EXACTLY the same? It is more likely that it is as Kone has described. You have to "synch" individual core/coils separately, and there may be a "formula" for this, but without enough data, we are not even close to making a guess as to what it is. The process Kone has gone through is probably one of the main reasons we have not seen a replication yet. Because every core/coil on every motor has to be treated according to what the motor "likes" as Kone put it. And just replicating the physical structure of Romero's motor is not enough. It takes the time and the minute adjustments to get it to work, and too many people want to slap something together and get it to work. At least here we have some folks who are willing to take the time to painstakingly make those little adjustments and document it for the rest of us who are still working on our builds. I salute you all. Your tireless efforts are much appreciated.

David

Quote from: chalamadad on August 30, 2011, 10:55:58 AM
Rebuild one. Works like a charm!!! Far better than standard reed switches.

I hooked this up to one coilpair in adding configuration and measured voltage on DC side of FWBR. Without the reed I have 3V (which is not much, got a fairly large gap between rotor and coils). With the reed and after finding the best spot I can get up to 60V on the meter. Waveform is going crazy, lots of spikes, and it seems to be shorting many times per cycle. This will slow down the rotor though.

And one more thing: That coil I prepared the way mariuscivic did, I moved that on top of the stator plate and the bolt goes down to the rotor where the coil was before, with this one shorted the rotor will speed up a little instead of slowing down. Now I have taken a look at the waveform and I've got a fairly nice sine wave here.