Quote from: JustMe on May 29, 2008, 09:52:53 AM
OUMan...are you available?

Potentially, JM - are you?

Quote from: aether22 on May 29, 2008, 05:14:00 PM
Vince thought of a better way, he measured the torque on the motor housing as that only increases if the motors power increases, and it did increase.

Vince's ad-hoc method was ingenious and hats off to him for that. But as he mentioned himself, it's not a dependable way of measuring the torque. I really think that measuring the torque with a calibrated torque sensor is what's needed. It's a crucial piece of evidence and if it's not measured in a dependable way its value is hugely reduced. It's only a matter of buying the appropriate torque meter and attaching it to the shaft.

Quote from: CRANKYpants on May 29, 2008, 11:24:53 AM
COMPARING SLIP ANGLE PERCENTAGE DEVIATIONS FROM "NO LOAD" TO "ON LOAD" AS RECORDED BY THE CURRENT INCREASE OR DECREASE WILL GIVE THE TORQUE NUMBERS VERY EASILY.

Cheers
The ThaneMEISTER

That may be possible if you've already calibrated the motor's power-torque-slip-speed characteristics but I don't think you've done that calibration so it is NOT possible to deduce the torque from the current. There are two problems with your proposed method:

1. It is at best an indirect method of inferring what the torque is and so it will be inherently much less accurate than measuring the torque directly. At this stage, when you're measuring only a few percentage points either way, the inaccuracies will kill you.

2. To develop the motor's characteristic curves (which you absolutely need in order to infer the torque from the current), you have to fully characterize the motor, which means that you need to measure the torque anyhow to do the calibration. Which means you need to use a torque meter.

Your whole proposition depends on what torque is being transferred along the shaft between the motor and the generator. So measuring that torque accurately and directly is very important.

Quote from: CRANKYpants on May 29, 2008, 06:40:56 PM

YOU GUYS SHOULDN'T LET THESE RESULTS SLIP BY UNNOTICED...
THE HV INDUCED FLUX IS ENTERING THE HC COIL ON THE OTHER SIDE OF THE ROTOR AND INCREASING THE INDUCED VOLTAGE IN THE HC COIL - WHILE ALSO CAUSING ACCELERATION.

Thanks for pointing that out, very interesting, not sure I agree with your cause as it could also be that in the high speed test the HV coils were distorting the flux going into the HC coils causing them to drop power and now that they are on the opposite side they don't have that influence. So now the HC coils, without interference, respond accordingly to the increase in rpm.


TEMPORLARRYCLY! Cool, but I usually strap to the spoke and also from the rim to the bottom of the magnet. I noticed the top left magnet slipped to the side in 2_HC_AND_HC_COILS_ON_ROTOR, rim strapping may help.


BTW, I know this is not said enough, but thanks for all the great testing and for passing on the info.


Regards, Larry

Quote from: OUman on May 29, 2008, 09:54:00 PM
Vince's ad-hoc method was ingenious and hats off to him for that. But as he mentioned himself, it's not a dependable way of measuring the torque. I really think that measuring the torque with a calibrated torque sensor is what's needed. It's a crucial piece of evidence and if it's not measured in a dependable way its value is hugely reduced. It's only a matter of buying the appropriate torque meter and attaching it to the shaft.

Can you find one for sale online?