RESSURECTED TINMAN NEW INVENTION.

[tinman]

Your force gage is not always tangential to the flywheel. Therefore your torque measurements are subject to the cosine error.
To fix this, pull on your force gage with a strong thread (blue) attached to the flywheel's perimeter further away (brown fastener), so the line that passes through the center of the flywheel (green) and through the point where the thread stops touching the flywheel (red), always forms a right angle with the straight thread.  That red point, the force gage and its anchor point should always be colinear.

While i agree with what you say, the point of the experiment was to show the difference between the two in way of a %.
As we carried out both tests in identical ways, then we can safely assume that the results show an accurate difference between the two in way of a %.

If we were after an accurate torque graph, then the way you show to carry out that test would be the way to do it.

Brad

[tinman]

@Tinman,


Great stuff, love to see the extra time to explain a bit more deeply in these videos what you are looking for and how to achieve it. I would like just a little more clarity on one thing if you will. The difference in the test between the two experiments. You mention grabbing the wrong torque plate when you did the test at your shop and that the one at your home performed better. I believe you stated it was 5 mm wider plate that you had best results with? To clarify, the test that you graphed the best results was in fact done with the "worst performing" of the two torque plates? Meaning that if you had been able to test with the material you wanted, it would have been even better difference between the two tests. Just making sure I understood you correctly.
So, if one was to perform your test exactly, would it be a torque plate 30 mm wide, and how thick? I have seen others mention getting best results with 1/8" thick. Do you concur? Your's LOOKS thinner, maybe 2mm thick? Or did you prefer thicker? Also, since I think you said the wider plate worked better, did you notice a width that seemed to no longer work any better? Was it the 30 mm plate?


Last thing I was wanting to clarify. It appears that the 1st video simply has the difference of raising the electromagnet up and out of directly in front of permanent magnet behind the plate. Am I correct in understanding that with that test, the permanent magnet is still 1 mm gap from the torque plate and still pulling the plate into the gap in both experiments, and that the ONLY difference in the test (except for grabbing different size plate) is the electromagnet is simply raised a bit higher and out of direct center alignment with the permanent magnet?


Thank you for your videos and clarification so far. I hope my questions are not annoying, as I have been working on the bench myself to fully understand what you are presenting to everyone. It's great work and I have learned so much already. Thanks again.

Hi captainpecan

The electromagnet was set in each test so as the core of the electromagnet was centered with the PM.
The extra width of the TP from 25mm to 30mm makes a big difference-about 30%. But this also depends on other factors as well, which is why it is important to make a good test rig, so as you can try different width and thickness TPs, and see which one gives you the best results.
The thinner TPs shown in the video are on the verge of magnetic saturation, which reduces the pull force. If you are too small with your TPs, it can actually have the opposite effect, and push the TPs away from the field gate. The 5mm wider ones offer that slight increase in mass, and decrease magnetic saturation of the TPs. The saturation point will also depend on thickness of your TPs, and the distance of free TP to the flywheel mass. You must use a steel or iron flywheel (a magnetic flywheel) for best results, as the flywheel will also carry the magnetic field, and reduce TP magnetic saturation.

The TPs i used for that particular motor are made from 2 inch exhaust tube, at 2mm thick.




Brad

[wopwops]

Stephan, will you please make Tinman the admin on this thread before it goes the way it always goes... Hopefully, he just posts everything ASAP.

[Floor]

While i agree with what you say, the point of the experiment was to show the difference between the two in way of a %.
As we carried out both tests in identical ways, then we can safely assume that the results show an accurate difference between the two in way of a %.

If we were after an accurate torque graph, then the way you show to carry out that test would be the way to do it.

Brad

Good point verpies.
       but
I think that Tinman's point is valid here.  The conditions being the same in the two sets
or even very nearly so, should result in a very good approximatioin of the ratios and
that is what he is looking to demonstrate in this case.

[captainpecan]

I edited this post. Answered my own question.

Thanks again, great work!