Howard Johnson Replication Tube Claim

[eastcoastwilly]

That is the other side of the coin from what LightRider and I have tried to do. But your technique requires that the same researcher have access to the original video and to the analysis method--but I agree it should be done if possible.

What about the countless other factors that could affect the outcome of computer based video analysis software such as  like lighting, camera shutter speed, grime on the lens, flare in the lens etc ... ?

Not to re-iterate what has already been mentioned about horrible compression algorithms, time base errors (if applicable), and general garbage quality from most home recording video equipment but what kind of medium was the video recorded on ?

Any or all of these things would have a major effect on the software results and render them useless.



Will

[TinselKoala]

What about the countless other factors that could affect the outcome of computer based video analysis software such as  like lighting, camera shutter speed, grime on the lens, flare in the lens etc ... ?

Not to re-iterate what has already been mentioned about horrible compression algorithms, time base errors (if applicable), and general garbage quality from most home recording video equipment but what kind of medium was the video recorded on ?

Any or all of these things would have a major effect on the software results and render them useless.



Will

Yes, you are right, that's the point of calibration and validity testing. If you run your video analysis on my video, and your graph looks nothing like the one I recorded "live" with a calibrated instrument, then you can bet it won't work on other random vids either. But,say, your graph looks like mine only stretched and skewed. Then a correction might be applied to your data or method to make it usable on other videos.
So, anyway, there's a target, and the data taken "live" are there too. I'm sorry I couldn't do it on the "real McCoy" but you'll have to take that up with Mylow.

[LightRider]

TK,
sorry for the delay...
Here the analysis.
LightRider

Edit:    horizontal lines formed by the dot are probably due to the high RMP and video of 29 frames / sec limitation

[TinselKoala]

Sealed bearing bicycle hub 

Excellent idea. Might be draggier than you think, though.
Listen, if you really want to get into this stuff I recommend that you bite the bullet and spend enough money on your disk (big, heavy, 6061-T6 alloy), your mounting (stable, easy to dismount, quality sealed bearing that you can remove the seals from) and your base/chassis assembly (stable, sturdy, level-able, portable) so that you can have precision and sturdiness and stability on your basic platform. The magnets are secondary, but if you are going to be spinning several kilos of magnets around at 100 rpm you need to think about these things carefully.
So use good materials and parts, develop a "backdoor" relationship with a local machine shop or individual machinist (beer is an excellent lubricant) and go for precision and quality on the basic wheel and mounting.

(Now people are going to say only a rough-cut, flimsy motor will work, since TK wants you to build well and heavy... )

[TinselKoala]

TK,
sorry for the delay...
Here the analysis.
LightRider

That's great, LR! I am surprised that it worked so well, actually. But I thought surely it would work better at the low speed end than the high...I'm a bit confused about that part. Is there something I can do to improve the low-rpm resolution, or in general?
But still, that's pretty impressive and should answer the objections about whether your analysis can be applied confidently to Mylow's vids. Clearly it can.