High Speed Video Info Request

[Mike Rinkunas]

Hey all,

I've gotten a request for some high speed videography of a mylar appication machine that likes to eat some of the mylar. The machine processes about 300feet/minute of an inch wide metal band.

I've been looking around the forum and i've read up about the after affects plugin twixtor and also DynaPel's motionperfect with jacking up the shutter speed.

Would using an XL-1s with one of these programs be enough? Motion Perfect seems like it should do the job -, although, it will render new frames from interpolating the video footage.

or should i be looking into rental of a true highspeed camera system, and if so do you know of anyone in the PA/NY/NJ/CT areas that i could rent from?

i've been requested to keep the project on a small budget - less then $1200, this way they don't have to do any competitive bidding and saves them internal headaches.

Is this possible?
Any thoughts?
~Mike

[Glenn Chan]

Those plug-ins just make up information to make slow-motion look better... they don't actually do anything about increasing the information coming in. Depending on the application, it may not be useful.

Jacking up the shutter speed is something you probably want to do to get rid of motion blur.

I would look into using a consumer level camera that you can get your hands on for (almost) free. Resolution on most of them can be very comparable to an XL1.
Shutter speed: If it doesn't have manual shutter, use the sports shooting mode.

To me, this seems like a quick way to research what you need. If it works, then you're done. If it doesn't work, then you'll know you'll need something better. If the resolution is just a little off, then go rent/beg/borrow/steal a better camera (i.e. DVX100, which should have slightly more resolution than an XL1) which would be within your price range.

2- How long do you need to record? A lot of high speed systems are limited in terms of how long they can record.

3- What exactly do you need to record? Do the engineers just want to see the machine screwing up?
Do you want the footage to look nice?

[Mike Rinkunas]

Glenn,

how long hasn't quite been determined....i'm pretty sure it will be one of those "however long we need it" probably not more then 5-10 min of footage in real time.

This will be used by engineers to look at the process to see what is screwing up, so i will need a decent level of clarity, but nothing broadcast quality....

~Mike

[Ed Liew]

mike,
if lighting is good, you can try using faster shutter speed. images look sharper when playback in slow motion or still frame. disadvantage is, certain in condition visual will tend to be strobing such as camera pan.

[Dan Diaconu]

Mike,
this may not be an "orthodox" solution.... but:
if you can shoot the same thing through a semi-mirror using two camcorders having their frames "interlaced" (shutter speed set at 250/sec) you would get 120 different fields every second. I do not have a solution for the "perfect delay" but I am sure someone will chip in a way to do it.
The rest is just a half disolve and slow down of both in NLE.
(and obviously using a triangular prism, you could use three camcorders and interlace their footage to get to 180 fields/sec... and so on)

[Mike Rinkunas]

Ed,

no panning is planned for this shot...just one shot, of the machine running.

Dan....

your idea is interesting, very interesting. Now, what do you mean by a semi-mirror?

and i'm working in preimere pro, how do you do a half dissolve? - i think i know what you are getting at, cutting out 1/2 of the interlace field and combining that with the same feed from the other camera.

~Mike

[Dan Diaconu]

See through (half and half) One glass (as some sun glasses- you can not see the eyes from the outside, but you can see trough the glasses)
One camera shoots trough this glass POSITIONED at 45 deg, the other one shoots the reflection of the same scene in the same glass. (Imagine a camera using tele prompter) As for the rest, both footage plays delayed (so they "fill in the time gap) and both get top play at 50% speed. Top video has half transparency so you see them both (like a permanent dissolve) Good luck.

[Jack Smith]

The speed of the metal is fairly slow(about 1 1/2 miles per hour)
2 inches of material will pass every frame (assuming 30p) if the problem occurs over a number of inches or feet a high shutter speed should give a good indication of what is happening.The XL1s would be fine.
If the problem occurs over only a few inches it may not show
As others have said interpolation isn't going help.

[Mike Rinkunas]

well...

after all of this, it turns out that the company execs ordered a high speed camera setup for the company without having talked to the local level management, who had already decided to hire me (verbally, of course too!). gotta love corporate communications! So if the local engineers can't figure out how to use it, i've been told i'm first on the speed dial.

thanks guysl for all of your help on this subject.

Dan, i'm still curious about the mirror system you mentioned. would using a teleprompter beam splitting glass work for this application? or are there special glasses made just for this dual camera purpose?

~Mike

[Greg Boston]

Mike,

I had a similar problem to solve when I was working in the semiconductor industry. We had a tool that would randomly mishandle wafers. We used one of those little camera modules mounted to look through the viewport(process occurs under high vacuum). We used a time lapse VCR and recorded 'as long as necessary' until it happened again. It's clearly visible what the problem was using this setup. I think I still have the footage here somewhere and could email it to you as a demonstration.

-gb-

[Mike Rinkunas]

Greg,

Thanks for the offer, if the company calls me back to try again, i'll take you up on that offer!

Thanks
~Mike

[Dan Diaconu]

Originally Posted by Mike Rinkunas

would using a teleprompter beam splitting glass work for this application? or are there special glasses made just for this dual camera purpose?

I would think this may be the most appropriatte glass made for this purpose. On a general idea, a glass should have a reflective coating on one side and both sides should have antiglare (on top of that) I do not have any filters in mind that would qualify. I have a 3"X4" glass like this (without the anti glare coatings) I could post a pic if you want in a few days (or even do a test as required by your application) (Did some two shooting with two cameras years ago but for a different purpose/early focusing system).

[Jan van den Hemel]

I really like this idea for slowmotion video by using multiple cameras. I remember someone talking about a bunch of guys who had used an array of webcams to do something similar: since all the webcams (dozens of them) were recording frames at slightly different times, they could make really good slowmotion out of it. Unfortunately I don't have a link but maybe someone knows what I'm talking about?

[Eric Stemen]

I just bought the motion perfect thing, and it does work really really good.
If anyone can tell me a free video hosting site i'll upload a 3 meg video of a comparison of before and after the slow mo effect.