[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]

Re: leader strike photo idea



Original poster: Vardan <vardan01@xxxxxxxxxxxxxxxxxxxxxxx>

Hi Jim,

I am working on an LM3914 bar graph "scope" to do this. The only thing I don't know is how fast the LM3914 will operate. It might have to go to near 1MHz!!

Cheers,

        Terry


At 07:23 PM 9/23/2006, you wrote:
At 01:07 PM 9/23/2006, Tesla list wrote:
Original poster: Vardan <vardan01@xxxxxxxxxxxxxxxxxxxxxxx>

Hi Jim,

I tried that here but with a generic photo from Dan.

http://drsstc.com/~piranha/PIRANHA/Corrected-PetersPictureWComents.jpg

The problem is, with the coil running at like 300BPS, to catch just the right strike to capture on the scope. If the camera had a flash "hotshoe" it would be much easier to sync the scope. Having the scope in the picture too would be cool but the screen persistence would just make a blur. Digital scopes are no hope. Even though they can clearly pick out a single hit, they have a significant digital delay time before the computer decides to upload the event to the screen. To get the scope in the picture too, one would have to play with mirrors to get it the same distance away but in a safe place to retain the focus.

I was thinking about combining the images in "photoshop" as opposed to optically. You'd capture a frame on the camera and a matching trace on the scope. Then, hand stretch and align. The trick would be if you're automatically capturing a sequence of spinning mirror frames, then you'd need a reliable way to capture the scope data and to keep the two sets of data aligned.

There are some ways to do this (for instance, you put a few LEDs in the corner of the frame that have the current "acquisition number" (in binary) from the scope data acquisition program displayed on them. They'll render in the spun image as stripes along the side of the image. You can then use a program to go look at the image and read the bits.

I hadn't thought about doing it optically, but with an analog scope it would be easy.. you just turn off the horizontal sweep, turn the scope so that the vertical axis is parallel to the mirror spin axis, and let the mirror do the sweeping.





Steve Conner suggested using an LM3914 or two as a bar graph LED driver that could just be placed to show up in the picture too. It would work just like a scope then. But we are not sure if the LM3914 can operate fast enough (at "least" 100kHz).? If anyone has tried to run the LM3914 at say 100kHz and above I would like to hear from you!


You could use a software program to go look at the brightness of the pixels reading the LED, and turn that into a trace?