Mini Spy Cam Recommendations

[davefoxx]

@Lemosteam,

Haha!  I appreciate and thank you for your "complications."  But, nah, I didn't miss your point.  When I was assembling the car, I was conscious to the fact the the gimbal would hang down below the end of the car, which is the main reason that I did not use the 60' bulkhead flatcar and went with a much shorter scratchbuilt car, which would get that lens back closer to the centerline in curves.  However, since I constructed that car, I can now see that the camera must be raised to get up to the engineer's height, that likely means that the gimbal will no longer interfere with the end of the car.  If that's the case, I will likely consider rebuilding a slightly longer car that will allow placing the lens over the lead truck.  It would also mean that I could put a coupler on both ends for some interesting shots from the rear of the train (with flat cars or gons in front of the "caboose," of course).

So, it's still a work in progress.  Stay tuned.

Thanks,
DFF

[C855B]

Here is a procedure to eliminate the camera's flexible neck. ...

Most excellent write-up, Pete! Thank you!

[NtheBasement]

By any chance is the lens a press fit?  If you can move it slightly away from the sensor that would give you closer focus at the cost of focusing at infinity.

Wondering the same about this $100 Ordro camera.  Video looks great, not thrown by camera wiggles, but 2 foot focus: https://www.youtube.com/watch?v=Y6b4J77d6XI

[wazzou]

@Lemosteam,

Haha!  I appreciate and thank you for your "complications."  But, nah, I didn't miss your point.  When I was assembling the car, I was conscious to the fact the the gimbal would hang down below the end of the car, which is the main reason that I did not use the 60' bulkhead flatcar and went with a much shorter scratchbuilt car, which would get that lens back closer to the centerline in curves.  However, since I constructed that car, I can now see that the camera must be raised to get up to the engineer's height, that likely means that the gimbal will no longer interfere with the end of the car.  If that's the case, I will likely consider rebuilding a slightly longer car that will allow placing the lens over the lead truck.  It would also mean that I could put a coupler on both ends for some interesting shots from the rear of the train (with flat cars or gons in front of the "caboose," of course).


So, it's still a work in progress.  Stay tuned.

Thanks,
DFF

I'd cut up a locomotive and glue a hood to something beneath the camera gimbal for the engineer's perspective....?   

[davefoxx]

I'd cut up a locomotive and glue a hood to something beneath the camera gimbal for the engineer's perspective....?

It's funny that you say that, because I thought an E- or F-unit window would be easily identifiable and cool.  The problem, on my camera anyway, will be focus.  Thinking about it, I'd also have to push the lens to the right side of the locomotive.  The gimbal makes this possible without having to throw all of the weight to the camera off center, i.e., angle the camera up and to the right and then adjust the gimbal straight ahead.  Hmm.

DFF

[peteski]

By any chance is the lens a press fit?  If you can move it slightly away from the sensor that would give you closer focus at the cost of focusing at infinity.

We are discussing multiple cameras here.  If you are asking about the one I took apart, yes the focus was adjustable until the sealed the threads with some sort of adhesive at the factory. It wicked into the threads.  I'm sure that to loosen the adhesive woudl take so much force that it woudl destroy the camera.

[eric220]

I'd cut up a locomotive and glue a hood to something beneath the camera gimbal for the engineer's perspective....?   

If you’re talking about putting a physical hood in front of the camera lens, forget about it. One thing I have been struck by watching these videos is the shallow depth of field. There is NO way you could get a camera to focus on something millimeters in front of it, and something six inches in front of it or further. It would be far more effective to digitally superimpose the image of a window and hood onto the video. There’s a guy building a full “Waldo” setup with an EMD cab who’s doing just that.

[narrowminded]

Mike, this is by no means negative waves from Moriarty, and you are soooo close IMHO both from a video quality and camera unit's perspective.

Aaaaand here is comes...

In the video the camera swings so far off track centerline, that I kinda lose interest.  On the straights of course it looks great.

A long flat car, with the camera just ahead of the lead truck bolster center, and slightly off center would put the lens in the vicinity of the engineer, IMO.

This is what I keep waiting to see in these types of videos.

Please don't hit me.

I agree with the idea that the focus of the camera not following the direction through bends dampens the experience a bit.  To fix that, maybe the camera mount should be over two cars with the camera lense area on a pivot on the lead car with a lever that engages a fixed fulcrum extending from the second car.  The lead car starting through a bend will shift in that direction and the second car's fixed arm engaging the pivot will effectively swing the pivoting camera in the proper direction leading ahead on the track.  The amount of the angle would be built into the lever fulcrums or maybe even adjustable to compliment the focus distance, maintaining focus on the point ahead.  Properly positioned the look ahead focus should be largely self compensating for varying turn radii.

Long cameras might have a scenery clearance issue with this swing effect but might accomplish the best clearance by having the pivot centered under the length.  This potential problem would be greatly reduced or nonexistent on those short box looking cameras or a very narrow camera.

Should I put this on my list of things to do?

[C855B]

My immediate thinking is mounting the camera element on a pivot so it turns with the front truck, or even by itself on a single truck, depending on how flexible we can get with a modified umbilical to the CCD head.

An additional thought remaining to be proven with a little math and engineering is to use a bit of mechanical trickery a la lever arms to aim the camera into the curve at a slightly greater rate than the front truck is sensing. This is how our minds anticipate changes in direction; we focus on what is about to happen as opposed to what is happening at that precise moment.

[wazzou]

What we need here are little gyros. 

[peteski]

Funny, that wide swing effect does not bother me enough to do something about it, especially in the camera I currently use, mounted in the nose of the loco (as my earlier video show).

And Mike's camera is also a no-go for me. Unless I shoe-horn it into a bi level passenger or a high cube boxcar, it cannot be fully concealed in a typical N scale loco.  The camera is plenty small, but the PC board is too large (too wide or too tall).

I suppose I could mount it in a Superliner, and get an observation car view" of the layout, but that is nowhere as useful as the locomotive view.

[narrowminded]

An additional thought remaining to be proven with a little math and engineering is to use a bit of mechanical trickery a la lever arms to aim the camera into the curve at a slightly greater rate than the front truck is sensing. This is how our minds anticipate changes in direction; we focus on what is about to happen as opposed to what is happening at that precise moment.

That's exactly what I was just proposing!  I think it could made made to work acceptably but the pivots and linkage will require very proper fits.   Probably should include pretty heavy cars, too.

[C855B]

... it cannot be fully concealed in a typical N scale loco.  The camera is plenty small, but the PC board is too large (too wide or too tall). ...

Ah. If I recall, you have an operational constraint - for the effect you want in a public display setting, it needs to be concealed. My objective is remote operation, so function trumps form if it has to. But looking around the shop here, I see candidates for loco mounting - Bachmann DDA40X and E-units, particularly, if you mount the board vertically so it dips into the fuel tank space. I've carved up a few Centennials for projects, and done that way it might even be possible to leave one end powered.

But I think my priority is the intuitive in-motion view on the screen from the lever-managed leading effect. The end result on the track may be something I cobble-up from styrene and Photon parts to look like a Rube Goldberg MoW contraption. We'll see.

[Lemosteam]

My immediate thinking is mounting the camera element on a pivot so it turns with the front truck, or even by itself on a single truck, depending on how flexible we can get with a modified umbilical to the CCD head.

An additional thought remaining to be proven with a little math and engineering is to use a bit of mechanical trickery a la lever arms to aim the camera into the curve at a slightly greater rate than the front truck is sensing. This is how our minds anticipate changes in direction; we focus on what is about to happen as opposed to what is happening at that precise moment.

Aye. Exactly what I feel.

[brill27mcb]

Yes, mounting it to swivel with the truck would only make the camera tangent to the curve. It will still be looking off to the outside of the curve ahead. The camera needs to turn farther than the truck does in order for it to look down the curving track ahead.

Rich K.