Small camera from an unexpected source.

[bicknell]

I saw a new video from the Lock Picking Lawyer, which is often interesting. But this one featured a super-small camera, which is not his usual material. It just showed me again how small cameras can be these days. Unfortunately not something that can be easily thrown together as a hobby creation, but I still have faith a manufacturer will get these into an engine in the not too distant future.

https://www.youtube.com/watch?v=DGdsIrAjp3k

(Oh, and BTW, don't watch his videos if you want to feel secure.)

[kingmeow]

You do need that fancy key cutting machine though.

In the world we live in today, I always believer if there's a will, there's a way. NOTHING is undefeatable!

[conrailandrew]

It's usually not the camera that faces the problem of being small enough, but the accompanying power source and wireless transmission/storage and associated circuitry that's the problem. The technology has been there for some time, I work at a place where we make tiny cameras that would be great for trains. It's the circuitry to make it all run that won't fit! Lol

It has been done though. I remember hearing about someone who put a cab camera and DCC into a fully operable E unit in N scale. I think Tomix makes a powered camera train as well?

Andrew

[bicknell]

It's usually not the camera that faces the problem of being small enough, but the accompanying power source and wireless transmission/storage and associated circuitry that's the problem. The technology has been there for some time, I work at a place where we make tiny cameras that would be great for trains. It's the circuitry to make it all run that won't fit! Lol

Yes and no.

I have a camera car, and once you strip away all the plastic shell gunk inside are 2 circuit boards, each no more than 10mmx10mm. They house all of the circuitry to make it work, and this is on an older 900Mhz unit I bought over 10 years ago. The one area where I somewhat agree is the antenna, antennas are a serious problem.

The problem is more one of volume. Generally these boards are built around a custom SOC. Everything is baked onto a single die. As a result they can be quite teeny, small enough to fit on a DCC drop in decoder for sure. But to design and build that requires moving a million units to make it profitable. If you're making security cameras or spy cameras there's a market to move that many units. In the model train world where a large run of locomotives would be 20,000, the economics just don't work.

The way it will eventually happen I think is for the non-model railroad world to continue to miniaturize until the packages are small enough. The PIC microprocessors many decoder manufacturers use today were not designed for DCC decoding. They are just computers on a chip that finally got small enough to use for that purpose. Larger PICs have camera interfaces, the super teeny ones do not yet. Larger pics have TCP/IP stacks, the smaller ones do not yet. But in a few more years they will. Features keep tricking down to smaller and smaller packaging. IOT is helping. Now that BGA circuit board mounting has made its way to low-volume services there will be more demand for small chips.

[NtheBasement]

I suspect the depth of field on the lock picking camera is only a millimeter or two. The antenna and most of the power consumption goes away if you are willing to "go dead" with a camera that saves to micro SD instead of broadcasting.

[bicknell]

I suspect the depth of field on the lock picking camera is only a millimeter or two. The antenna and most of the power consumption goes away if you are willing to "go dead" with a camera that saves to micro SD instead of broadcasting.

I'm not too worried about power consumption. My old 900Mhz camera, which is not efficient, runs for about 6-7 hours on a 250mah 9v rechargeable battery battery. That's a draw of only about 41ma on average. Most DCC users wouldn't even notice an extra 0.04amps on their system.

It's the physical size of the antenna that's the major problem. While there are "PCB Antennas" they are still too large for N scale (see https://www.ti.com/lit/an/swra117d/swra117d.pdf for one with a USB-A port for scale) -- so it's back to some sort of wire and that wire really wants to be a bit long for N scale. Going to 2.4Ghz helps get a shorter antenna. I believe the math is a 1/2 wavelength antenna is 6.25 cm, or a 1/4 wave is 3.125cm. An SD70, if my quick math is right, should be about 13cm long in N scale. So we're talking half to a quarter of the locomotive length for the antenna.

I wonder if someone could make a PRR unit where the train phone antennas were the actual 2.4Ghz antennas.......

[NtheBasement]

Not sure I would go the loco route. How much traction is st going to lose when you carve metal out of the front end to make room? Maybe put it in a separate car?
[IMG][/IMG]

[bicknell]

Not sure I would go the loco route. How much traction is st going to lose when you carve metal out of the front end to make room? Maybe put it in a separate car?

My current one is a separate car.

What I want is a "decoder" with a camera on the front pointed at the window that requires no additional space over an existing decoder. No metal removed, no traction penalty.

[MoPac]

Not sure I would go the loco route. How much traction is st going to lose when you carve metal out of the front end to make room? Maybe put it in a separate car?
https://www.nscale.net/forums/attach...6&d=1607438152

Maybe you should have used an SD90MAC to push that camera around?