Engine Cams - how well do they work?

[MisterBeasley]

I'm planning on a layout with a subway system, with a completely underground loop but also a ramp up to the surface. As most of the track will be hidden, I'm wondering about an engine-cam to see what's going on.

How well do these things work in tunnels, and how do they adjust to going into and out of dark areas? How does a bright spot like a signal light in a tunnel appear? Is one manufacturer significantly better than another, or do they all use the same basic components from some off-shore distributer, and just re-package the stuff? Any comments on black-and-white vs. color?

[cacole]

The only one I've seen operate doesn't work in tunnels due to the lack of adequate light and the loss of radio signal between the camera and receiver. It transmitted a beautiful color picture when in the open and close to the receiver, but range of these devices is extremely limited due to the high frequencies and low transmission power involved.

[DigitalGriffin]

you might want to try www.smarthome.com. Since you are only concerned about what happens inside the tunnel, you might be able to hardwire a infrared camera inside the tunnel and save some moola. :-)
~Don

http://www.smarthome.com/wirelesscam.html

[MisterBeasley]

No, I'm more into the idea of the Engine Cam itself. While I wouldn't equip a fleet with them, I'd like to have one. Of course, it means that I'd have to scenic the inside of tunnels, but I'd like the image of coming out into a subway station or a climb back to the surface.

If the transmission power is so low that it can't get through a foam tunnel wall, though, that would be a problem. Since a lot of these are battery-powered, I can see that Engine Cams aren't going to be putting out much. Fortunately, I don't need to do this right away so I can wait a while for the technology to improve.

[mcouvillion]

MisterBeasley,

I've mounted an RF Systems Lab TC-9 camera in an Athearn Trainmaster engine behind the lower headlight opening. It is track-powered. I have a second TC-9 camera that will eventually be installed in an engine, but is used in "temporary" applications on gondolas, boxcars, etc. The camera requires AT LEAST 10 volts DC for a good, clear color picture. Battery power is much better than track power, though track power works reasonably well. Depending upon the cleanliness of your track, engine's wheels, and location of the receiving antenna, your pictures can range from poor to excellent. There is more interference when running on track power than when running on battery power. I have connected two 9-volt batteries in series for 18 volts, or as they drain, more than 10 volts. This is a good way to get extra life out of batteries used in DCC throttles after they are a little weak from running the throttle. A single battery just doesn't have the power, or the life, to justify working by itself.

As for use in tunnels, this camera picks up infrared light very well. It cannot see in a tunnel unless there is an infrared source inside (i.e., an infrared LED on the front of your engine). You and I can't see infrared, but the low-light security cameras use IR LEDs to provide enough light to "see in the dark". I have a very bright white LED in the upper headlight of this engine, it's like a flashlight, and I can't see a thing in the tunnels with it. I can see when near the security cameras with the IR LEDs. A small problem is the filaments of searchlight signals, and the emitters of LEDs, show a singular bright white source in the camera, making color recognition of lights difficult. Ground color is OK since the light souce is elsewhere, but some, not all, bulbs are a little hard to see color.

As far as transmitting through the tunnel material, you should be OK as long as you don't use a wire-based support for the tunnel (chicken wire, hardware cloth, etc.). It transmits a weak microwave signal at 2.4 GHz, just like Wi-Fi. The more power, the better the signal. Some structures interfere with the signal, and some people do too, so we have mounted the better Train Antenna from RF Systems Lab on the ceiling for essentially "line-of-sight" signal transmission from the camera to the antenna. Your subway system should work just fine. You'll have to scenic the inside of your tunnel, however, since it really picks up the details and your tunnel will look horrible if you don't. (who would have thought that we would need to scenic inside tunnels! Some of us have enough trouble scenicking the railroad.) Things in miniature look real and full-scale.

In my opinion, and of those who have seen it in action, the train camera works great as long as you have enough power to the rails to adequately supply it. Go with DCC or greater than 10 volt battery power. You will be extremely disappointed if you try to do this on DC. At voltages low enough to make the train speed seem realistic, there is not enough power for the camera to transmit a good, reliable signal. At voltages high enough to do so, the ride appears to be in the cab of the TGV, flying around the track at unrealistic speeds.

Personally, I think the camera is a great addition, in the right application. I intend to operate from this perspective, as it is as close to getting at the controls in the cab of a real engine as I will ever get. Our operating group has a retired SP engineer who has seen it and said it is a very realistic view of the railroad from the cab. It is amazing how well the scenery looks, even minimal scenery. Others agree that it will add a new perspective to operation.

Let me know if you have any further questions. We intend to make a video of the railroad from the cab camera in the near future. Maybe I can get it digitized to post at least a portion of it for you.

Mark C.

[MisterBeasley]

Thanks! Right now, I've just made out my materials list for benchwork, so I've got some time before I get to the electronics. I've filed your posts for future reference. At 2.4 GHz, just about any sort of metallic framework is going to interfere, so I'll stick with foam and plastic, which was the plan anyway. It sounds like my priority should be to install DCC first and then go to the camera after that. It would be hard enough to get 1 9-volt battery into a locomotive, let alone 2. (Of course, there's always that second car in the subway train.)

Are you actually talking about infrared, which would be invisible to the eye, or is it just particularly sensitive to visible red if there's nothing else? It seems that a second LED covering the visible-to-near-infrared spectrum would be worth installing, particularly on a subway where it would look like a marker light. Above ground, there's enough light that you'd never notice it in daytime, and it would add some low-level illumination if you're running at night to counteract some of the "hot spots" of building and signal illumination.

And yeah, please do try to post some video! I've watched a few demos from MRR magazine, and there's a nice link on Tony's Train Exchange.

[Anonymous]

I have a small collection of traincams starting with an X10 colour cam on a flat car an extremely low lux b&w and another b&w that takes 16mm movie camera lenses.

The problem I find is that the fixed camera gives a picture in the direction of the car that it is fixed to. When the car goes into a curve the camera points off into the scenery instead of looking down the line.

My latest traincam is a colour camera mounted on a gimbal made from the bearing that drops into the hole in a 5 1/4 floppy drive. I added a 'steering' arm to the back of the mount. The cam car is followed by the steering car which has an adjustable actuating arm that automatically turns the camera as the cam car enters the curve. The resulting picture keeps the view down the track quite nicely. Of course the camera can also be adjusted to give angled side views that compensate in a similar manner.

I have a formidable collection of video tape but as yet very little has been digitised other than the early X10 efforts.

If anyone cares I could take some pictures of my tracking traincam for you.

[Trainman2001]

Yes, I would like to see a schmatic on the "tracking cam". I bought one of those tiny ones with the audio pickup last year at the York Train Fair. It does just what you say, points straight ahead when going into a curve and I have been trying to think of how to articulate it so it follows the track contour. But... if you've already designed one that works, I would love to steal the idea. Stealing a good idea is always superior to inventing your own any day...

I was also thinking even a little more wildly by mounting the camera to the actuating arm of a RC Model servo and using the RC transmitter to steer the camera independently of the train's movement. This would allow you to pan the countryside, follow engines that are passing, and generally operate the camera as you would your own vision if you were in cab. It would be excellent for making a subsequent video tape of the trains in operation.

Myles M.

[XG01X]

On the other side of this discussion; I know of a guy who did this(subway thing) and put pexiglass on one side so you could see in. He detailed it with bums and trash in tunnels it was pretty neat.

[Anonymous]

Hello Myles,

Here's a picture of the prototype, I'll try and get a few pics of the current model over the weekend
The prototype revealed problems with the power leads and antenna binding or fouling. The camera base pivot needs to be laterally rigid to prevent wobble yet free enough to rotate without jerkiness. I'll try to take pics that addresses my solutions.

This camera fits well onto a lego worm drive gear box making radio control for 360 panning simple but tilt gets a bit cumbersome. Then came the question of a radio controlled cam car for getting leading or following shots. DCCers already can do this.

[Anonymous]

I have installed several of these. The lighting is the only issue. You can either put bright white LED's in the nose and/or use lights in the tunnel roof. If you use LED's then they never burn out and you won't have to dig up your tunnel to replace low-light incandescents. The cameras that I sell and install are not gimbled but could be if you just wanted an "inspection" car. See http://www.trainbuddy.com/public/47C/f7a_athearn_genesis_dummy.htm for an install in an F unit. Track pictures from the camera under low-light conditions are available under http://www.trainbuddy.com/public/47C/47c.html
This shows some dark and light areas, turnout position and signal lights. Most use 2.4Ghz with a fairly narrow band. The ones that I sell are all color and are get rave reviews from my customers. I have installed in a caboose, SD70, F unit, and a Sperry inspection car. The power is either battery or battery eliminator from DCC track. Reception is very good. At least 300 feet up to about 800 feet depending on interference conditions.

TheTrain Buddy.

[Anonymous]

I don't know the specifics of these cameras, but some cameras are IR sensitive. You could try making an array of several IR LEDs - you could put more light on-target for the camera, without having a huge white spotlight at the front of your train.

Rob

[ben10ben]

O gauge 3 railers have had excellent luck in the past couple of years using the X10 cam. Some have mounted them on flatcars(and a handful of companies are making them already available mounted on flatcars), and others have installed them in engines. They usually require some careful trimming and fitting, but results have been excellent with good picture quality and high reliability.

Given the current progression of technology, an updated X10 or similar small enough to fit in an HO engine will be available for the same price as the current model. X10s are dirt cheap for the picture quality, if they just we're so large.

[Anonymous]

I'm in the process of putting one together right now with a $40 wireles camera I got last week. It's set in the front of an Athearn F7A with the battery in a B unit and the antenna alongside the super power weight. The range is good, but it needs voltage regulation at the camera and noise filtering in the power leads. I think the growler motor is causing some noise problems as well. It's slated to get a can and a bigger flywheel in the near future.

Pictures at: http://dpaton.net/hobbies/trains/traincam.shtml

Lighting is critical with something like this. There will be some superbrite LEDs installed in the future to help even out the lighting issues, probably about the same time I make the unit DCC compatible and add the outboard power supply to run it off of a function output. I'm also looking into a more reliable transmitter/receiver pair and a better CCD element to bump the resolution and Dmax. I'm still waiting for an HD element to be available in something small anough to pack into the front of an HO shell.

For now, I just tape a mini Maglite to the top and run it around the oval. After all, this is about having fun [:D]

-dave

[Anonymous]

Hello Ben,

An X10 will fit an H0 layout and they are nice cheap cameras but the picture is a little grainy and the flat directional antenna will phase in and out as you circle. It's not much of a problem adding a whip antenna but you need to pay attention to getting it the exact length


My first traincam, the camera was taken apart and spread over 2 flatcars.

[mcouvillion]

Mister Beasley,

Sorry for the delay in responding, I've been out on vacation for a week.

The RF Systems Lab TC-9 camera is sensitive to infrared light. It defintely sees in the IR part of the spectrum, as the IR LEDs of the cameras used to monitor hidden staging look like floodlights to this camera. I think, but have not tried yet, that one or two IR LEDs on the front of the engine could be inconspicuous to people but provide the "light" necessary to see in the tunnels. The LEDs could probably be mounted in the cab behind the windshield or possible behind the marker lights of certain engines. As mentioned in some of the previous posts, power is critical and batteries seem to work better. Maybe a small voltage regulation / capacitor system could eliminate the need for batteries altogether, but I haven't had time to try to develop one.

Some of the cameras pictured above are certainly large. Mine are pinhole cameras, about the size of gambling dice. The lens is a true pinhole. The gimboling needed for seeing around curves comes into play when the focal length is fixed and the radius of curvature is small. I don't have either problem. It is truly like sitting in the engineer's seat in the cab.

Mark C.

[MisterBeasley]

Mark - Now I've gone beyond the needs of the subway into some technical curiosity. Do these IR-sensitive cameras actually shift the wavelength up into the visible so you can see what's going on? If so, then installing IR LED's in the front of the train would be the ideal solution. Also, how fast do the cameras "recover" when they leave the tunnel and they are brought out into visible light?

Geometrically, I would mount the cameras as far forward as possible. That way, they would "lead" the rest of the train into the curve and gimbaling wouldn't be necessary.

[mcouvillion]

MisterBeasley,

I don't think the camera "shifts" the wavelength; I think it's "vision" bandwidth is broader than ours into the near infrared. Why it does not see the light from the bright white LEDs is a mystery to me. As I said earlier, it is like a flashlight. The IR LEDs look like floodlights under the benchwork (in the only area of the layout that has security cameras). I have not had the opportunity to mount an IR LED near the camera yet, though I have tried to find one in my collection of loose parts. I'll get to it soon.

I don't have any trouble with "recovering" the image when the train exits the tunnel. Things start becoming visible as the train nears the exit and are in focus all the way.

The pinhole nature of the camera allows things to be in focus up to about 1/2 inch (or nearer) in front of the camera. The handrail immediately in front of, and below, the camera is always in the picture and always in focus. So is everything across the room. The camera's peripheral vision is limited a little by the headlight hole it peeks through, but the loss is minimal. Otherwise, it's similar to standing on the front walkway of the engine (centered between the rails, behind the handrail).

My next effort will be the camera mounted from the ceiling of the cab where the engineer sits. I have already test-fit and powered it, and I do get a little blockage from the hood and areas around the windows - so does the engineer! I would love to have one in the cab of a Big Boy or Challenger, where the boiler blocks the vision along one whole side. That would REALLY be a challenge to operate safely on the layout - just like a real engineer has to contend with.

I don't think gimboling comes into play unless you have the larger lens with a focal point several inches in front of the engine. It misses everything up close. I did something like this about 10+ years ago when I strapped my camcorder to the grabiron of the short hood of an SP diesel while they did a short "turn". I put the lens to tele-photo rather than wide-angle. Going down the straight track, everything was fine. Going around the curves, all I saw was trees. The focal point was too far away. Had it been on wide-angle, I would have seen all the cars on the siding in the curve. I did not see anything but trees. (You would be amazed at how many people go around the crossing gates in just a few miles of operation. It made me very apprehensive to see it on video. I can only imagine that it gives engineers "heart attacks" and constant high blood pressure.)

I think the specific camera has a role in how you need to set it up. The TC-9 has three small wires coming out the back, opposite the lens - two for power and one that acts as the antenna. It is much smaller and "cleaner" than the ones pictured in the posts above. Again, it is a pinhole camera and does not have the large lens as on the other cameras.

Good luck in your installation. If I had a digital camera, I could post pictures of my installation. No such luck. Sorry.

Mark C.