Drones Over the Layout

[BigDaddy]

doctorwayne

Anyone interested can view the photos HERE

Well worth clicking on that link.  Wayne I love your irregular shaped buildings.  What size wye's did you use here and what space does it take up?

[mlehman]

doctorwayne

I'm glad that Barney did those photos when he did, since there's now a second level over parts of the layout, limiting the view somewhat.

Wayne,

That brings up another point. My plans are to shoot in the main layout room, where there's more airspace. The Cascade Extension is high enough it's a lot like the second deck on multideck layouts. There is only a couple of low spots where I'd have enough clearance to try flying a drone in.

[doctorwayne]

dknelson

....Of course, the entire notion of photographing the otherwise invisible portions of the layout runs headlong into the traditional advice, given by Tony Koester among others, that the sides of structures that face away from the aisles can be replaced by plain styrene to save time and materials.

You're right, Dave, and my around-the-room-style layout has many of those plain back walls.  I do, however, make a point of painting both sides of all rail, just for those occasions when a camera on the layout can give me a nice picture and a view of something I would never otherwise see:

As for drones, I've seen some nice railfanning videos done with them, but I doubt I'd get much use of one over my layout.  
I did, however, do a sort of fly-over of the parts of my layout which existed at the time.  Anyone interested can view the photos HERE

I'm glad that Barney did those photos when he did, since there's now a second level over parts of the layout, limiting the view somewhat.

Wayne

[mlehman]

speedybee

I meant a light source fixed to the ceiling, not the drone... fixing it to the drone would mean everything on your layout would cast moving shadows, which would look silly.

OK, gotcha, I was thinking that sounded pretty high wattage, but LiPo's can do some amazing stuff. As I noted about layout lighting...

I may need some fill lighting if I find dark spots shooting from the drone, though. As Dave noted, the backsides of some structures are, errr, plain and they may need work, because what's unseen gets shortchanged sometimes. The drone will know...

That does remind me of a couple of things that may apply to my penchant for night lighting.

What are the low light capablities of the drone camera?

Can the LEDs mounted on the drone be switched off?

Would a spotlight on the drone (OK, but not 100w probably) prove effective in making some interesting night scene pics, for instance simulating a helicopter rescue, etc?

[speedybee]

mlehman

 

 

speedybee

Regarding the blurry photos: One idea is to install a super ultra bright light source. Quadrupling the amount of light hitting your layout will make your shutter speed 4 times faster.

 

 

speedybee,

Thanks that's a great tip to know about these cameras.

Lighting in the layout room is pretty substantial, so may not need onboard illumination for the camera. That would be a trick to pull off in a small drone, but could be a useful thing if it's possible to install.

 

 

I meant a light source fixed to the ceiling, not the drone... fixing it to the drone would mean everything on your layout would cast moving shadows, which would look silly. And as you say, a 100W led is not going to fit on a drone. It would need a huugee battery pack and a fairly big heatsink. Better to have it wired directly to a wall outlet.

[dknelson]

When I think of the whole idea of being able to "fly" over the inaccessible portions of a large layout and take pictures, I immediately think of the Gorre & Daphetid and such areas as Angel's Camp, Akin, and Sowbelly.  John Allen would have  been intrigued by the possibilities.  Of course, the entire notion of photographing the otherwise invisible portions of the layout runs headlong into the traditional advice, given by Tony Koester among others, that the sides of structures that face away from the aisles can be replaced by plain styrene to save time and materials.  

Dave Nelson

[mlehman]

IRONROOSTER

And have your 1:87 drone license handy or the 1:87 FBI agents will arrest you.

Paul,

I know drones must be registered if they exceed 0.55 lbs in weight. That would be a largish drone for inside use and could really cause some havoc if it went down on the layout.

Not sure about the operator licensing requirement, I've read different things on the internet. But a drone in your basement is most unlikely to attract official attention, but am not an attorney, so I'm not saying anything about the reg if so, except that it's a largely irrelevant question if the drone stays inside the house.

[mlehman]

speedybee

Regarding the blurry photos: One idea is to install a super ultra bright light source. Quadrupling the amount of light hitting your layout will make your shutter speed 4 times faster.

speedybee,

Thanks that's a great tip to know about these cameras.

Lighting in the layout room is pretty substantial, so may not need onboard illumination for the camera. That would be a trick to pull off in a small drone, but could be a useful thing if it's possible to install.

[speedybee]

Sounds fun! Regarding the blurry photos: One idea is to install a super ultra bright light source. Quadrupling the amount of light hitting your layout will make your shutter speed 4 times faster.

Also, as long as the ceiling is decently high, a single super bright light source as high up as possible should cast some very realistic looking shadows in your layout.

edit: perhaps something like this 100W warm white led + driver?

http://www.ebay.ca/itm/LED-SMD-Chip-Bulb-10W-20W-30W-50W-100W-LED-Driver-Supply-High-Power-Waterproof-/231949443858?var=&hash=item360143a312:m:m2C9CmIJ8oq-5PqUA70c7Nw

if the 6000-7000lm rating is to be believed, that's as much as several car headlamps. Of course you'd need a housing with a reflector and a substantial heatsink as well

[IRONROOSTER]

mlehman

but here are some preliminary points of what capabilities a drone should have in order to be successful and avoid any 1:87 NTSB investigations.

And have your 1:87 drone license handy or the 1:87 FBI agents will arrest you.

 

Paul

[LakeMichijim]

As an Hubsan X-4 owner I would agree with the selfie stick.  The smaller drones just aren't stable enough.  Regardless, on your average layout you really have limited space to fly.  A moments in attention and ...  I have a little Mobius camera that I bought for use with my RC aircraft.  That can be put on to a flat car and go anywhere on my HO layout.  It does a really nice job.

[mlehman]

BRAKIE

Ed,At 68 my reflexes is well above normal for my age.My doctor thinks it because I play video games my reflexes stays stronger then others in my age bracket.. That may be true but,I think its a combination of my activities including our hobby.

Larry,

This is a good point. I'll be testing the other end of this spectrum, as I don't play video games at all. Those who do play will likely enjoy some advantage, but don't get too far ahead of things since the damages of a flight mishap will be real, not virtual.

[mlehman]

Just so we don't head down the distracting path of why even bother with this, I'm well aware of the limitations of other methods and the dangers of flight over the layout. Got plans for risk assessment and dealing with that. If people want to start a thread about why this is an awful idea, please do so. I'm more interested in finding out what is reasonably possible than letting the fact that something can go wrong discourage me. Otherwise, the Wright Brothers would never have got off the ground...

I'm going to write up my thoughts in detail a little later when I have time, but here are some preliminary points of what capabilities a drone should have in order to be successful and avoid any 1:87 NTSB investigations.

The drone should be "headless." This is so there is little or no confusion in which way the drone will respond to controller inputs in the confined spaces of an indoors layout. A headless drone responds to inputs directly, no matter which way it's oriented. Otherwise, the operator has to keep track of what's forward and reverse in moving the drone. Since I can be a bit dyslexic at times, I really need this. An expert drone operator might be able to do without this.

Hover/Altitude Set: This is basically an auto throttle tied to some means to keep a stable altitude once set, usually controlled via the left joystick. The amount of lift in a quadcopter is controlled by the throttle input. It can be a very tricky thing to do manually, even with lots of practice. This feature is just now reaching into the very smallest drones which are the best fit for flying indoors. Some use GPS for this, but indoors your GPS signal will be weak or non-existent, especially if the layout is in the basement. Others use a barometric sensor, which should work OK inside. Then the question becomes how much does the drone vary up or down once the altitude is set?

Auto Liftoff/Auto Landing: This is typically the riskiest part of any flight. Since lift is controlled by throttle input, bleeding it off slowly is essential to smooth landings and makes liftoffs more controllable also. Assuming you haven't modeled a heliport, I'm going to assume your "home" spot would be in the aisle or other location near the layout. Depending on ceiling height and the way the hover is set up, you want the drone to rise up, but not go so far as to hit the ceiling.

Other features that could be useful, but may not be strictly necessary

Fly by Track: Some drones now allow you to draw a "flight plan" on a smart phone used to control the flight or other means. That would allow the operator to concentrate on taking pictures and monitoring for safety. If this depends on GPS, may not work inside, but if this is implemented via inertial navigation it's utility would depend on the accuracy of the system. You don't want to bounce off backdrops or mountains any more than you want to hit the roof.

FPV (First Person View) This allows the operator a direct view from the drone. Given the confined spaces most layouts reside in, the drone will most likely not be out of sight, but this might allow flight paths closer than would otherwise be possible to terrain.

Collision avoidance: Something currently found only on higher end drones, this feature might be neat if it worked over the layout. However, when reading on the limitations of such systems, these indicate that the average layout wouldn't "read" on the sensors reliably. But things are always improving, so if such a system did work for layouts it would be useful.

Generally, a drone useful for layout imaging will be on the smaller side. This is both so that its performance envelope "fits" inside and in order to limit the damage if something unfortunate happens.

 

[BRAKIE]

Guy Papillon

For hard to reach places, I would rather suggest the use of a cell phone with a selfie-stick. A lot less fun but far more safe.

I will agree simply because a drone flight would be seconds over my 12' ISL. I would like to learn how to fly a drone for railfaning reasons.

---------------------------------------------------------

Ed,At 68 my reflexes is well above normal for my age.My doctor thinks it because I play video games my reflexes stays stronger then others in my age bracket..

That may be true but,I think its a combination of my activities including our hobby.

[Guy Papillon]

mlehman

I've long thought about using a drone to photograph parts of the layout that are otherwise not seen regularly because I have a tough time getting a camera there. 

For hard to reach places, I would rather suggest the use of a cell phone with a selfie-stick. A lot less fun but far more safe. 

[mlehman]

Frank,

OK, drones 1-0 vs model RRers.

Yeah, gyroscopes by themselves aren't going to cut it here. You need some sort of auto-altitude-hold, too, I'm pretty sure. Will discuss more after some sleep.

[zstripe]

Mike,

Good Luck! LOL...Tried that months ago over the layout...bad part was one side of the layout has a sloped ceiling (its in My attic)and I kept running into it. Hovering was a real challenge inside. Outside I did ok. I had one that had Gyro control that supposed to be easier to get it to hover...but I just didn't have the knack for it. Wound up giving it to My young 9yr. old grandson and that little sucker can do it. He takes it to the RC racing track He frequents near His house and it's wide open spaces out there...must be an age thing for Me though, My reflexe's have slowed quite a bit in My 74 yrs.

BTW: The one I had was not a cheapie either.

Take Care! 

Frank

[mlehman]

I've long thought about using a drone to photograph parts of the layout that are otherwise not seen regularly because I have a tough time getting a camera there. Then Santa surprised me with a tiny Brookstone camera drone. After almost two weeks of flying, I wanted to report some findings and hacks, as well as to inquire if anyone else is contemplating or accomplished such a tomfoolish thing.

Just to keep this on topic, I'm not trying to start a general discussion on drones. Kalmbach has a magazine for that now, Drone 360, which probably will have it's own discussion forum eventually. See: http://blogs.discovermagazine.com/drone360/#.WG830bmLXZV

What I want to discuss is about using drones to photgraph layouts. Seems fairly straightforward if you have a garden railway line outside. It's when an indoor layout becomes the subject when things get dicey.

Unless you're already an expert drone pilot, the first thing to do is practice flying. A drone crash on the layout could be disastorous. Just imagine those blades slicing and dicing through a lot of your modeling!

So try to get the crashing out of your system as much as possible seemed like a good route to take. The goal of taking good pics with a quadcopter drone is to learn to reliably hover, otherwise you get results like this.

Nothing wrong with abstract art, but I'm hoping to achieve, err, more focused results over the layout. Practicing flying is important, but so is taking pictures in the air. The shutter speed of the digicam on my drone seems rather slow for the app, but I suspect it's a more general problem with the "HD" 720P digicams common on small drones. Thus the emphasis on hovering in my training regime in order to take pics as close to a standstill as possible.

I'll be the first to admit the little Brookstone drone I'm using isn't particularly suited to this application. With the stock blades, it's just too fast and dodgy, at least for me. Your 15 yo might find the skinny little two-bladed props eminently suitable because of youthful, nimble fingers, so YMMV.

For me, I decided to try something a little different. I cadged some nice fat four-bladed props designed for the Horizon Blade Inductrix FPV drone (P/N BLH8507, set of 12 bladses). The Inductrix, as its name suggests, uses ring-like ducts to house the props protecting them while at the same time making more efficient use of lift.

The net effect is that the props with more surface area absorb a lot of the excess energy that causes the drone's dodginess with the original power. Control is greatly improved, but still not enough to use over the layout in most cases IMO.

Here's a pic taken today after some flying in the crew lounge.

I had the drone about as stable as I can get it, but it's not a really good pic. I've got some ideas on how to improve that, but they mostly involve upgrading my equipment to a better drone.

Enough for now. I'll discuss some minimum requirements that seem to exist in order to safely fly over the layout soon. For now, I'd be really interested in anyone else's experiment in using a drone over the layout.