Two clients want me to fly their RED EPIC cameras - advice?

[ChrisViperM]

To be honest...apart from the fact that I disagree with a lot of what you wrote, I don't like your attitude, and I am not alone. Since you are apperantely a "developer" and know each and everythig, how do you explain this:

This is what you said in your post above:

"Now, as to the comparison of using a Heli to do this vs. a Multicopter, I have flown both big and small helis with the Arducopter System, as well as a quad with Arducopter. I really wouldn't say that there is much difference at all"

...and this is what you said in another post on this forum:

"I came here to research multicopters, as up to now I've been concentrating on Helicopters. I've got a 600 and a 450 I've been using to develop the code, but crashing them is costing me a fortune so I decided to try a quad"

I don't have more to say to this, but with respect to Benjamin Rowland this "sideline" should be closed now....rather watch some of his films, which are really great !

 

[R_Lefebvre]

The crashing, simply put, has nothing at all to do with *how* the heli flies (ie: how easy it is to fly and how it feels). And everything to do with testing code which may or may not work. As well as the effects of mechanical vibration on the flight controller, where that vibration is at least one order of magnitude harder to deal with on a heli, compared to a multi-rotor. And then when you consider the cost and complexity of repairing a helicopter after even a minor crash, maybe you will understand why I decided to get a quad. A quad can crash from 20 feet up and cost you $50 to fix. A helicopter can tip over without even leaving the ground, and cost $200 to fix.

Simply put, a helicopter has many, many single points of failure which will cause a crash. Any one of the 4 servos. All the linkages. The drive gears, the Jesus bolt, etc. etc. An Octo, if it's built properly, and has a flight controller than can deal with a motor failure, has very little if any single points of failure which will cause a crash.

So this is goes directly to the original topic of the thread. Somebody suggested he should look at a helicopter to lift this camera. I'm saying, I think that's the wrong way to go. Stick with the Octo, and use oversized ESC's if you want to.

 

[Str8 Up]

I don't know if it would ever become an issue or not but most every octo I have seen flying the Epic or Scarlet uses a lipo for camera power. It came straight from Red that this voids the warranty should a problem arise. Since we own the Red we fly we chose to use the Redvolts for power considering this. The extra weight isn't a problem since its still much less than the R1's we were flying.

We just wrapped a film yesterday that involved a series of vehicle tracking shots. For all five locations we would get airborne and do multiple variations/takes of the shot, staying airborne for 20 - 25 minutes at a time. I think a director who has worked like this would quicly become frustrated with an octo that would need to land every 7 minutes to change batteries. When you have a whole crew, police/traffic control and many vehicles involved, a director can get impatient very quickly. Often it would't matter but in this case it certainly would.

 

[R_Lefebvre]

IMO, that's the only place where a helicopter has a real advantage. You can probably get twice the flight time out of a similar size/weight setup with a heli than you can a multi. And then if you have a gas powered heli, double it again.

 

[Str8 Up]

I notice I am far more relaxed flying my gas versus electric helicams as I am not always constantly worrying about how much time I have left and can focus more on getting a good take.

 

[sim_io]

I've seen videos of an octo flying with one motor down on the WKM but they weren't carrying anything underneath so it is possible, maybe MK boards are better dealing with this? I've had a hexa naza with one motor out just crash out of the blue and it was carrying a light dslr, it was because an ESC decided to quit (the heat in the south). I don't think any other fc board would have saved it, power was gone from the right rear and with the load it just sloped down at 45 degree angle.

If the octo cinestar has just enough power to lift the Epic for a 5-6 minute flight I can tell you that a motor failure would hindenburg everything to the ground, lets hope you weren't too high up.

 

[R_Lefebvre]

I've seen videos of an octo flying with one motor down on the WKM but they weren't carrying anything underneath so it is possible, maybe MK boards are better dealing with this? I've had a hexa naza with one motor out just crash out of the blue and it was carrying a light dslr, it was because an ESC decided to quit (the heat in the south). If the octo cinestar has just enough power to lift the Epic for a 5-6 minute flight I can tell you that a motor failure would hindenburg everything to the ground, lets hope you weren't very too high up.

Really, it's that bad huh? I'd been moving in the direction of Octos because I assumed they did offer failsafe. But if they don't, and those 8 motors and 8 ESC are now each a single point of failure... that's actually not attractive at all. So is that how it is? An Octo carrying a weight can't fly with an engine out anyway? Then why not just use a quad with smaller motors?

A helicopter has the option of an autogyro if the motor fails, but the likely hood of success, if you are hovering, at low altitude, carrying a big weight, probably isn't very great.

 

[sim_io]

Really, it's that bad huh? I'd been moving in the direction of Octos because I assumed they did offer failsafe. But if they don't, and those 8 motors and 8 ESC are now each a single point of failure... that's actually not attractive at all. So is that how it is? An Octo carrying a weight can't fly with an engine out anyway? Then why not just use a quad with smaller motors?

A helicopter has the option of an autogyro if the motor fails, but the likely hood of success, if you are hovering, at low altitude, carrying a big weight, probably isn't very great.

Maybe someone with more experience with octo's can give you their input, I'm in the process of building a cinestar 8 from the knowledge I've gained with the ex-hexacopter, now a quad.

My thought process is that if your stress out the motors too much with this heat one way or another one will quit on you, I didn't have any problems on winter testing or spring I go in the heat one day and wham... everything fell out of the sky, frame/camera/gimbal bad. Funny thing was when I took it all home I just plugged in a lipo and all motors turned on and sounded well except for the one on the side where the esc quit. (my bad choice of ESCs has something to do with it) Although I have a good flying quad now for the gopro while waiting for the octo!

With an octo the load is spread evenly so each motor works less, hopefully with the gh2 and the lightweight gimbal is just light enough to land with one motor failure, I might do low altitude tests over grass and weights.

The conventional single rotor helicopter setup is the most proven one and used for hollywood but you can't fly that around people, the thing can kill you on a bad turn.

 

[R_Lefebvre]

If the octo cinestar has just enough power to lift the Epic for a 5-6 minute flight I can tell you that a motor failure would hindenburg everything to the ground,

I think this obviously is the critical point in the whole discussion. If your all-up-weight requires much more than 50% throttle to hover, then you have a serious problem if you lose a motor. Because you don't just lose 1 motor, you're going to effectively lose 2, because the opposite motor has to basically shut down to keep the thing stable. Then you need reserve power just for stabilization. Also, obviously, the higher your loading, the more likely you are to have a failure in the first place.

What ESC were you using?

I am only planning on flying with something like a Nex7. At least for the foreseeable future.

The other place I think Helicopters have an advantage is when the job would require a long, high speed flight path. Chasing a race car, or surveying a road, something like that. But they're just far more complex.

I had a quad from box to successfully flying in 8 hours. It takes WAY longer to get a Heli flying. And every crash requires almost a total rebuild.

 

[Str8 Up]

To fly a $50K camera you had best have all your crashing behind you. The attitude that if you fly you will crash has no place in this scenario. We employ the zero defects philosophy where all crashes are preventable if all reasonable precautions are taken. It takes a full four hours to do a pre-job inspection as many components are removed and inspected. Little is left to chance. The entire process is well documented even though i have it all memorized. Critical wear parts have well defined flight hour limits, etc. The only flying done is for work or testing as other flying only shortens the mean time between component replacements. With only two forced autos in seven years we know the process works but requires considerable experience to make happen. Crashing an Epic is akin to dropping a newborn infant, you make damn sure it does't happen!

 

[jdennings]

According to everydays logic, with everything "a little bit bigger"...and I am on the save side. That might be true on some - or most - occasions in life, it certainly is NOT true in electricity and elektronics. If you could not figure out from the links I gave in my previous post, I'll try it in a very simple way: If you mainly drive a car on small country roads or in the city, you will be fine. ..no overheating, good fuel consumption. This is called "efficient" . If you want to keep up with the Porsche in front of you with the same little car on a highway without speed limit, it won't take long and your car will make "poeeefff"...gone up in smoke. This is called "UN-efficient".....If you take a big strong car going on a highway 160 km/h is in terms of the engine layout, gearbox ect. and fuel consumption, "efficient". With the same big car going in city traffic in summer will make the big car overheat...it cannot work in it's efficient range,...lots of gear changes and only using very little of the possible engine power will push fuel consumption up and generate heat in the overall system..so it's unefficient.

There is only ONE bullet proof way to make your copter run cool, power-efficient and RELIABLE...that is choosing the components - motor, esc, LiPo, propeller - as close as you can - with app. 25% reserve - to their peak efficency. Everything smaller and everything bigger is contra productive. Fact is: Motor or/and Esc too small = lot's of heat and "poeeefff"....Motor and Esc too big = copter is very difficult to handle....Esc too big = creating heat due to it's internal structure = unefficient.

It took me a long time to understand all this electronic stuff in RC, but there is an even simpler way...try it out. Invest the time and try all possible combinations and measure the results. But for the theory, try to find out what the FET's are doing when they are "underused"....


Chris

Chris, thanks for your input. I think I understood very well the links, I've been doing electronics for a long time. Power losses due to voltage drops in diodes is electrical engineering 101. (by the way, a simple esc may not even have diodes as in the schematics in the links). The links you provided, while interesting, do not make the point whatsoever that "a larger esc than required" will lead to over-heating. The only thing they point to is that the larger the current, the larger the power loss given the voltage drops. This of course will lead to increased heating, if and only if more current is drawn. Same current: same power losses Same power losses on larger esc: less or equal heating (depending on design, heat sinking, etc ...). That's it, that's the theory, and there's nothing more to it.

Possibly less efficient? Sure. Because of the (small) weight added going to a larger esc.

In practice? I've been flying 3 multi-copters for a few years. Played with all sorts of prop/esc/motor/frames combinations, both on the bench and flying. I don't pretend to be an expert by any means. But never saw more esc heating when up-sizing to a larger esc. Sure, if I upgrade 8 pairs of wires from 16 gauge to 12 gauge after upgrading escs the weight is going to get up there, and my motors if not sized right may overheat from the larger thrust and resulting higher current draw.

But, again, the larger escs will not over-heat.

> "Esc too big = creating heat due to it's internal structure"
- No.

That said I am with you in terms of sizing every right for the max overall efficiency, in terms of flight time.

JD

 

[jdennings]

> ... Sure, if I upgrade 8 pairs of wires from ...

Should've said 8 triplets of wires from ..., of course. Typed too fast.

 

[Benjamin Rowland]

I've been running a lot of numbers and variables through eCalc, and putting together the right combo is no easy task with the current hardware available. One of the main issues that concerns me is that much of our shooting takes place around Atlanta, GA. And it is HOT here. We've had several days over 100 degrees lately. That doesn't bode well for keeping a motor cool.

I'll throw out some numbers that I'm working with...

Our Cinestar 8 with gimbal and downlink, but without ESCs, motors or batteries is 3kg ("without drive" on eCalc). The Red Epic with lens, battery, and media is about 3.9kg, so I've rounded up to 4kg.
So the weight for Cinestar 8 and Red Epic (without motors, ESCs, and flight batteries) is about 7kg.

Motors I'm looking at: Axi 2826/12, Kopterworx KW8, and Plettenberg. I'm sure there are others worth looking at too.

The KW8 motors seem to have a good rep and will likely be in stock in a few weeks. On the site, they advertise max thrust as 3.2kg (given the right props, batteries, ESCs, etc.). That's an absolute maximum of 25.6kg of thrust in a perfect world. Lose two motors and you've got 19.2kg of thrust. Given that my AUW is going to likely be in the 10 to 11kg range, this is still looking pretty good.

Now all I need is a good featherweight nuclear power source so I can be up in the air for as long as my neck muscles can take it...

If you have experience with these motors, please let the comments fly!

 

[sim_io]

You're in Atlanta? Cool. I'm in Atlanta right now just visiting and yes the heat is as crazy like home you just can't be standing around in the sun, tomorrow is my last day here before I head back.

I still don't understand how well eCalc figures out all the information I've put in various types of multis and it figures out the time to to 30s margin of error.

I've been running a lot of numbers and variables through eCalc, and putting together the right combo is no easy task with the current hardware available. One of the main issues that concerns me is that much of our shooting takes place around Atlanta, GA. And it is HOT here. We've had several days over 100 degrees lately. That doesn't bode well for keeping a motor cool.

I'll throw out some numbers that I'm working with...

Our Cinestar 8 with gimbal and downlink, but without ESCs, motors or batteries is 3kg ("without drive" on eCalc). The Red Epic with lens, battery, and media is about 3.9kg, so I've rounded up to 4kg.
So the weight for Cinestar 8 and Red Epic (without motors, ESCs, and flight batteries) is about 7kg.

Motors I'm looking at: Axi 2826/12, Kopterworx KW8, and Plettenberg. I'm sure there are others worth looking at too.

The KW8 motors seem to have a good rep and will likely be in stock in a few weeks. On the site, they advertise max thrust as 3.2kg (given the right props, batteries, ESCs, etc.). That's an absolute maximum of 25.6kg of thrust in a perfect world. Lose two motors and you've got 19.2kg of thrust. Given that my AUW is going to likely be in the 10 to 11kg range, this is still looking pretty good.

Now all I need is a good featherweight nuclear power source so I can be up in the air for as long as my neck muscles can take it...

If you have experience with these motors, please let the comments fly!

 

[jdennings]

Really, it's that bad huh? I'd been moving in the direction of Octos because I assumed they did offer failsafe. But if they don't, and those 8 motors and 8 ESC are now each a single point of failure... that's actually not attractive at all. So is that how it is? An Octo carrying a weight can't fly with an engine out anyway? Then why not just use a quad with smaller motors?

A helicopter has the option of an autogyro if the motor fails, but the likely hood of success, if you are hovering, at low altitude, carrying a big weight, probably isn't very great.

Yeah, it all depends on how much "reserve" thrust you have on the motors on an octo. I am pretty sure (well ... speculating) that a Cinestar carrying a Red Epic can be built and safely brought back to ground after a motor failure. But how much flight time once you've built it and installed bigger motors, esc's, etc ...? Four minutes? Three minutes?

This video on an octo losing one motor is interesting: http://www.youtube.com/watch?v=WYSXSjO0ZDg
Definitely brought it back in one piece. But can't say control was 100%

Somehow this has got to be safely testable. Tether the octo with a couple feet of cable, with dumb weights, and "program" a failure?

 

[R_Lefebvre]

Somehow this has got to be safely testable. Tether the octo with a couple feet of cable, with dumb weights, and "program" a failure?

Programmed failure could be simulated in the FC if the code is open source. Or, you could probably rig up a big FET feeding one of the ESC's that you can turn off with a radio switch.

 

[R_Lefebvre]

To fly a $50K camera you had best have all your crashing behind you. The attitude that if you fly you will crash has no place in this scenario. We employ the zero defects philosophy where all crashes are preventable if all reasonable precautions are taken. It takes a full four hours to do a pre-job inspection as many components are removed and inspected. Little is left to chance. The entire process is well documented even though i have it all memorized. Critical wear parts have well defined flight hour limits, etc. The only flying done is for work or testing as other flying only shortens the mean time between component replacements. With only two forced autos in seven years we know the process works but requires considerable experience to make happen. Crashing an Epic is akin to dropping a newborn infant, you make damn sure it does't happen!

You sound like a true professional. Obviously I'm nowhere near that yet. But this is all a great discussion guys.

 

[DucktileMedia]

BH insured the Sony fs700 for $500 against anything. thats a $9k camera that can compete with the red.

 

[jrlederer]

You sound like a true professional. Obviously I'm nowhere near that yet. But this is all a great discussion guys.

Well put LeFabvre, Str8up does sound extremely versed and exceedingly professional. I like the bits about the maximum number of hours allowed on certain component, in addition to sharing the nation that work helis are for only work, no cruising around for similar reasons. Some of the most level headed things I've come across on this board since becoming a member.

Have a nice day to all!

 

[jdennings]

Programmed failure could be simulated in the FC if the code is open source. Or, you could probably rig up a big FET feeding one of the ESC's that you can turn off with a radio switch.

Come to think of it: Test lift-off with one motor purposely made unfunctional (disconnect a wire or something)?If the octo can go up, even in a not too stable way, one would think that would garantee a somewhat controlled landing in the event of a real failure in the air. Anyone tried this? Doubt a closed source FC like DJI's would let one do this, I'm going to try this with mine though one of these days.

Will MK let you try lift-off with one missing motor? Ben, what FC stack are you using? I am assuming MK given DJI's woes with heavy-weights, 14x4.7 and up props?