What we know about the new Sony Airpeak drone

It's a shame sony doesn't have more high-end PZ lenses - with the weight budget on this you should be able to carry on of their nice F4 zooms with 100 g left over for the PZ mechanism

Jun 12, 2021permalink

Sony's digital "Clear View" zoom yields an effective 1.5x zoom range with most cameras and lenses. Sony professional PZ lenses are relatively large and heavy compared to their non-video counterparts. They're barely manageable on a DJI Ronin S gimbal, and an unlikely fit for the new Sony drone.

Jun 12, 2021permalink

The 24-70 f/4 fits in the weight budget with about 150g left over, and zoom motors aren't that heavy.
Hand held gimbals are a different story - a drone is a smoother platform than a monkey

Jun 12, 2021permalink

Zoom motors require a substantial base plate, dual 15 mm rails and lens support. The motors require 12-14 vdc power, presumably from a battery. Then remains, what controls the motor? The entire assembly must be rigid enough to withstand the substantial reaction force between lens and motor without slipping out of gear.

I use a rig of this sort for land based videography, which builds a 2 lb A7 camera into a 12 lb rig, plus another 3 pounds for the battery.

Jun 12, 2021permalink

It's a lot lighter integrated into a PZ lens. That's all I'm suggesting, a PZ version of the same optical formula

Jun 12, 2021permalink

There are several PZ lenses for APS-C. The best match for the 24-70 in a PZ lens is the PX 28-135/4, which weighs 1212 grams (42.9 oz) vs 886 grams (32 oz) for the 24-70/2.8. As I noted, it is a hunk. The newly announced PZ 16-35 T3.1 would have a focal length better suited for drone photography. I don't have a weight, but any lens over 4" in diameter at the objective with 16 elements is probably a hunk as well. At $3600 it's still cheap for a cinematic zoom (Zeiss cine zooms start at $15k).

Cinematic quality is not for someone who hesitates to spend $600 for a lens.

Jun 12, 2021permalink

oldtimes

One word.... Sigma
Also, I wouldn't hook a full 7 series into this, who needs a viewfinder when mounted on a drone?. That's where the 7c will shine with lighter weight.

Jun 16, 2021*permalink

Distorted_Light

One thing doesn’t make too much sense. Sony should create a dedicated camera gimbal as an additional choice, just as the DJI approach. The present offering looks pretty crude when Sony actually possesses all the technologies to produce a good gimbal camera to provide longer flight time. 12 minutes flight time is not a convincing proposition as a new comer in drone business. That is like going back 5 to 7 years in drone technologies.

Jun 12, 2021*permalink

The gimbal is similar to that of a DJI Ronin, with 3-way adjustments for dynamic balance. The Gremple yoke is enclosed rather than one-ended, which be lighter if sonewhat less flexible than the DJI version.

Jun 12, 2021permalink

Distorted_Light

By integrating a good quality camera and gimbal together would make it a much more attractive proposition for most of the professional, if not all. That can be done in a Ronin approach too. All the buttons, dials, LCD screen, viewfinder and hand grip with battery inside are dead weight for a drone that is very weight sensitive.
And the existing gimbal with market available cameras can be offered as an option. What I meant is the system will be far more complete with a dedicated gimbal camera for people who need that, especially when Sony has those technologies.

Jun 13, 2021*permalink

Just Another Photog

It’s way outside my comfort zone. $9000 for the Drone, say $1,750 for the Gimbal.

THEN the Sony camera. Entry level Full frame: $2600.00. So then the training. I have no idea, but I’ll bet it’s atleast $2000. So we just spent $15K and we start trying to sell shoots, but competition has spent far less than you have. So they can do the job cheaper.

Sony misses the mark on this one

Jun 12, 2021permalink

Malling

But how many of the cheaper has a FF sensor and the ability to change to some of the best lenses on the market.

People who are going to get this is probably already in the Sony Ecosystem being either photographers or filmmakers. There are drones that cost tens of thousands out there, so while it might be expensive for the average joe it’s not something that would hold many pros back. 10k is peanut in the drone and cine industry, even in lenses you cannot even get the 600mm lens for that.

If you find this expensive then your not it’s target audience, but Sony will probably launch lower entry models down the line and probably higher specs too. Be glad that there will be more options.

Jun 12, 2021permalink

(unknown member)

So Sony missed the mark on this, since they did not make a drone specifically for your wallet?

If the drone helped you earn money, what then? That's what professional tools are supposed to do.

Jun 12, 2021*permalink

Just Another Photog

I have owned and operated a Video and Photography company since 1994. We ran a great business for our clients.

I’ve owned a great many expensive bits of gear, including a $30,000 beta cam camera. I operated 4 HD edit suits, gear for/ multi cameras and much more.

I understand the value of top end gear. I’ve owned more of it than you ever will.

It still doesn’t make financial sense to have $15-$20K in gear that will take more time to justify and payback before the technology needs upgrading.

Now having said all that, if aerial photography/videography is your total gig, and you live in a community where much of commercial and entertainment industries then this would work, and you can justify it financially.

The lower cost competitors BTW would still be in larger drone, but remember the Phantom 4 Pro has a M4/3 camera and high quality glass for a lot less money.

Oh, you have to factor in a camera. I wouldn’t put my A7R3 with $2K of glass in jeopardy when I have another shoot.

Jun 12, 2021permalink

(unknown member)

Then the drone probably is not the right tool for the kind of business you are running.

Jun 12, 2021permalink

Teila Day

@just - You're presupposing Sony "missed the mark" if you don't know what market they're focused on. By your post, in my opinion, you're not the kind of business/individual that they're trying to appeal to.

Jun 12, 2021*permalink

I just finished binging on the British series, "Doc Martin," which was filmed in a small fishing village (St.Isaacs) in Cornwall. They used a lot of drone photography for establishment. Imagine moving and setting up a 60' crane in that location.

A commercial drone license costs less than $150 if you are a good self-study. That teaches you when and where you can legally fly, not how or why. Flying the beast is actually quite simple. It stops and hovers if you releast the controls to get your bearings. Up,,down, turn right, left, go forward/backward/sideways. Coordinating the gimbal angle with flight is probably the hardest thing to learn, besides dodging trees and wires.

Anyone who can juggle kitchen knives while riding a unicycle should have no trouble.

Jun 12, 2021*permalink

@Just Another Photog
For a record Phantom 4 Pro has 1" type sensor camera not M43. Inspire x5s has M43 type sensor

Jun 13, 2021permalink

photography-lover

I would consider it if I were still shooting Sony.

Jun 11, 2021permalink

Does it need one or two batteries to fly?

Jun 11, 2021permalink

Dale Baskin

It uses two batteries.

Jun 11, 2021permalink

(unknown member)

How much is the gimbal?
Isn't that an important part of a drone?

Jun 11, 2021permalink

Dale Baskin

Sony didn't announce a price for the gimbal, but indicated that it will be based on the Gremsy T3, which currently retails for around $1750.

Jun 11, 2021*permalink

(unknown member)

I see the Gremsky T3 is compatible with many camera brands. Maybe the one for the AirPeak will cost less because it doesn't need to support those other brands. It can be more stream lined.

Jun 11, 2021*permalink

Revenant

Or it will cost more because it's developed just for Sony, and because they'll sell fewer of it.

Jun 12, 2021permalink

Wes s Smith

It looks to be an exact replica of the Gremsy T3 V3, I already own this gimbal and have kept this hoping to use it with the airpeak, unfortunately I have had confirmation from Gremsy it will not be compatible.
Arghhhh

Jun 15, 2021permalink

User1234567890

Cool.

Now , if you are in Africa and see a cheetah chasing a gazelle, you can chase both closely. Even better using Real Time tracking of A9ii/A1.

Pixel shift shot from volcano entrance.

Jun 11, 2021permalink

Francis85

Wouldn't do the latter with a 12K drone that only has 4 engines. Only one needs to fail and it will crash.

There is no chance a quadcopter survives a single engine failure.
Props are always positioned:

A - B
B - A

A and B spinning counterwise to eachother.
If one fails 'lift' is gone and can't be corrected by the other engines.
So resulting in a crash.

Then there is the fact that most wildlife parcs in Africa won't allow for drone flying without a commercial license.

Jun 11, 2021*permalink

"no chance a quadcopter survives a single engine failure"

Physics says otherwise. Just as a tripod only needs 3 legs for stability, a 3 rotor copter can be stable assuming the software is smart enough to compensate (and if the hardware can accommodate); e.g. use two rotors for lift and the 3rd for stability and control. Google "can quadcopter survive a single engine failure" (or similar) for multiple links on how this can be done. However, it is open question if this drone can do so.

Jun 11, 2021*permalink

The crash issue I want to hear about is a parachute or other means to qualify under the new FAA guidelines for flying over people. To the best of my knowledge, there are still NO class 2 or 3 drones that meet the new FAA guidelines....

Jun 11, 2021permalink

tangbunna

@astrostan 3-copter arrangement is different from a quadcopter with a failed prop. you can imagine about stability arrangements of each arm and prop.

Jun 12, 2021permalink

Can they spin propellers backwards for downforce? Because you'd need the third prop to have that option to balance if one motor died on a quadcopter (assuming there's low enough inertia in the prop that you can switch direction fast enough)

Jun 12, 2021permalink

@AstroStan
Physics as theory say many things but applying them to practice is often more challenging than one can imagine.The props are fixed pitch type and as such they can only generate lift in upward direction. Amount of lift is controlled by speed of rotation. If one motor fails you'd need to switch off the other one diagonally opposite. Otherwise you get an uncompensated one directional torque causing the drone to spin. It might be able to sustain altitude on two motors if they could generate enough lift for the load. You could not control attitude of the drone in desired direction with only two motors running. As soon as you engage the third motor the quad drone will start to spin, tilt and likely crash. If a wind would push it in one direction it would be impossible to tilt it against the wind with only two motors running and bring it back. The drone might be gradually brought to the ground somewhere but the landing would more like be a crash landing.

Jun 12, 2021*permalink

Francis85

AstroStan - I see you don't know much about drones and yet never flown one either. Nor do you know a lot about aerodynamica. Yet you think to know.

However a 3-copter, 5-copter of hexacopter works all very different from a 4 rotor drone.

And ProfHank - Despite the new FAA regulations for parachute.
This is not going to help a lot, most of these drones are flying low altitude. Before that parachute is being deployed, the drone has already crashed.

Jun 12, 2021*permalink

A Owens

You can just stay in the hotel bar and send the drone out. Firmware 2.1 has the cheetah on gazelle recognition and sends the footage to your phone and you select the frames to post on Instagram. FW 2.2 does the selection.

Jun 12, 2021permalink

Francis85: The parachute is to reduce impact force by limiting terminal velocity. If the drone is flying very low, it doesn't reach terminal velocity when falling. Besides, the key point is that the impact force limit is an FAA rule requirement for flying over people, not an option, and the rule is currently in effect with NO class 2 or 3 drones qualified.

Jun 12, 2021permalink

le_alain

Ah Ah Ah "Physics " ........ :)

Jun 12, 2021permalink

Francis85:

Why do you have to be such an ass? If you were an aerodynamic engineer then you would reply with meaningful content instead of spewing puerile insults. You obviously did not bother to do the least amount of research. There are several sources easily found via web search that could help you dispel some of your ignorance, such as
https://www.digitaltrends.com/cool-tech/quadcopter-can-fly-with-three-rotors/

It is obviously complicated and no current off-the-shelf consumer drone will do it out of the box (so far as I know). But it can be done as has been demonstrated.

Jun 12, 2021*permalink

P.S. Here is a video demonstration of enhance control software on a standard quadcopter losing a propeller in flight and recovering:
https://newatlas.com/quadcopter-failure-algorithm/30031/

Jun 12, 2021*permalink

https://newatlas.com/quadcopter-failure-algorithm/30031/

That is brilliant! I can see where a wild-looking spin could give nearly complete control even with just one prop working!

Jun 12, 2021permalink

@AstroStan
I am sure that even you can see that the video shows purely a scientific experiment with heavily modified quad using balancing weights and counterweights strategically positioned to mitigate the effect of one specific prop failure. This does not carry much bearing to real life application, IMHO. In a lab you can make even a pig fly😉 if you stretch its skin into a shape of a wing and put a prop in its ass. The only workable solution at present, and I dare to say foreseeable future, to save a quad drone from falling from the sky at terminal velocity is a parachute. Bud that is additional dead weight and on a drone with max 12min flight time, that is the last thing you would want to add...
But the prosummer drones have proven to be very reliable machines and prop or motor or esc failures have been extremely rare over the last 5 or so years of broad proliferation of drones. So, this would not be my biggest worry when considering purchasing this drone..

Jun 12, 2021*permalink

@ProfHankD: > “That is brilliant! I can see where a wild-looking spin could give nearly complete control even with just one prop working!”

Hmm, I have no experience with flight control software — but that IT MAY BE POSSIBLE was kind of obvious. I wrote about this “Kapitsa’s pendulum” effect during the previous discussion of Airpeak:
https://www.dpreview.com/news/2550610046/sony-reveals-new-information-about-its-upcoming-airpeak-drone?comment=4134886250

What WAS unknown to me was whether the inertia of rotors is large enough to make the algorithm unimplementable.

Jun 13, 2021*permalink

@mmarian: > “even you can see that the video shows purely a scientific experiment with heavily modified quad using balancing weights and counterweights strategically positioned to mitigate the effect of one specific prop failure.”

Hmm, cannot you JUST READ WHAT IS WRITTEN THERE?!

Which part of

“… requires no physical changes or additions to the aircraft.”

is unclear to you?!

Jun 13, 2021permalink

@ilza
A see. I have to admit that I only watched the video and did not reed the article. So, my appology also to AstroStan. The drone has some sort of ball attached to the arms and there is also one on top which led me to believe that these are some sort of added weights to help stabilize the drone while spinning. If this technology is really viable I wonder why it has not been addopted and implemented by any of the current drone makers...

Jun 13, 2021*permalink

ilza: The beauty of the fast spin is that it allows even a single motor to essentially counterbalance itself. However, I suspect this requires precise control over motor speed combined with knowing instantaneous rotational position... which might not have been viable until fairly recently. Put another way, air is a pretty viscous fluid once your control loop time constants get small enough....

Jun 13, 2021permalink

@ProfHankD: my point was kind of different (though not written down clearly enough — sorry!).

① In the scope of the PROPER “Kapitsa’s pendulum” effect, no “control software” AT ALL is needed: the system would “stabilize itself” when “the modulation frequency” becomes high enough.

② My hope¹⁾ was that even if the system is not in the scope of “such proper Kapitsa’s effect”, still a morally-similar-to-Kapitsa arguments may shows that even minor possibilities of control may lead to a complete “stabilization of the system” when one spins it quick enough.

¹⁾ This was half-a-year ago, before AstroStan found the PROPER references. — BTW, thanks to AstroStan!

Jun 13, 2021*permalink

@mmarian: this is how “complicated math” percolates into the real world! As one person said: “as time goes, deep things float closer to the surface”.

After the “software developers” internalize these algorithms, this is going to be the new lingua franca of the game. Until then, “the scary factor” rules the ball.

Jun 13, 2021permalink