Dear All,
I was using Cannon S100 previously for mapping purpose with CHDK, 1/2000 shutter speed, IS turned off and focus set to Infinity but still get around 50% blurry image.
Now I will use Sony Nex 5n camera on Flatmap1 gimbal with 16mm stock fixed lens and IR triggered from APM for more ground coverage and quality image (hope so..)
Please share your Nex 5n camera setting (Auto or manual or shutter priority?) and any experiences, pros & cons using Nex 5n for mapping purpose.
Let me start from what I was reading from some forum (forgot the link):
Sony Nex 5 series Camera Setting for Mapping:
- Mode: Aperture Priority (by Krajst)
- ISO: 400 or 800 (by Krajst) (if you want faster shutter speeds, you'll have to increase your ISO) check: http://www.wikihow.com/Use-Your-Digital-Camera%27s-ISO-Setting
- White balance: Daily/Sunny or Cloudy (by Krajst)
- Sharpness: infinity (but it is tricky with 16mm lens so I use auto (by Krajst)
Recommendations:
- IR Shutter from Hobbyking (by iskess)
- Use high speed SD: SANDISK EXTREME PRO SDHC SD 16GB, 95MB/s (by iskess)
Pros:
- More pixel compare with S100 = More ground coverage
- Fixed lens 16mm = less blurry image result
- ??
Cons:
- Heavier compare with S100
- No GPS onboard
- Nex 5n Time between shutter > 2 sec = S100 faster < 2 sec (if you need to fly really fast)
- ??
I will update first page regularly that may be useful for others.
Many thanks in advance for your contribution.
Waladi
Replies
Hi Arga,
can you please complete your info, which lens, AGL, speed?
thanks
Wolfgang R.
Using sony 16-50mm set to 16mm, flying 140m agl on 14 m/s (actual speed 12 m/s). Using highest resolution on jpeg (using raw will kill my computer).
Total got 865 photos for 215 Ha area. Post processing took 12 Hours on my monster laptop (core i7 3.2Ghz octacore, 16Gb ram, 4 Gb graphics card).
The photo's exposure is really consistent (usually got slightly darker or lighter between images on Aperture or shutter priority). Focus is a bit blurry due to drift focus on flight (forgot to put a tape on the focus ring), but the post processing result is really satisfying.
I'm using 9 gcp surveyed with static rtk gps.
thanks, great info
that's complete ;)
Wolfgang R.
Hi Waladi,
The following info may be of interest to plane users.
I wonder about the low shutter speed most of you can
live with. Are all of you flying copters or turtles ;)
I finally got a used NEX-5R with the Sigma 30mm f/2.8.
Now I can compare between the Canon SX-260HS, a Sony A7R
and the NEX. All wing mounted to my Skywalker 1900.
The A7R I had for a weekend only and it's far beyond my
budget. But a major step is going away from the Canon to
the NEX size. I run a few missions with 600 shots each.
No single blurred picture! The Gentles IR trigger cable
works fine for me. For my missions the missing CHDK isn't
an issue. The distance trigger of the APM 2.5 is sufficient.
For the cam I setup everything manually. Today I choose 1/2500s,
ISO400 and f/4. The pictures are taken in RAW format only. I'll
hope this will allow me to fix low light issues later.
The Skywalker is going for up to 15m/s at least and even by
low wind I've to calculate for 20m/s.
As I'm concerned about the shutter speed my calculation is
as follows. The nice Pteryx whitepaper mentioned to go for
shutter speed related to 1/2 pix or even better use 1/3 pix.
Now my math
AGL: 100m =>
pixel (height): 52.2m/3264pix =>0.01599m (measured, not calculated)
pixel (width) : 81.7m/4912pix =>0.01663m
1/3pix size: 0.016m/3 => 0.005333m
15m/s: 0.005333m/15m/s => 1/2812s
Some variations for AGL and speed
1/ ((0.522 * 100 / 3264) * (1/3) / 15) = 2812 [1/s]
1/ ((0.522 * 120 / 3264) * (1/3) / 15) = 2344 [1/s]
1/ ((0.522 * 100 / 3264) * (1/3) / 10) = 1875 [1/s]
1/ ((0.522 * 120 / 3264) * (1/3) / 10) = 1563 [1/s]
(http://web2.0rechner.de/ for copy&paste)
It's hard to judge the subpixel setting but for me it
sounds reasonable to go for high shutter speed.
A second point is overlap.
The paper mentioned here to go for 60% or better 80% (3d modelling).
For a plane this isn't that easy. Here the calculation
overlap: 60/80% height: 100/120m speed: 15m/s
(1-0.8)*(0.522*120 )/15 = 0.8352 [s]
(1-0.6)*(0.522*120 )/15 = 1.6704 [s]
(1-0.8)*(0.522*100 )/15 = 0.696 [s]
(1-0.6)*(0.522*100 )/15 = 1.392 [s]
That needs a closer look at the write speed of the NEX (-5R).
From what I found on the net it seems that the limit is
somehow around 45MB/s to SD. I bought a
SanDisk SDSDXN-032G-FFP Extreme SDHC 32GB UHS-I Class 10 U3 60MB/s.
A test on an 80MB/s did not give any better results. My
measurement for the 60MB/s shows 0.89s/pic. This means you
can save some money compared to the mentioned 95MB/s. But
be careful and look at the write speed. The math shows that
trying to go beyond 60% overlap is asking for good calculations.
A very nice point I didn't expect is the low power consumption.
The 600pics consume less than 20% of my battery. That means
I can go for up to 3000pics in one mission. This is much much
more than I did expect.
Be aware that my figures are from some very first missions and
not collected over months. But I thought I should contribute a bit.
HTH
Wolfgang R.
Hi Wolfgang
How do you find the wing mounted camera? Any pics would be great.
I've done that on my Maja too, but obviously it creates a lot of drag, and I feel like I have to add weight to the other wing to balance it (extra weight = more drag).
How did you decide how far out to mount it. I planned it so that I could bank at 25 degrees and not get the fuse in frame. If I was doing it again I might bring it in towards the fuse a little, as it never needs more than 15 degrees bank except when turning at the end of passes.
Hi moglos
I do not use a gimbal but simple velco strips.
For me the results are really fine.
But of course there is a major downgrade of the Skywalker
performance. I need some higher cruise speed. I fly
it for a year now either with the cam or an identical
dummy like seen in the picture. Note that I've a Sigma
30mm. The lense center is about 8.5cm from the fuse.
You see the counterweight on the second pic.
With my 5000mAh/3S I'm close to 30min. But I plan to
migrate to high power Panasonic cells which should give
me 1h flight time. For 2h I'm awaiting a 3m Skywalker.
I was a little skeptical at the beginning but now I feel
it a perfect low cost solution. I tried a Vamp before
and have two Maja pilots next to me. The Skywalker is
so much easier to work with.
Wolfgang R.
P1020060.JPG
IMG_0683.JPG
Looks good. I see the fairings for the camera too, that should help.
I wondered for a while if it would be worth cutting a relief in the wing to allow the camera to fit in flush (or somewhat flush). I even did some tests and found the wattage saving was 15%. But I wasn't brave enough to go through with it and actually cut my wing.
I use the Panasonic cells, I think they're great. Note that they only have a 2C rating, so you need to take care not to use full throttle for very long on takeoff. 2C * 3.4mAh * 3P = 21A max.
How exactly did you get these 15% values?
Wolfgang R.
@moglos
yes, I'm aware that this is the downside of the high capacity
batteries. But I'm sure I can stand away from their limits.
For a couple of months I flew a tiny Hacker with 14A max for
this payload. This motor was a bit too small but only little
more power is needed during launch.
@Martin
I wouldn't think much about modifying the wing. A recent article
about the atlantiksolar convinced me that a streamlined hanging
box works very well. It is a matter of minutes to have a reasonable
working version.
For the Lipos I think about two 3S4P (2x6400mAh, 600gr). A bit more
than my Turnigy (420gr). But this is for next year and I will check
what 4S is giving me more. In the meantime I'll hope that a large
cheap plane comes up. That would help more for efficiency.
Wolfgang R.
Anyone tried mounting the camera on top of the wing and shooting down through it. How would this affect the flight characteristics?
It might be a simpler method of mounting.
@Wolfgang, did you think about going up to a 5000mAh 4s batteries to increase your flight time or even 2x5000 mAh?
Thanks.
-
1
-
2
-
3
-
4
-
5
of 6 Next