FIrefly6, Mad jelly.

I dunno...I think the picture quality/color reproduction might be better with the Yi. I do like that I can configure the Firefly without needing an app, and I want to try the Gyro feature when mounted to my gimbal.

I've just gotten the gimbal reattached. With the 12v BEC, my voltage to the controller has increased from 11.65v to 12v. I had hoped the difference to be insignificant, but it has introduced buzzing in some positions, so I'll need to tweak the setup...maybe just decrease vmax.

From what I've seen of the Firefly it looks much more crisp than the Yi. Do you have a retina or 4k display? You can clearly see the difference on a high-dpi display.

I totally though you guys were talking about this, http://www.birdseyeview.aero/products/firefly6 , Not an action cam.

I have a 30" Retina display. I thought the one picture I took was murky...dull colors. Of course, I'll keep playing with it, and who knows... maybe it'll get airborne this Spring.

Did you record in 1440p?

No, I just snapped some photos. I gave it another try today and I take all the criticism back. The photos were crisp, the video (just handheld) was very clear on the 2.5k video at 30fps.

Looks like the larger one might have an LC filter?

Ah, after Googling, I believe you are correct. Thank you! So, the larger one will clean up power supply noise, if I see any of that, in a FPV/video transmission setup.

Sounds like I need to order some FPV gear and experiment! :wink

Would definitely be nice for framing your shoots.

Well, that experiment was short-lived... while the smaller sized regulator performed well, I just started having too many accessories glommed onto it's output pins. I probably should have done this to begin with, since this is a 4S system, but now I will connect a 12v BEC from the main power board to a mini-PDB from which I can then more easily connect gimbal, LEDs, vTx, FPV camera, ...

Interesting idea. I haven't seen a multirotor with wooden props before.

There's a few videos out there, like this one, but none to show pros/cons.

I'll probably end up tossing $35 to find out for myself...

They look quite nice. I wonder how the rigidity will affect the flight.

We're going to find out in a couple days... Curiosity got the best of me... I ordered a set.

Awesome! That should look pretty nice too. A little old school and new school in a single build.

Wooden props!

I love the way they sound! Mine are due to be delivered tomorrow.... but it'll be so much longer, if ever, until I can fly like that!

Awesome, looking forward to seeing how steady you can get your video. They do sound quite nice, but I'm not even envious of that kind of flying. I can't imagine how many broken parts he had to go through to get to that point. I'm happy with slow and steady.

First impressions? Wood sounds AWESOME! No more bees! More of an extended "Zing" sound.

I just mounted them up quick without balancing to get a flight in before the rain comes. Jello appeared in the video, so I know balancing is needed; I'll do that tonight. I also heard some oscillations, so another auto tune is in my future, too.

Ugh, rain! And I was going to do an autotune this afternoon.. The wood sounded really nice in that video I shared a few posts back. Looking forward to seeing some pics and video. Speaking of which I need to take some new pics of my rig now that I've rearranged the battery and added the SunnySkys.

Props are all balanced now. 5 of 6 needed adjustment to the hub. Is it sad that I think it's cool to be able to use sandpaper? Only two of them needed adjustment on the blades. I still used tape for that. I don't really know why... old habit, I guess.

I hope they give you a good result on the footage!

Me, too, once I get the autotune done.

It took 4 attempts on 3 different days to get a workable autotune result... sheesh! Seemed like every time I attempted, there'd be some gusts of wind during the process and kept giving me overly twitchy outcome.

Video looks good now so long as the camera is in its default position. There's a few twitches which I think/hope I can remove with a slightly larger yaw deadband. I get a lot of jello when the camera is pointing straight down, though; I don't know what I'm going to do about that yet.

I found that the yaw deadband can only do so much. It's more important to tune the rig to hold a steady heading. What Stabilize Yaw P and Rate Yaw P settings are you using? I'm up to 6.0 and .30, but I think I need to take those down a few notches. 5.5 and .25 were a little too low, so I need to find a nice middle ground.

Oh, I think it follows the desired heading pretty well. It kind of twitches when, say, yaw left, the deadband lets it lag behind to a point and then seems to snap forward. I turned down yaw speed some time ago to 10 degrees/second which I thought would be a good idea at the time. I figure I was turning faster than that and then ran out of 5 degrees of deadband...just a guess, really.

My values are 4.95 and 0.15. I've been tempted a few times to upgrade to 3.3 to see what autotune would do for yaw (and then revert back to 3.2.1).

Why not just upgrade and keep it? I'd love yaw autotune.

Wow, I didn't realize how cheap those wood props are. Very tempting..

I dunno. I'm suspicious of all the release candidates...on rc9 now. 3.2.1 is stable and does everything I need it to, except for the re-written autotune...all 3 axes, configure it to just do one at a time if you want, etc. So far, I can wait it out.

Well, could be worth a try to upgrade, do the autotune and see how it handles for a couple flights. If you don't like it you can always revert back.

Wow, I just noticed what you saw...those wood props are on steep discount... Much more so than when I ordered a week+ ago. I might need to stock up on spares at these prices... Sorry honey!

Now that you mention that I'm tempted myself. That's less than I paid for my nylon/carbon props.

It'll be about $11 for a set of 4 11.4 shipped. Not bad at all. My only question is how big is the hole in the hub? Does it come with some sort of adapter to fit in there.

They sell a bag of adapters separately. The specs say the bore hole is 6mm. I believe my motor shaft to be 5mm (they use a M5 locknut after all). So, I bought these adapters that say they reduce 6mm to 5mm.

Well, when I received the props, all of them were tight fitting to the shaft. I had to screw them on..the threads were biting into the wood a bit. I never used the adapters...never needed to.

They're a good prop...I'll be ordering more today just because of the price. Very easy to balance, too. Most of the just needed a little sanding on the hub...just enough to remove the finish on one side.

My motor shaft is 5mm as well, so I wonder if I would need the adapters. Mind linking up the adapters so I can have a look?

A day later, 3.3rc10 is now out...that's moving fast! The thread over on diydrones suggests 10 could be last one before the official 3.3 release since it was only minor defect fixes from rc9.

Sounds like it's almost time to make a permanent upgrade ;)

Yep, almost... Through rc7, 8, 9 (and probably back to rc1 even) folks were reporting unexpected vibrations and strange crashes. I don't want any part of that.

Oh my! I wouldn't want any part of that either.

I just ordered these. After my little mishap the other day I think I might need some spare props at some point. Can't beat $2 a prop for something of this quality.

Yep, well worth it. I ordered a few spares Thursday morning; they were in the mailbox this afternoon...

On a whim, decided to swap the wood propellers for the APCs I used some 7 or 8 months ago. It's louder, a touch heavier, but somehow so much more efficient in flight. Two batteries today...19 minutes each!

Not bad! Get any good footage?

Some of it looks usable, yeah, even though it's not a very colorful time of year. I did discover a cool near miss with a bird in one of the videos. Will make it to YouTube eventually.

Looks nice in black!

Seems to be a newly available model/color option. I still needed to adjust the focus, though. Interesting to learn that it weighs 78g. The Sj4000 was only 55g...which might have been a factor in my video stability issues.

Oh my! Could it have been a short through the arm?

Amazing, it's taken me a month(!!) to get out and record a video with the Yi. It's nothing exciting... just went up, switched to Loiter when one of my neighbors happened by to visit, spun around and came back down.

This is with all the same PIDs I tried (unsuccessfully) to use with the SJ4000. It appears much more stable aside from some odd Yaw waggle. Maybe I can fix that with a change to the deadband and/or hysteresis values.

Looking much better! I can still see a little jello in there, but it's far better than the SJ4000 ever did.

I got rid of the yaw waggle. Changing deadband and hysteresis values didn't help. Turns out the motor wires were under too much tension... probably from a few of the times it spun like a poltergeist with the SJ4000!. There's still the slight jello, like you see in that video...mostly when I'm face on to the wind, so I think I'll try to stiffen up the pitch response and see what happens.

Awesome! You should be able to pull off some really nice footage with that setup. Can't to see how it looks when you've got it geared in.

Awesome, I'd love to see what kind of GPS fix you get with both of those running.

Right now, only one or the other is supported. The 6M seems to acquire satellites first, so Copter starts with it. Soon after, M8N catches up and I can get 11 satellites indoors, so Copter switches over. In my last flight today, practically as soon as I got outside 16 satellites were found and Hdop dropped as low as 0.86.

Do you look at uavforecast.com? For the 2:00 and 3:00 hours today, it reported Kp=0. I don't recall ever seeing a value below '1', so I arranged some free time and charged up a battery. The difference was amazing! GPS #1 (the 6M) found 11 satellites and a hdop of 0.86; GPS #2 (the M8N) found 16 with a hdop of 0.73!

I also got to experience what happens when the Pixhawk decides to switch from one GPS to the other. There was a musical tone played and then it rapidly moved about 10ft west. I was still below rooftop level when it happened, so it was a little unnerving!

Afterwards, now looking at the logs, the two GPSs never agreed on the same Lat/Lng (I guess from the difference in precision). I'll have to mindful when flying at home to not rely upon a GPS mode unless I'm above all obstacles.

Interesting, so hopefully the software update does something about that.

I found a fairly recent discussion thread where another user had the same experience (a rapid change of position) as I. The tone of the conversation seemed as if the devs were surprised by that behavior and then they launched into a debate of how best to manage the transition (I could probably re-read what they were talking about a dozen times and never understand it).

In the meantime, I figure there are a couple things I can do to limit my exposure to it. One, leave it on the ground longer to allow for the switch before arming. Two, not use any GPS-dependent modes until safely above all obstacles. Three, unplug one of the GPS'es if options one or two isn't possible.

I'd probably avoid GPS modes unless high enough. I haven't even been using my GPS lately as I find stabilize much more predictable and fun.

I didn't know about the RCOUT value. So that indicates the amount of power each motor is consuming?

In a way, yes... It's only able to monitor PWM signal, though. Higher signal = more demand from the flight controller = increased amp pulled by the motor.

In a flat hex, quad, tri, you'd probably never need to look at RCOUT...just assume its all the same. In the coax layout, there's a lot of fluid dynamics at play and, from the sources I've found, its really not all that well understood at the RC scale.

What I have learned is that lower propeller is subject to a column of accelerated air from the upper...and that accelerated air reduces the effective pitch of the lower motor. And now I'm learning that the separation distance between the propellers has some influence, too apparently. These T-Motors aren't as tall as the others, so the props are probably 8-10mm closer together vertically.

Super fast shipping from Master Airscrew! 11x7 propellers arrived in today's mail.

Hope they work well!

Took me a while to make the time for balancing the props, and today's gusty winds probably aren't the best testing conditions, but the new propellers definitely do make a difference. I note that in the prior configuration, Ch5/6 averages were 1529/1709. This morning's hover with the new configuration returned results of 1439/1559. I'll do more testing, but at the moment, I don't think 11x8s are needed to get the outcome I'm looking for.

Hope to see some footage soon!

Me too! For however long I've owned that Yi now, I think I have 9 minutes of video! I'm getting a little impatient with all the tinkering (all the while, FPV stuff is sitting on the bench waiting to be wired and mounted).

Going to attach your monitor to your TX for FPV?

I'll use AV Out from the Yi to the TX. Before I add that I want to redo my 12v feeds.

I bought one of those mini PDBs Ive seen on other builds here. I'll feed it from the main PDB and then use it's built in 12v regulators for the LEDs, gimbal, vTx and 5v regulator to power the Yi.

I mean your Taranis TX. What will you use to watch the video feed? The Yi does lag slightly, but it's not horrible.

Oh, now the question makes more sense...! I have a bracket on the Taranis for mounting the monitor. I don't really like it...too heavy, gotta use the neck stap, etc. Time to put together a ground station with antenna tracker? Lol....this hobby never quits!

Might as well just buy a headset. That's really the way to go. I have a couple linked up on my proposed 210 build that seem like a good value. I'm waiting for more reviews on the higher resolution display. It looks promising.

Maybe so... I figure I was collecting so much gear I was going to have to take along a second person to carry it all out to the field. Before it all has to be lugged back, they might as well pull up a chair and watch! :-D

Antenna tracker seems like a cool thing to put together, but the crazy long range it allows for is probably only achievable with a plane.

Strange, I've never had that happen to me. Usually some form of input would cause a pause, but that's not the case here.

Next opportunity I get, I'm going to try it again... see if it's reproducible. There was no pause between pitch twitching/tests and the yaw.

I'd probably just discount it as a glitch since AutoTune isn't something you do all the time.

I couldn't stand not knowing... :-) I re-ran AutoTune and observed no glitches this time. It's a good thing they added the feature to do one axis at a time... trying to do all 3 at once takes ~15 minutes, leaving very little time before the battery warning to try out the settings it comes up with.

That's a nice feature because I always worry I'll run out of battery before the cycle has completed.

Nice, what are you going to do with the old motors? Maybe build an S500 =D

Motors didn't sell on RCGroups or eBay. Now, along with a set of ESCs, I guess I'll be "forced" to build something like the Greek Drone

Doh! Well they're looking pretty nice. If you change the ecalc links to read like xcoptercalc.php?rotorbuilds instead of xcoptercalc.php?ecalc those links will work for everybody.

Awesome, looks like they work. So you'll get 3 more minutes of flight time and a little more thrust.

Putting spacers within the arms was a lot more difficult than I had planned for. I couldn't hold the spacer in the arm with needle nose pliers like I expected. I ended up fishing the spacer into the arm on a bent paperclip. Then securing the spacer in place with a partially inserted screw from the top. Which then held it in place just well enough for me to run the main bolt through the lower plate, into the arm and into the spacer. As that bolt advanced I would back out the screw from rhe top. Figuring out the process and doing it yielded 2 hours to attach one arm and then the other two went more smoothly at ~30 minutes each. Good thing it's not a quad! =). Most importantly, though, it served it's purpose. I cranked down on those bolts and the arms did not buckle under the stress.

Next up: plugging the motors to the ESCs...simple enough.

At least it did the trick! Working on these things is always a test of patience and when you've got a family it's not just your patience that gets tested ;)

"simple enough"... famous last words! I blew two ESCs! One gave up a little puff but it would still twitch the motor. I verified that the motor wasn't damaged {whew!} by connecting it to another working ESC. The second ESC had an ulterior motive to pop, blow off it's heatshrink, smoke, and likely catch fire if I had let it get that far.

EDIT: It was an oversight to not check the motors attached to both ESCs. While the twitching motor turned out to be OK, the motor attached to the really burned up ESC was shorted.

I learned a pretty interesting and simple test for the motor: When disconnected from the ESC the housing should turn pretty freely. When you join 2 of the 3 motor wires together, there will be a noticeable resistance to the turning...join all 3 to feel even more resistance. Attempting all combinations of the 3 wire set, if any did not give resistance, then there's a problem.

Another test I ran was connecting the motor to a drill, and a pair of the motor wires to a multimeter. If all is well, while powering the drill, the rotation should produce a small voltage that can be read by the multimeter. Test all combinations: no voltage is no góod.

What to do with the busted motor? Well before I chuck it in the garbage, I might give a try to rewinding after watching a few Youtube videos...

I suspect that the heatshrink I staged over the connectors (without melting) must've slipped off the connectors and shorted with one another.

Now, my next question: I ordered 6 ESCs from a US-based retailer (so I can get them quickly). Do I need to replace all 6 to match, or can I replace just the two failed ones?

Are they the same as the originals? You generally want the same across the board, but due to the nature of the Y configuration maybe you can get away with using something different in the back?

I will likely just replace all of them for peace-of-mind but "same as the originals" is what got me thinking, what does 'same' mean? There are 100s of vendors selling ESCs... I can't really believe they are all manufactured separately/uniquely. I want to believe that one 30A SimonK ESC is the same as another...

If they're similarly branded you might be okay, but if you don't mind the extra expense you can replace them all and sell the originals to somebody building a quad.

Taken it out for a test flight since these upgrades yet?

Not yet; patiently waiting for no tornadoes, gale winds, or cold temps...Sunday, maybe?

Sounds good! I'm currently traveling so no flying for a little while here.

Hovered for about a minute and a half yesterday before packing it in. Too much wind and gusts to really feel comfortable or enjoy myself. I say 'hovered' but, really, once it was above tree top level, I just pushed it south and let the gusts blew her back to me. Still patiently waiting to really shake it out and test the extended flight times....

At least you know it's working!

Yep, seems to be working pretty much to eCalc spec. I stepped outside yesterday afternoon and noticed it was practically windless so I grabbed a fresh battery and up she went. After a few laps around the property and an Autotune cycle, I landed at 80% consumed, 16:23. For 80%, eCalc predicted 17:24 strictly for hover so I think I could best that by some seconds if I ever wanted to just sit in a Loiter for that long (boooring!).

Do you use Tower at all? I basically rely on it for my voltage readings.

I use Tower for plotting Auto missions only, now. Everything else it does I get on the Taranis display with the added benefit that I can program feedback features into the controller. E.g. alerts for when battery remaining is <50% and Dist from Home > 200m... Or when current > 70amps sound a warning... Or when in stabilize mode and vertical speed is < -3m/s (falling too fast), trigger the lost copter alarm... Just a bunch of things that aggregate data.

Very cool. Sounds like the tip of the iceburg for what the Taranis can do. BTW, I updated the discussion layouts here to hopefully make them more readable. It'll always default to showing only the last 2 comments for each thread.

For sure! That's all simple a<x expressions stuff. I haven't gotten into the Lua scripting language, yet. You said it once before: We're flying computers...with computers. Its fun to tinker with!

Neat, I've never thought to use strobes. Do they show up in your video though?

No sign of it in the video...hmm, though I haven't tried it with the camera pointing straight down.

What I really like about the strobes is that they'll grab your attention especially on a cloudy day when it can kind of fade into the sky at high altitude.

With extra time on my hands, I programmed my Taranis to assign switches to channels 9-12. Now the strobe only fires when switched on.

Great! I picked up a Turnigy 9X during the HobbyKing holiday sale, so I'm looking forward to adding some extra channels to my rig. Maybe I can finally add my LEDs and do a toggle as well.

Oops! I mistyped. I meant to say Turnigy 9X (Got it for $30 including shipping). The Taranis is too far out of budget for me. I'd rather spend that on a little 210 FPV rig.

Sounds like a good project, how about some pics? 128g is a bit of weight, but your rig should be able to handle it. Did you end up installing those thicker arms yet?

Yeah, those stripes might need a redo :) Covers everything up nicely though!

Looking good. How are those props treating you? I used mine for a couple flights and they did well, but I had a take-off flip which broke one and scuffed another. After replacing the broken one my jello was pretty bad. I need to balance them. My other problem is my shaft is too small for these, so in order to get them perfectly centered I need to wrap tape around the shaft to get a snug fit, which I didn't do. In the mean time I've reverted to my carbon mix props.

This prop combination is the best I've tried...and I've experimented with a bunch of them.

When I first ordered them, I bought the aluminum sleeves that reduce the hole size from 6mm to 5mm. As it turned out, I never needed to use them, so they're in the spare parts bin. The props fit snugly on these motor shafts... the thread even bites into the wood a little. I have discovered that they all need balancing; I wouldn't run them straight out of the package like some other brands claim is possible.

I'm really fond of the prop I'm running on the lower motors from Master Airscrew. They have yellow painted tips, which you can plainly see the rotor outline when I'm flying low.

Mine didn't fit snug on my SunnySky motors. I tried to center them as best as I could by resting it around the shaft, but that's far from ideal. I'll revisit them at some point, probably when I break one of my carbon mix props.

Could you invert the prop washers? Maybe that cone shape on the washer could self center the prop around the shaft?

That's an idea! I'll try that next time I pull those props out.

You're making me wish I had Pixhawk. I'm starting to realize the limitations of my 8bit APM FC.

You'd especially like the improvements in Yaw. My old 3.2.1 logs would show some jittery behavior when graphing Yaw vs. DYaw. I did the firmware upgrade fully expecting to be able to do a yaw autotune to smooth that out. Well, I haven't done it yet, but the latest graphed Yaw vs. DYaw was super consistent and stable. Autotune not required, imo.

I'm waiting for the camera battery to recharge to see if there's an observable improvement there, too.

That sounds great, looking forward to seeing how your video turns out.

Wow, that sounds great! The video can use some work, but I love the sound of those props.

The issue with the sunset shot is... The sun light was too bright, which caused the camera to adjust it's exposure, hence why the tree area became dark. These little action comes have poor dynamic range, so it will exposure for the high or the low (pick one). You can play around with the exposure compensation, but I doubt it will help much. The metering mode of the camera will determine what is used as a reference point for it's exposure. It most likely some sort of evaluative metering mode, but I doubt it can be changed. Either way, experiment with the camera angle during sunsets, so the camera adjusts for whatever you're shooting. :)

Thanks! I know how to get that shot with my DSLR but I guess you're right..the action cams are over simplified. The SJ4000 has a HDR setting, too; I want to try that next time, too.

The Xiaomi Yi has far better dynamic range than any of the other budget action cams. I can tell a huge difference over my old Mobius.

I'll probably just start with red/green like a boat/plane setup. Orientation can't be any easier than that, right?

I bought a 3m sticky strip of RGB 5050 LEDs for about $4 on eBay. I haven't had a chance to install them yet, but they're quite bright.

Did you get 5v or 12v ones? I first thought about the 5v option and power them from the output rail of the Pixhawk. Then I figured the 12v would be brighter; I can power them from the same voltage regulator that feeds the gimbal. Sitting here, now, though I think I also want a small switch to turn the lights off during the daylight hours. Are they so bright they might bleed into the camera field of view?

They're the 12v ones as well. I just added them to my build. I was thinking the same concerning the camera, so I guess experimentation will be the only way to know for sure. It seems lots of folks use these though, so I don't imagine it's a major issue.

Aww, I saw that sale but I wasn't sure about buying it. Maybe it just wasn't meant to be. I wasn't sure if my rig could handle all that weight.

Yeah, I bought one... For $27-28, I tried to ignore it, but failed.

I really like the Y shape... I've been thinking about a 250-sized build over the winter, and maybe that'll be a true tri-copter.

I've been reading about a lot of folks having issues keeping a straight trajectory with ti-copters. Apparently they're prone to yaw wag. That may not be the case all the time, but it seems to be a hurdle some have had to cross.

Yeah, I've read about that, too.... there's probably even more topics on a reliable servo to operate that tail rotor position. No doubt, in my mind, that the two concerns are related.

...and I do agree, a 804g battery would be quite the paperweight for your 450.

I did ask on the reddit thread concerning that battery and the consensus was that it'll work on my F450, but I'm still dubious. I think I'll opt for a 5000mah. Another idea is to run both of my 2700's in parallel.

I would be dubious, as well. I don't recall exactly, but I thought the thrust of the SunnySky 2212s was around 800 or 900g each. That'd be like asking one motor to do nothing but lift the battery, no?

I had one of the batteries in question, and bought 2 more when they were on sale last week, love em.

Oh, but be certain to reset the AHRS_TRIM_X/Y back to 0

I saw that as well. I don't have any issues in the steadiness department, but I plan to give it a try at some point anyway.

I didnt either... I thought, but it made a huge improvement.

Good to know! I'll definitely give it a try then. I think I might be able to use some juice bottles..

Ah, I look forward to seeing it! uavforecast.com showed a high Kp index for parts of Saturday and Sunday... any issues with GPS lock?

No flying for me this weekend but I was visiting with my sister who shared she's picked Natural Bridge as her wedding venue. I told her I wanted to get with the photographer about taking some aerial photos/video before and after the ceremonies.

Yes, I had some GPS issues yesterday in the mountains. I was stuck at HDOP 2.3 and it only dropped to 2.0 after several minutes. It was trying an RTC any time I enabled PosHold, so I kept it in AltHold. The GPS was great on Friday though.

I'd love to see Natural Bridge someday, but it's far too pricey. Looks like a great place to get some aerial video though!

The battery alone is right about 1/3 of my total AUW. I'll get 24 minutes running to ~20% capacity remaining. Personally, flying that long in a single session is exhausting though I've done it a few times now. It's much more common for me to take it out for many short duration flights, and just charge less frequently. My little SkyRC b6 charger takes nearly 3 hours to charge this thing!

I haven't added it to the parts list, but I started with an 8000mah Zippy from HobbyKing (16-19 minutes), but I found the weight of this one to be the same... 845g if I recall the specs correctly.

I know what you mean. I have a friend who told me he won't get into multirotors until the battery life increases. I told him 10-15 minutes is plenty. The stress of keeping such a large investment in the sky for so long wears at my nerves, so it's quite a relief to have a safe landing.