Your right. I have that in my checklist for rapid loss of altitude. It's a critical action procedure, it is treated like a stall recovery. I treat the first 3 steps as the most critical and assumes ground impact is imminent. I treat the fly out maneuvering as near critical, because their may not be time to fly out. The last 3 assume that time is available for the fly out and stabilization of the aircraft. Here is my entire procedure for it. You will notice the camera operator is instructed to stow the gimbal. I brief them to put it in the furthest tilt up position possible keep the tilt stack as far back as possible and not to remove their finger until instructed to so. This is in preparation for ground impact in hopes to mitigate as much damage on the camera as possible (p)'s are pilot actions and (c)'s are camera operator actions. Uncommanded Rapid Loss of Altitude (Altitude Hold on) 1. Altitude Hold – Off (p) 2. Throttle – Full (p) 3. Gimbal – Stow, Full UP(c) 4. Air Speed – Gain (p), as required 5.Hover – Stabilize (p) 6. Battery Level – Check (p)
@Andy, I have question. With the 2.2 do you notice the Variometer on your HUD/OSD representing the ACC with the Baro? Or is it still responding to the affects of the Baro alone? I am curious if the data is representing the inputs for the compensation when AH is on. If not, I can see this as an issue because you may end up erroneously responding to inappropriate data. Shaun
Hi Shaun: Hmmm. I'm not sure how I could tell which it was responding to. The variometer value is calculated based on two consecutive readings of the altimeter as I recall -- but I'm away on vacation and don't have access to the machines on which I prepared the Rather Good Beginner's Guide to MK_GPXTOOL (which does have the formula in the materials). I'll send myself a note to look the formula up when I'm back home. Andy.
Here is an interesting link on how a similar 2 Axis acceleromter works. The one used for the 2.2 is a 3 axis, but based off of MK's description as well as the official data sheet the principle is the same. http://www.parallax.com/dl/docs/prod/compshop/SICMemsicTut.pdf
Shaun, I've been noticing the wobble and it's good to know the science behind what's happening. Thanks for taking the time to write all of that up! I have a less complicated question Land Gear: What are you guys using to color your landing gear? Is it a reflective paint I can get at Home Depot? Or is it a yellow sticker you're cutting with an x-acto knife?
Get some florescent Gaffer's Tape. It's available in Orange and Green. For example: http://www.amazon.com/Polyken-Premium-Fluorescent-Gaffers-Tape/dp/B000QE9OW0 Then you take off the landing gear, take the two side legs off, and lay them on the bench, stick a length of tape along the "outside" surface, flip it over, and using a sharp X-acto knife blade (don't be cheap -- use a new blade! ) cut around the profile of the side leg. One side will have the threaded inserts on, so you can cut around those. Hope this helps. Andy.
Yeah....it's a helluva lot cheaper than paying for the gas to go down to the local stores, I'm afraid -- and, if you put any kind of value on your time, there's a double saving there! Andy.
There are some vibrations and bumps I'm hoping to get rid of but could be an issue of flight speed (~15mph) or pan balance (roll and tilt are balanced spot on). I could just be over critical too, but there are a few points as I am approaching a slower airspeed, bumps and vibs hit the cam. Beginning to understand air's effect on the blades (Shaun) I think you can expect vibs/bumps. I am using red/black/red/black/red/black "O" rings on my vibration dampeners. Also I am using default radian settings with no deadband (I found no deadband to be smoother in my opinion... but I could be wrong!) Also at night, it seems like the roll puts the camera at an angle during a few turns. I think my roll gain is 90% -- I'll have to check! These are a few shots from last night. 5DM3 - 30fps - 24mm
Here's the formula for how the variometer value is calculated -- bear in mind that the Previous Altitude means "the altitude as it was a second ago": New Variometer = (PreviousVariometer + (2 * (CurrentAltitude – Previous Altitude)) / 2 Line 613 spi_slave.c (NC)] Andy.
EDIT:Actually its does NOT look that bad. I don't think its possible to get rid of bumps 100% on this setup. O Rings, there's a whole discussion on O rings and the hokus-pokus involved. Unfortunately, I have not found the right O-ring recipe yet. Some say all blue, or all black or blue black and red. You can dial down the tilt gains and play with then a bit and see if that helps. You can try dialing down the tilt and roll gains a little. Make sure all of the pulley belts are tight. I use 70 on my tilt and roll gains and find that if I am not racking the AC around it seems to do a descent job. Everyone will have their own magical number that works for them. You will find that CS6 Warp stabilization will do a good job smoothing out the footage as long as there the pan rate is slow and controlled and the jitters are not too aggravated.
That's interesting. It looks like a smoothing function. I guess to keep the Variometer from being too erratic. On that note the next question is... Are the functions CurrentAltitude and PreviousAltitude considering data from both sources, the Baro and ACC in the calculation? Or what I am trying to ask is the ACC normalizing the data from the Baro? Or is all of the altimeter data still only a strict barometric function as it is with 2.1? Shaun
Which is what Holger stated without the math on the Wiki. To a small circuit board to be soldered with an acceleration sensor on the FC2.1 because the accuracy of the old Z sensor is not enough (under vibration). The ACC Z value is fused into the altitude level and the climb/descent rate (Vario). This makes the measurement of the altitude many times faster, which is beneficial to the entire height control. The MK can respond to fluctuations in the level already before an apparent height change occurred.
Beginner question... The heat sinks sit VERY close to wire and solder when glued on top of the black squares (BL). Under high power demand, will the heat sink (metal) conduct a solar bridge by jumping from one of the wires to the next via the heat sink? Thanks guys!Zach
That's why placement and not getting the silver epoxy on anything that can conduct electricity. You are correct that all of these parts are close together. But as long as you don't have any stray wire hairs you should be ok.
Just as an aside, I tested the Arctic Silver, and it is not conductive to electricity. I was pretty surprised as it's a suspension of silver particles that conduct heat, but it has a very high electrical resistance -- that's not say you might not get a batch that will conduct, but what I had, when tested with a Fluke volt/ohm meter, did not have a measurable resistance. Andy.
Thanks guys for your time... 1. Should I run a silicone pass on the circuit leads before I glue the heat sinks on? Also, a real quick flying question: I was shooting some footage today. My POI was about 300-400 feet away, 100 feet high. There were a lot of surrounding obstacles, most 40-60 feet high. My max distance from the copter and myself was 500 feet and 120 feet high. Average speed was about 9 feet per second. 2. Is it normal to feel uneasy at these distances? Especially as a beginning flyer? The copter was difficult to see and distinguishing depth of field was very difficult, I'm wondering if it would be much safer to fly with a pilots FPV. There are optical illusions I notice at high altitudes. This emphasizes why I would think an FPV would be much safer and less nerve racking of piloting experience. 3. Also -- is anyone recording with waypoints anymore? I know it's a old feature but I was wondering if it accelerates from point to point at a smooth rate.
No need for silicon. For the flying question yes, especially as a beginner it is hard to get the depth perception thing figured out. You need a lot of practice and a lot of communication with your camera op. They are your eye's when your that far away and trying to fly proximity with objects. If you don't feel comfortable, don't do it and move closer. Josh