Long time lurker, first time caller-
just built a quad with the prebuilt APM2 and the DIY drones kit and Ive been having issues everywhere that dont seem to be fixed by calibration, re-calibration, reinstalling or cursing. SO before anyone asks, yes, Ive calibrated the escs 30+ times, reflashed APM multiple times, recalibrated the tx and kicked the dog 50+ times. On the plus side, i havent yet busted anything, but then again i havent managed a real flight either.
The first major issue occurred due to the lousy motor mount methodology of 2 screws through the arm. It appears any torque on those screws which id call safe or responsible resulted in torque being aplied to the motor shaft causing binding. This took me a bit to figure out, but eventually i bit the bullet and used the x mount that came with the motors and secured all 4 screws to two cross bars underneath the arm largely reducing or eliminating any binding issues with motors.
The second major issue came after syncing ESCs collectively and individually. At least 10x id boot the copter and it would start the incessant ESC beeping that happens when it feels the ESCs need recalibration, so every time id recalibrate only to have it happen again a few flights later or when i reflashed it.
The second major issue appears on takeoff, where the copter will oscillate as it appears engines are not equally throttled. it seems to recalibrate enough to takeoff eventually but then fairly consistantly, will spin to the right, at varying speeds. Ill set it down, and spin up again and it seems to have stabilized itself for a moment or two, but then will drift while the motors cycle and not maintain altitude.
I am currently in stabilize mode, declination set to 6 decimal places, calibrated, lowered, louvered and two failed flights from embarking on a three state murder spree.
TX spectrum 6i w/ ar5000 rx
channel 5 is set for mode toggle ( ch7 in MP )
No other electronics or systems on board though i do have a 1.2ghz FPV tx and camx as well as inline OSD which i plan on integrating later.
There are multiple things going on here, obviously, but most issues seem to be in the controller or escs. Any help would greatly appreciated.
Bart
Tags: calibration, esc, hovering, not
Permalink Reply by Vernon Barry on June 7, 2012 at 11:56am Ok, I had massive problems with ESC calibration and while the fix is not easy, once it's done life is much better when you never calibrate again and never have the beeping.
This post outline the single best fix you can make and should be compatible with your hardware
http://diydrones.com/forum/topics/flashing-3dr-escs-or-rc-timer-esc...
Next, after you fixed the ESC problem permanently
Are you running the latest arducopter firmware 2.5.5 that had a very important fix just for orientation problem?
Next, now that the ESCs are right and the Firmware is the latest and you leveled it
Now, this test with the props off:
That checks the wiring and the stabilization, but I need to make another video that if you pass the wiring test, you then need to pass the radio channel reverse test.
In other words, move to mid throttle so the motors spin evenly but the heli is level on the ground. Then when you move the right stick on the transmitter (AKA elevator) forward, the back motor should speed up (assuming the forward arm or apm direction is pointing away from you). Stick back = front motors speeds up, and moving the stick left (AKA aileron, not rudder) results right motor spins up and the opposite for moving the stick right. Finally, the left stick on the TX is throttle and rudder and if rudder was reversed you could never arm or you would have to move the stick down and left instead of down and right. Basically, the easy test again just like the tilt test is to run the motors up and move the right TX stick corner to corner clockwise and the motors opposite on the heli should speed up as you go around. If it's on the same side as the TX stick, then either the aileron or elevator channel is reversed. You can either fix it at the TX or fix it in Mission planner radio setup.
Just to recap, the problem is likely a couple of things and the first one is the stock ESC firmware sucks. Next, since it continues to rotate, I suspect either you don't have the ltest code which did fix a major problem on the new APM2.0s, or there is a wiring problem (motor order), or maybe it's just the ESCs and the stock firmware doesn't want to work for you.
Yes, you need some type of AVR programmer, but in the long term of this hobby, if you own even one Arduino, you really should have an AVR, because one day, you will kill a bootloader and then that is your way to bring it back from the dead.
Permalink Reply by Vernon Barry on June 7, 2012 at 12:28pm This sucks, the forum just ate my long post.
Permalink Reply by Jake Stew on June 8, 2012 at 7:50am Don't worry! The Ning team is working hard to make this forum better. They assure me that in another 10-15 years it will be almost as good as the leading forum softwares currently are.
Permalink Reply by Bart Mowrey on June 7, 2012 at 3:03pm Ahah! You my friend, have just clued me in to an issue i wasnt aware of re: ESC firmware.
I had updated the apm firmware last night to the newest available version so the heading issue should have been resolved, though FAIK the rotation was due largely to ESC screwyness. In truth, im probably a lousy heli pilot and couldnt tell you what behaviours truly result from Ground Effect and what from APM or ESC, but either way way i do know something aint right. I have verified motor speed and directional functionality long ago and as part of the battery of tests i repeat on a thrice daily basis. Now to find an AVR in town here....Hopefully that'll resolve it
Shortly after the first posting i was able to get the thing airborne and lose my orientation nearly immediately and had to set her down before all hell broke loose, as a moony eyed neighbor kid appeared right into my flight path. Most controls were responsive and not overly so, though it still had a directionality issue due to undirected rotation. Hovering was the initial plan.....
Plus side, I was able to set down in one piece and upright without Ginsuing the rotten little ankle biter.
Solution: head to the park behind the old folks home, at least i can see them coming, ever so slowly.
Thanks for the input and your time, Ill update this thread with results
Permalink Reply by Bart Mowrey on June 7, 2012 at 4:19pm For those in denver and surrounding areas, sparkfun has one
http://www.sparkfun.com/products/9825 and though i couldnt quite pickup this afternoon, I can get it tomorrow if i just schlep to Boulder, around 15 mins away.
Permalink Reply by Bart Mowrey on June 8, 2012 at 5:50pm Hi Vernon - i have gotten an AVR programmer from sparkfun but im confronted with the hard reality that im a barely functional idiot in this and most regards. I have zero clue how to establish the pin order from the AVR unit.
The link they had to the schematic 404s ( http://www.sparkfun.com/datasheets/Programmers/AVR-Pocket-Programme...) Do you happen to know what they are? FWIW i have in a tech support note to SF regarding this as well.
Thanks in advance
Bart
Permalink Reply by Vernon Barry on June 9, 2012 at 2:17am I actually covered you on this in my post in the other thread
I got the standard ICSP header pinout (for the programmer side) from here http://www.evilmadscientist.com/article.php/avrtargetboards
You should have a 6 or 10 in header off that programmer, that matches the standard. On the plug there will be a tiny triangle showing pin 1.
Another pitcure showing the key tab (ignore the chip below it)
http://diydrones.com/forum/topics/flashing-3dr-escs-or-rc-timer-esc...
Permalink Reply by Bart Mowrey on June 9, 2012 at 5:20pm Indeed you had- I had not scrolled down far enough to see the good stuff. My apologies, Its difficult to ascertain what is indeed a standard methodology ( e.g. pinout standards ) when one is a neophyte. Im used to very little standardization in my own disciplines and expect things like this to vary from one brand/retailer to another. Thanks again for the direction, the expertise and the RTFM reminder :-)
Bart
Permalink Reply by Vernon Barry on June 9, 2012 at 6:59pm Sorry, didn't mean for that to be a RFTM, just honestly trying to give you pictures and make it easier.
Honestly, I just started out in this not long ago so we are both learning.
Permalink Reply by Bart Mowrey on June 10, 2012 at 1:24am No worries- the help is invaluable and i really didnt interpret it as otherwise. Really was just acknowledging my own ability to totally miss the painfully obvious. Im still having a hard time finding the orientation, as the socket case doesnt have a small triangle or anything of note with which to orient oneself. On which side is the orientation tongue as relates to the above diagram? Right or left would you think?
Sorry to be so didactic, I just dont want to burn up gear unnecessarily or brick an ESC
TIA
Bart
Permalink Reply by Vernon Barry on June 11, 2012 at 6:32am Oh crap, just saw that the picture confuses the situation because it's "keyed" backwards to the ribbon cable. That is a pooor example and sorry I didn't catch it.
Link no longer 404s http://www.sparkfun.com/datasheets/Programmers/AVR-Pocket-Programme...
Ribbon cables are simple when done right, the red stripe indicates pin one and normally the Key faces away from the ribbon cable. Thus, that crappy picture technically has the numbering wrong pin one would be the the right corner, not the lower left corner based on the red strip and the key. Sorry for any confusion that picture caused.
Permalink Reply by Bart Mowrey on June 15, 2012 at 5:57pm After several days and innumerable poorly made cables, Im kind of at a loss. So if anyone could answer a few questions regarding this process, Id be grateful for input. Ive used the kkmulticopter flasher as well as just using avrdude alone. (All its really doing is creating a command line from dropdowns and consolidating firmware). Im using the usbtiny clone from sparkfun and it appears to be largely the same as the original. USBlibs are installed
Pardon the sophomoric questions, This is my first foray onto microcontrollers-
1. after creating a cable and rewiring any number of times i managed to finally connect only to have it drop. Ive checked and rechecked continuity as well as the wiring

2. if the connection drops in process, is the esc bricked? If not, is there a way to screw up enough that it is bricked or can new firmware always be put on even if corrupted?
3. How does the flash process take? 5 seconds...30? 18 minutes? I realize that the tgy.hex is only 8000k or so but i never recieve confirmation of conclusion when i do manage to connect.
4. The two i did manage to connect to and flash appear to have taken the flash successfully, but as a result of this no longer acknowledge the USBtuiny connect handshake. Is that normal? They do sound as if they are nicely synced and far quieter. IM assuming this has to do with the engine speed being closer.
5. Why for the love of all thats holy, does this seem to only occasionally connect? Power always seems to be there, MISO and MOSI are correct SCK, RESET VCC and everything else is correct, has connectivity and everything yet only once out of 50 tries actually connects. Maybe a better question would be which of these cables facilitates the handshake? Im guessing theres a cold joint there somewhere. ( Yes- i suck at soldering) or is there some huge thing that im missing other than missing my calling of ditch diggery or janitorial services?
Thanks in advance
Bart
Season Two of the Trust Time Trial (T3) Contest has now begun. The fourth round is an accuracy round for multicopters, which requires contestants to fly a cube. The deadline is April 14th.106 members
1279 members
671 members
183 members
182 members
© 2013 Created by Chris Anderson.
Powered by
