coreyperez
Member
My initial plan was/is for a pelican rifle case. Nothing else I have would fit it.
Corey
Corey
Tarot 680pro & Naza M v2: will it work?
- Thread starter jomk63
- Start date
coreyperez
Member
Simon,
I just watched the video again, at 2:30(ish) I see where you were descending (rapid, probably due to seeing the oscillations and wanting to get it down quickly). I don't have any video of mine doing this act, but I have encountered this same act nearly every time I've made a rapid descent. Once again, I believe it comes down to the settings. I'm not OVERLY willing to try mine in a rapid descent (Tarot), but I"ll try and record my Disco doing it tomorrow (only because I've done zero-throttle descents and I know I've got it set at a comfortable level where it will maintain self-level).
In the instance you are referring to (2:30) I really believe that the settings are too high, granted, not exactly knowing how the settings impact the results, it could be they are too low (you mentioned 100's previously and are not much higher). I look at it like this, you are giving the NAZA it's "Left and Right limits". Telling it how far it can go in either direction. If they are too wide (figuratively) then it may respond too slowly (the aircraft will pitch/roll too much to allow the motors/props to recover it), where as if it were set too narrow it would be like a professional ping-pong competition, the FC is slamming the motors full throttle (left/right or up/back) to attempt to recover/keep the craft level, these oscillations become too great/fast that eventually when one side spools up and creates thrust, the FC is already behind the power curve (one reason why people are flocking to the NAZE32 FC) and the Naza cannot now correct because it was too much/too late hence the increasing throttle/wobble you experienced.
Personally if I experienced what you did, I'd have restored the power to 50% (Appx hover power), this SHOULD have narrowed the band that the NAZA was trying to correct within. This is a theory anyway, I'd believe that by bringing ALL the motors to 50, the power swing and application would be much smaller (vs the 25% or so you probably had to "land" the craft, this would require the significant increase in power (from the flight controller) to attempt the recovery). The other motion I probably would have done is given it some directional flight (IN THE DIRECTION OF PITCH/ROLL!!! - more later). Two reasons for this: #1 Vortex Ringstate (Settling with power). Its a situation where a helicopter "Settles" into its own downwash. I'm a helicopter pilot, and this stuff is for real! Our full size aircraft with over 20,000 shaft horse power will fall out of the sky faster than you would imagine.
The second reason: That downwash you are descending into doesn't provide "clean" lift, so now, even what the FC is TRYING to do (with theoretical numbers DJI programmed into it) are only a "ball park figure" because they will figure a descent with even thrust on each motor SHOULD develop level descent, but if that air is disrupted, that one motor/side will not develop the thrust it "should" develop, so the FC will increase the thrust on that motor to correct, now you are getting that rocking motion... There are a lot of other aerodynamic principles that are involved (Induced thrust, drag, AOA, Relative wind, and even crazy stuff like stall regions of the rotor (reverse flow, negative stall, negative lift, positive lift, positive stall. ALL of this changes hundreds of times per second, every second. There is no way DJI (or any flight controller would "KNOW" what will happen in every instance (different wind directions, modes of flight etc.) They just have to figure it out on the fly.
Now quickly, why would I have given it directional flight in the direction of the motion?? Specifically because let say your craft was going left/right (opposite yours in the video ie: fore/aft). If it was rolling left and you gave it a hefty 30-45% left direction with moderate application of 50% throttle, you have completely eliminated the need for the FC to "figure out" what to do, now it's thinking "GO LEFT"... Huh... I AM going left! COOL! With that the FC is no longer "panicked" and not going to induce too much power. With that in mind, slowly bring it out of the left motion and recenter the sticks, slow your craft down and make a SLOW descent. I'd also recommend keeping some motion in until you get closer to the ground to reduce the possibility of damage, keep a slight motion going (brisk walk) until you get a couple feet from the ground.
OK, the point of that...
I'll take my 690 out this weekend (weather dependent) and give it a couple adjustments and flights, I've already given my baseline numbers and had a completely successful (3 minute) flight. Since I only have a couple 5000mah 4S batteries, I'm limited in what I can fly. But, I'll record my current numbers, including a couple rapid descents and I'll try Info@key-photo's settings. Especially since he has had success with his.
I'll report back in about 18hrs (its nearly 10pm Korean time on Friday) 1250Z Friday
Corey
I just watched the video again, at 2:30(ish) I see where you were descending (rapid, probably due to seeing the oscillations and wanting to get it down quickly). I don't have any video of mine doing this act, but I have encountered this same act nearly every time I've made a rapid descent. Once again, I believe it comes down to the settings. I'm not OVERLY willing to try mine in a rapid descent (Tarot), but I"ll try and record my Disco doing it tomorrow (only because I've done zero-throttle descents and I know I've got it set at a comfortable level where it will maintain self-level).
In the instance you are referring to (2:30) I really believe that the settings are too high, granted, not exactly knowing how the settings impact the results, it could be they are too low (you mentioned 100's previously and are not much higher). I look at it like this, you are giving the NAZA it's "Left and Right limits". Telling it how far it can go in either direction. If they are too wide (figuratively) then it may respond too slowly (the aircraft will pitch/roll too much to allow the motors/props to recover it), where as if it were set too narrow it would be like a professional ping-pong competition, the FC is slamming the motors full throttle (left/right or up/back) to attempt to recover/keep the craft level, these oscillations become too great/fast that eventually when one side spools up and creates thrust, the FC is already behind the power curve (one reason why people are flocking to the NAZE32 FC) and the Naza cannot now correct because it was too much/too late hence the increasing throttle/wobble you experienced.
Personally if I experienced what you did, I'd have restored the power to 50% (Appx hover power), this SHOULD have narrowed the band that the NAZA was trying to correct within. This is a theory anyway, I'd believe that by bringing ALL the motors to 50, the power swing and application would be much smaller (vs the 25% or so you probably had to "land" the craft, this would require the significant increase in power (from the flight controller) to attempt the recovery). The other motion I probably would have done is given it some directional flight (IN THE DIRECTION OF PITCH/ROLL!!! - more later). Two reasons for this: #1 Vortex Ringstate (Settling with power). Its a situation where a helicopter "Settles" into its own downwash. I'm a helicopter pilot, and this stuff is for real! Our full size aircraft with over 20,000 shaft horse power will fall out of the sky faster than you would imagine.
The second reason: That downwash you are descending into doesn't provide "clean" lift, so now, even what the FC is TRYING to do (with theoretical numbers DJI programmed into it) are only a "ball park figure" because they will figure a descent with even thrust on each motor SHOULD develop level descent, but if that air is disrupted, that one motor/side will not develop the thrust it "should" develop, so the FC will increase the thrust on that motor to correct, now you are getting that rocking motion... There are a lot of other aerodynamic principles that are involved (Induced thrust, drag, AOA, Relative wind, and even crazy stuff like stall regions of the rotor (reverse flow, negative stall, negative lift, positive lift, positive stall. ALL of this changes hundreds of times per second, every second. There is no way DJI (or any flight controller would "KNOW" what will happen in every instance (different wind directions, modes of flight etc.) They just have to figure it out on the fly.
Now quickly, why would I have given it directional flight in the direction of the motion?? Specifically because let say your craft was going left/right (opposite yours in the video ie: fore/aft). If it was rolling left and you gave it a hefty 30-45% left direction with moderate application of 50% throttle, you have completely eliminated the need for the FC to "figure out" what to do, now it's thinking "GO LEFT"... Huh... I AM going left! COOL! With that the FC is no longer "panicked" and not going to induce too much power. With that in mind, slowly bring it out of the left motion and recenter the sticks, slow your craft down and make a SLOW descent. I'd also recommend keeping some motion in until you get closer to the ground to reduce the possibility of damage, keep a slight motion going (brisk walk) until you get a couple feet from the ground.
OK, the point of that...
I'll take my 690 out this weekend (weather dependent) and give it a couple adjustments and flights, I've already given my baseline numbers and had a completely successful (3 minute) flight. Since I only have a couple 5000mah 4S batteries, I'm limited in what I can fly. But, I'll record my current numbers, including a couple rapid descents and I'll try Info@key-photo's settings. Especially since he has had success with his.
I'll report back in about 18hrs (its nearly 10pm Korean time on Friday) 1250Z Friday
Corey
Last edited by a moderator:
Info@key-photos.co.uk
KennElliott
Following my flights this afternoon my gains are settled at 160 160 160 175 150 150
I've had them as high as 180 180 180 170 160 160 and the machine flew well on all of the settings - a little over reactive with those atti settings but it goes to show that the gains aren't mega critical at least in my case carrying 6200 lipo and arexcraft gimbal with go pro. And as I understand it the gains have control over the aircraft in it's reactions to it's surroundings whilst atti gains govern inputs from the Tx
As to the auto take off I may remove my curse from it. Maybe. By deliberately adjusting the battery position I could cause the tarot to swoop left, right, forwards and backwards on take off but not all at the same time. I will be paying closer attention to trimming as part of my pre-take off checks.
I'm ready to fly now, may have left it a little late for autumnal colours but if a bright day happens along I'll be out - f/cast for the weekend here in the northeast is not so good. But this does mean I need a carrying bag - not sure about the pelican case. I asked Zhang at Tarot if he stocked them (was sure I'd seen some somewhere but can't find them on the tarot site) he replied "Hi yes I am have" but that was all, no catalogue number. I shoot stills and video with the gopro and a canonG15. The attached is gopro from 1200feet (all permissions obtained inc Ncle air traffic)
I will continue to follow this thread with interest.
I've had them as high as 180 180 180 170 160 160 and the machine flew well on all of the settings - a little over reactive with those atti settings but it goes to show that the gains aren't mega critical at least in my case carrying 6200 lipo and arexcraft gimbal with go pro. And as I understand it the gains have control over the aircraft in it's reactions to it's surroundings whilst atti gains govern inputs from the Tx
As to the auto take off I may remove my curse from it. Maybe. By deliberately adjusting the battery position I could cause the tarot to swoop left, right, forwards and backwards on take off but not all at the same time. I will be paying closer attention to trimming as part of my pre-take off checks.
I'm ready to fly now, may have left it a little late for autumnal colours but if a bright day happens along I'll be out - f/cast for the weekend here in the northeast is not so good. But this does mean I need a carrying bag - not sure about the pelican case. I asked Zhang at Tarot if he stocked them (was sure I'd seen some somewhere but can't find them on the tarot site) he replied "Hi yes I am have" but that was all, no catalogue number. I shoot stills and video with the gopro and a canonG15. The attached is gopro from 1200feet (all permissions obtained inc Ncle air traffic)
I will continue to follow this thread with interest.
Attachments
Thanks Corey, I recognise much of what you say and have thought about trying to fly out of trouble but my immediate reaction which is instinctive is to try and calm the aircraft and stabilise it back into a hover. I have ended up by trying to descend even from that low height because I recognise that a crash is imminent so want to reduce the amount of potential damage. At the moment that's not difficult because the ground's so soft. Incidentally when the aircraft did the initial mad over-correction the throttle was somewhere between 50 and 60%.
I hear what you say about descending through the props own downwash which is why I try and descend in forward flight wherever possible so that I get a smooth flight; either that or just descend at a very slow rate. My other two a/c handle this situation really well, sure they don't like it but certainly don't flip like the 680 keeps on doing. The other 2 that I refer to are a dji F450 with Naza-M V1 with V2 upgrade (added PMU and V2 Firmware flashed), and a Cinestar8 with Wookong-M both really stable and rock steady in flight. You can really throw them around without any loss of stability. I'd like to add that I don't fly like that when filming but do when testing the airframe post maintenance. Very grateful fro your intuitive input and sharing your professional experience. Weather is awful at the moment so no chance of flying for a bit. Best.
I hear what you say about descending through the props own downwash which is why I try and descend in forward flight wherever possible so that I get a smooth flight; either that or just descend at a very slow rate. My other two a/c handle this situation really well, sure they don't like it but certainly don't flip like the 680 keeps on doing. The other 2 that I refer to are a dji F450 with Naza-M V1 with V2 upgrade (added PMU and V2 Firmware flashed), and a Cinestar8 with Wookong-M both really stable and rock steady in flight. You can really throw them around without any loss of stability. I'd like to add that I don't fly like that when filming but do when testing the airframe post maintenance. Very grateful fro your intuitive input and sharing your professional experience. Weather is awful at the moment so no chance of flying for a bit. Best.
coreyperez
Member
So, this morning I went out and conducted a good bit of testing. I ran through a couple battery packs and flew all the following settings:
I'll let you watch the video, but I'll say that Info's 180 settings were nearly deadly (as far as the HEX was concerned). REALLY on edge. I by no way would ever recommend those settings on the 690.
I have not decided at this point where I"m going to end up, but I have not changed it from the last settings. The stock DJI settings seem a little "on edge" and lethargic as far as available "recovery power".
Battery 1&2: 3s Turnigy NanoTech 5000 45-90c (Purchased in December 13 for my F550, use in my Apprentice for nearly all its flights, never really pull power down below 35%)
Battery 3: 4s Zippy Compact 4000 25c. I had no choice but to use this battery, my other two were exhausted/weak. But, upon getting home that afternoon I found the batteries to have over 50%, I think the size of the motors would sag the batteries thus causing numerous "lack of power" situations. I don't anticipate using anything but a 4s or larger on this Hex.
Corey
**Battery 1**
Flight 1 (Stock DJI - out of the box) settings
125/125/120/130
170/170
Flight 2 (Simon's settings)
170/170/150/140
170/170
**Battery 2**
Flight 3 (John's)
140/140/120/150
140/140
Flight 4 (Info's)
180/180/180/170
160/160
Flight 5 (Info's)
125/125/100/125
125/125
**Battery 3**
Flight 6 (Youtube review/video)
175/175/125/150
110/110
Last edited by a moderator:
Motopreserve
Drone Enthusiast
Corey,
great at that you got some testing in. I think your gut feeling about 4S is spot on. I think these rigs need at least this much juice to maintain. I almost wish I had set up for 5S (cheaper than 6S these days).
Keep us updated.
great at that you got some testing in. I think your gut feeling about 4S is spot on. I think these rigs need at least this much juice to maintain. I almost wish I had set up for 5S (cheaper than 6S these days).
Keep us updated.
Info@key-photos.co.uk
KennElliott
Hi, I did qualify my settings by mentioning the load the 680pro was carrying - the gimbal, camera and 6200 4S battery. For my set up the 180 settings worked and as I said it was a little sharp and I settled for 160's.
when you showed the imbalance in your cog I was surprised that the machine went straight up I expected at least a swoop forwards. Envious.
when you showed the imbalance in your cog I was surprised that the machine went straight up I expected at least a swoop forwards. Envious.
coreyperez
Member
One thing I've noticed (about me) is I always have mine a little nose heavy, all my multirotor aircraft I've built are balanced that way. I've found it always works (for me) and probably comes from my learning from full size aircraft. Even our UH60 Blackhawk has a forward tilt of the rotor system designed in due to the advantages (aerodynamic) and requirements.
Corey
Corey
Hi Guys,
Been off air working on other things. Just thought I'd get back to you all and let you know I managed to sort out the problem with my T680. I lowered all the gains and put a 4S battery on it and it works a treat, no more flips etc in fact I've managed to fly it with my Attitude Gains as low as 40% though 80% has a better feel to it. The other gains are 125%, 125%, 120% and 130% respectively. The alarm that can be heard in the high altitude flip was the low voltage battery alarm that had been fitted to the battery. I can only assume that as the gains were set quite high it was correcting quite quickly thus draining the power out of the relatively meagre supply from a 3S 2200mah battery which meant that it did not have sufficient power to stabilise the correction made. Quite by accident I recreated a similar situation with a 4S in that I did not check the voltage when taking off after a long break in between rain showers, as with many air accidents large or small very often it is human factors that contribute in this case pilot error by not being strict with his preflight checks. An important lesson and one that illustrates (to me at least) why training flights are so important. I reconnected to a spent battery from a previous sortie and she went into a similar death flip immediately after takeoff. Because of the shortage of spare props I have even been running on props that have the odd nick in them from the previous exploits and she has been fine. I will post some footage later on today once I have uploaded to YouTube. Many thanks to all contributors. Fly safe. Best wishes. Simon
Been off air working on other things. Just thought I'd get back to you all and let you know I managed to sort out the problem with my T680. I lowered all the gains and put a 4S battery on it and it works a treat, no more flips etc in fact I've managed to fly it with my Attitude Gains as low as 40% though 80% has a better feel to it. The other gains are 125%, 125%, 120% and 130% respectively. The alarm that can be heard in the high altitude flip was the low voltage battery alarm that had been fitted to the battery. I can only assume that as the gains were set quite high it was correcting quite quickly thus draining the power out of the relatively meagre supply from a 3S 2200mah battery which meant that it did not have sufficient power to stabilise the correction made. Quite by accident I recreated a similar situation with a 4S in that I did not check the voltage when taking off after a long break in between rain showers, as with many air accidents large or small very often it is human factors that contribute in this case pilot error by not being strict with his preflight checks. An important lesson and one that illustrates (to me at least) why training flights are so important. I reconnected to a spent battery from a previous sortie and she went into a similar death flip immediately after takeoff. Because of the shortage of spare props I have even been running on props that have the odd nick in them from the previous exploits and she has been fine. I will post some footage later on today once I have uploaded to YouTube. Many thanks to all contributors. Fly safe. Best wishes. Simon
Motopreserve
Drone Enthusiast
Glad to hear you got it all sorted out. Definitely a good lesson to make sure that a pre-flight check determines air-worthiness before lift-off.
My old boss used to say to me me after I'd made a mistake: "don't worry about it...... It was just human error.....and YOU'RE the human!"
My old boss used to say to me me after I'd made a mistake: "don't worry about it...... It was just human error.....and YOU'RE the human!"
coreyperez
Member
Excellent!
Good to hear and congrats on the indepth troubleshooting you conducted! Happy flights!
Corey
Good to hear and congrats on the indepth troubleshooting you conducted! Happy flights!
Corey
I have a Tarot 690S with NAZA M V2. At first it maneuvered well with pitch, roll and thrust. Held position in GPS mode nicely.
Now, after several test flights to become better acquainted with the system, it is now giving me issues. Has flipped a few times, broken landing gear, tried to RTC in GPS mode and anything else it can do to not respond to controls properly. The Tarot frame is tough to handle this abuse.
Last flight was okay for first 5 minutes or so. While in ATTI mode, I had the bird hovering in place and then applied yaw. It overcompensated and become uncontrollable and unresponsive, then crashed. The battery tested good
All flight attempts after this were not good with battery charged, compass and IMU calibrated and motors/props configured correctly. Decided to pack up and research possible issues.
My first thought is that the NAZA works good until it's internal circuitry gets warm from use and then becomes uncontrollable. I have ordered a CC3D REVO and will test the bird, motors and ESC with that FC.
If all is good with REVO, then I will replace the NAZA with a new one as it is still relatively new and under warranty.
Until then, I am still exploring other reasons why the system is not trustworthy.
AirShooter
FlyCam UAV
I just finished a Tarot 690 I using a Naza V2 and the DJI E600 power system, I must say I'm thrilled with it's performance so far.
I've got the 900mhz down link PC ground station (works flawless) Immersion 600mw tx IOSD mini (fpv) tarot 2 axis GoPro 3/4 gimbal (pretty damn good for a $100)
two ubecs one for a solid 12v 3amp source for orientation LED's gimbal and whatever I want just at 12v as the ship is powered by a single 10000mah 6s
the other ubec is a 5 volt for gopro charging. getting a solid 25 minute flight times with everything going, 900mhz DL etc. It's fast as hell, climbs like mad and will totally just kick a Phantoms ass all day long AUW is 7.4 pounds. We have a P2 here at the shop even though it's not fair we're going to do a head to head test on video very soon.
I've got the 900mhz down link PC ground station (works flawless) Immersion 600mw tx IOSD mini (fpv) tarot 2 axis GoPro 3/4 gimbal (pretty damn good for a $100)
two ubecs one for a solid 12v 3amp source for orientation LED's gimbal and whatever I want just at 12v as the ship is powered by a single 10000mah 6s
the other ubec is a 5 volt for gopro charging. getting a solid 25 minute flight times with everything going, 900mhz DL etc. It's fast as hell, climbs like mad and will totally just kick a Phantoms ass all day long AUW is 7.4 pounds. We have a P2 here at the shop even though it's not fair we're going to do a head to head test on video very soon.
Hi, I have a similar setup to you and love the results! I was wondering if you care to share your gains? I have mine set at default at the moment and am just about to start to tweak them. Any info you could provide would be greatly appreciated.
Thanks,
Jon
AirShooter
FlyCam UAV
Hi John,
My gains are Pretty close to stock as it flew pretty good with them that way. I just rasied up things slightly and have not seen much difference. I'm pretty happy with the way it performs now.
BASIC:
PITCH=130%
ROLL=130%
YAW=105%
VERTICAL=105
ATTITUDE:
PITCH=125%
ROLL=125%
Happy New Year,
Best,
My gains are Pretty close to stock as it flew pretty good with them that way. I just rasied up things slightly and have not seen much difference. I'm pretty happy with the way it performs now.
BASIC:
PITCH=130%
ROLL=130%
YAW=105%
VERTICAL=105
ATTITUDE:
PITCH=125%
ROLL=125%
Happy New Year,
Best,
AirShooter
FlyCam UAV
It does get a bit windy around here, are you saying raise all the gains more? to what numbers would you recommend to start, it seems to fly pretty good as is.
Best,
Best,
johnthorn
Member
Hi you only need to increase the gains if it starts drifting off on the wind if you go too high it will start to rock back and forth if you get to this stage reduce by about 10 and it will settle down. As you say it is flying well you hopefully won't have to make any adjustments. For AP and video you want nice soft controls so the lower gains you can run the better,you could use my gains but remember I have a lot different setup to you it weighs 4kg so all settings can vary. Hope this helps
John
John
Hi Jeff,
Thanks for the response. I tend to agree, my hex was pretty much rock solid with the defaults. I did a similar increase with minimal difference, except for the YAW which is at 115% and is much more responsive.
Thanks again and have a awesome 2015!
Cheers,
Jon