P1 From Start - To Start!

[ScouseMouse]

Hi everyone,

I thought I would share my experiences with you as although I am new to the Phantom I have had lots of experience with 250 size FPV quads and decided to dip my toe in the water with a second-hand P1 back in December.
Within an hour I made sure the P1 armed and at least wanted to take off by setting it up indoors but not flying it - all was good! The weather here in the UK hasn't been good and with the short daylight hours and work commitments I knew I wouldn't be able to truly test it for a good few weeks (months actually!), so I invested in a used 3-axis gimbal, an upgraded potentiometer and knob for the transmitter along with self-tightening 9" props and taller landing legs.
Physically fitting these wasn't an issue, getting the gimbal to work correctly was a real pain but it now works fine and the independently made control knob/pot is a dream, seemingly much better and harder wearing the the original (now unavailable) lever for the back of the transmitter. Currently I have an SJ4000 camera in the gimbal which seems to be pretty good for the money, the next step is to fit FPV to the P1 via the live video-out feature on the camera and a 25Mw Tx mounted nicely in the shell somewhere.
Interestingly, when I connected the P1 to NAZAM the other night to check settings, it gave me an instant alert saying that my transmitter settings were out of range - panic! When I checked the throttle/aeleron/rudder commands, the rudder wasn't being recognized at all so I had to open-up the transmitter and check for faulty connections etc. Not sure exactly what was wrong but when I re-assembled it and checked again it worked fine, the only problem was the drone wouldn't arm..! An advanced IMU calibration sorted the issue but I will have to keep an eye on the rudder control in case it's a fault that comes back mid-flight.
So - all this before the P1 has even truly been flown! From starting with a second-hand drone and buying the bits off EBay (total cost so far less than £230 including the P1) to being where I am now - about to start flying!
Seems to me to be a great way to start with DJI products and I have gleaned so much info from this site I just wanted to let people know my progress so far and to say thanks to all who have posted. Happy flying!

 

[Pilot_FML]

Some very curious things going on here....

If just taking things apart & then just reassembling them cures a problem, I'd be very wary about flying. At the very least, I'd try to duplicate the problem(s) a few more times to see if they re-occur.

I also don't see why an IMU calibration has anything to do with curing an arming problem!

Adding a Vtx also has the potential of interfering with the proper GPS operation and probably only should be done after you're sure there are no other problems.

I wouldn't even mount the camera & gimbal until after several shake-down flights verify all's well.

Bottom line: Take small steps and resist trying to do everything at once....just my humble opinion.

Best of luck...

 

[ScouseMouse]

Your comments are much appreciated and I had already decided the first flight is going to be REALLY close to me at low altitude so at least if the rudder problem recurs I will be able to land straight away.
Good point about adding Vtx so I will follow your advice until I am confident all is well. Thanks

 

[ScouseMouse]

Pilot_FML - So far so good, no rudder problems experienced again but it is something I am aware of and will kepp a close eye on.
With reference to your comment about a Vtx possibly interfering with GPS, is that a common issue? Lots of people have FPV added to P1's so I am curious and don't want to risk anything!

 

[Pilot_FML]

Pilot_FML - So far so good, no rudder problems experienced again but it is something I am aware of and will kepp a close eye on.
With reference to your comment about a Vtx possibly interfering with GPS, is that a common issue? Lots of people have FPV added to P1's so I am curious and don't want to risk anything!

Not really a widespread or common problem but just something to be aware of.

A poorly designed Vtx could emit signals that interfere with GPS signals. Not getting the proper number of satellites OR a good home lock before takeoff may indicate a problem in this area.

A Vtx placed between the P1's GPS antenna and the path of GPS satellite signals could also physically block GPS signals. Confirm where your P1 GPS antenna is and mount the Vtx accordingly.

Nothing guarantees that you'll have good GPS satellite locks your entire flight but by practicing in the ATTI mode, you can get around unexpected GPS problem and bring the P1 home without GPS assistance.

 

[ScouseMouse]

Thanks for the info! If I DO go ahead and fit a Vtx it'll only be a 25Mw which is our legal limit here in the UK and I think I'll mount it away from the upper part of the shell and GPS module. Your comments are extremely helpful and I will be doing a lot more close proximity flights to check existing GPS lock and also fly in ATTI to make doubly sure the rudder issue has been cured. Thanks

 

[Pilot_FML]

Thanks for the info! If I DO go ahead and fit a Vtx it'll only be a 25Mw which is our legal limit here in the UK and I think I'll mount it away from the upper part of the shell and GPS module. Your comments are extremely helpful and I will be doing a lot more close proximity flights to check existing GPS lock and also fly in ATTI to make doubly sure the rudder issue has been cured. Thanks

You're more than welcome.... enjoy the P1.... it's a reliable and "uncomplicated" bird to fly. Never regretted my choice to buy a couple of these.

 

[Some-E]

Cheers from Finland as well! I got my very first drone, the P1 just last autumn I'm a beginner with everything that flies or generally RC world. I'm beginning to value the P1 robustness and that it just flies. If it fails, it's a pilot error, one way or the other. I've had some periodic problems with the GPS reception and drifting while flying. I'm receiving a new GPS+compass puck (uBlox Neo 8 series) soon, and some have told it has improved the flight. I hope it could also fix the drifting for good. Enough about my babble!

Oh, your rudder issue: the rudder potentiometer could be worn, and thus not always having a connection between the wiper and the resistive element -> out of range. It should not go bad completely, but if you encounter strange behaviour, just wiggle it to get better contact (that's what we do when an audio gear pots crackle). I haven't heard about this before with RC, but that's my general analog electronics experience talking. It would be easy to verify if you could connect headphones to the potentiometer, spot / disprove the crackling.

EDIT: monitor the rudder movement in the Naza-M Assistant display while you move it slowly. If it jumps around even a bit, it could be a pot issue.

 

[ScouseMouse]

How good to hear from Finland..! I have always admired your country and in particular your driving!
From an early age I was involved in rallying and of course the "Flying Finns" left a lasting impression on me back in the late 1960's and 70's. The Rally Of The Thousand Lakes was always a favourite and now that WRC events are covered by DJI drones it makes Rally Finland even more spectacular to watch on TV.
I am glad to say that since I last posted on here I haven't had any issues at all. Although I keep an eye on the rudder controls the aircraft is behaving perfectly. The only issue I have now is getting rid of unwanted "Jello" on the footage from the SJ4000 but I think I can cure that.
Enough of my droning-on (forgive the pun) about rallying (I could go on for hours!) and like me I hope you continue to enjoy the P1. I love it.

 

[Some-E]

You didn't name your gimbal, but the weight of the SJ4000 seems to be an issue for some gimbals: it's too lightweight! I also have SJ4000, but with 2D gimbal (Walkera G-2D plastic). I was suggested to add weight, and I did (66g -> 84g for starters) and it made a difference. I could tighten the dampers more without gaining so much jello. I added the biggest weight on the back of the camera frame, not the camera itself, as there is not much room vs. finding an acceptable balance, maintaining removability and button operations etc.

2D is good for understanding the vibrations of the whole system: there is nothing to compensate it, everything shows as horizontal movement. Just what I need for good jello! It is been balancing between wobbling (lower frequency, large vibration) and jello (small amount on higher frequencies). I've done that by adding rubber bands (cut from mtb bike inner tube) through the holes in the two mounting plates and tightening. That compresses the actual dampers. The optimum tweak would restrict only the horizontal movement / roll. I do have something clever in my mind, but cannot discuss it openly for now. All in all, getting an old drone has given me much more than I expected! I still don't get the video stability I expected, but I've learned a lot, and still keep learning more.

 

[ScouseMouse]

We share a similar outlook obviously! My gimbal is a metal DYS (Chinese) 3-axis model so it does have some weight to it. I have experimented with placing 15mm lengths of quite stiff drinking straws pushed down through the holes in the rubber mounts to stiffen it up and although this seems to have helped, I am still getting jello. I have ordered an ND filter for the SJ4000 so I'll see if that helps calm the effect generated by the rolling shutter. I have also adjusted the power output to the pitch motor as it was vibrating just a little.
I have seen so many videos about different ways to cure jello, from fixing everything as rigidly as possible to adding various types of damping materials. I really do think it's a case of what works for one machine / flying style may not work across the board. Now that I have some "bad" footage I will post some (hopefully) "improved" footage on here when I see some benefit!

 

[Some-E]

I will not try ND filter, as there is not much sense in restricting the amount of light. SJ4000 doesn't have that sensitive camera/chip. And while making the exposure longer could reduce jello, it's the same as flying in dim light. You can test that. I've done a lot of those street light tests, just because the daylight time is so short during the winter, and yes, I cannot really see jello, but I also don't get sharp image because of the vibration. You get blur instead of jello. The vibration is what needs to be reduced, otherwise your're just denying the problem. I have to admit I've seen ND filter videos, and I've seen pretty fantastic results - but I haven't perhaps watched in full size. I often watch videos on my phones, as my old computer isn't really capable of YouTube 1080p... sigh.

I can now get great image when flying very slowly, in good light that is. I only have the final horizontal fix missing (no yaw motor). I also get smoother image when flying backwards compared to forward flight. That suggests wind induced problems, so I went to thinking that I need a transparent dome to protect the system from the wind, but after doing a crude test with transparent film as windshield, I dropped the idea, as I didn't see real improvement. As a bonus I got reflections, even from the tail lights. But what if it is the balance... (lost the thought).

If you can, mount the camera directly without a gimbal. To see how the drone contributes to the vibration. I tried that once. I drilled two holes through the baseplace of the action camera set, so I could screw the plate to the bottom and slide in the "GoPro" mount accessory. For me the image was surprisingly good! Of course the view is bad, due to constantly changing attitude, and I did it with street light time. That test told me that I may not need to spend time on balancing the motors. It could as well be the airflow from the props giving vibration?

Now it would be very good to have edited clips of the videos to share. I myself have one big question: does 3D gimbal solve all the problems? No, according to what I've seen before and reading now here. The cheapest 3D I've seen from eBay is something like USD 60.

Speaking of 3D, I got a "2D - 3D conversion kit" with motor, but without electronics. I already had a 2D controller with accelerometer, so I wanted to mix those two (three, including the Walkera!) pieces as one. It doesn't work, because the 2D controller really takes in pitch and roll data, and you cannot rewire digital accelerometer. I'd need a single channel contoller that reads which ever accelerometer channel I need. But sorry, I realize this goes off-topic!

 

[ScouseMouse]

This is partly why I bought a P1, I like to know HOW things work so that I can understand the limitations of the aircraft, flight controller, GPS - and of course me as a pilot! So I have no worries about fitting the gimbal and eventually an FPV transmitter. It would be a totally different proposition if I had spent £800+ on a P3 as in any case DJI have made it more expensive/harder to replace things like the ESC's.
I appreciate your comment about the SJ4000 low-light capability, the ND filter may be a waste of time but at least it'll iron out one other variable. My DYS gimbal cost me £30 off EBay, brand new in box but pre-owned and never fitted to anything. Instructions are appalling so had to rely on YouTube!
I have the original fixed camera mount for my P1 so one day soon I will also do a test flight with the camera in a rigid mount, this will also tell me whether the jello is partially induced by the gimbal.

 

[Some-E]

What a pity I don't have the original camera mount! P1 has apparently been used by two owners before me. There certainly can be some hidden variables lurking

Oh yes, solving out the variables. That's been the topic for quite a while now There is one big factor: we really don't know what happens. We only see the end result. We can knock the gimbal-camera assembly manually and observe the vibrations. What I really want to do is to stick four accelerometer chips below each motor, and log all those XYZ accelerations and then overlay the data on the video. I know the hull is really rigid, but still I want to see the data from each motor. The motor speeds and sync (phase) would not be bad either. Humm...

 

[ScouseMouse]

Haha now you are getting REALLY technical..! But I agree that nobody really knows how to cure it on a particular aircraft, what suits one many not work on others. I'll update you as soon as I have more results

 

[ScouseMouse]

 

[ScouseMouse]

Really bad jello. I posted this clip because it shows how the vibration instantly stops when I kill the motors after hand catch. The gimbal is still powered so it proves the fault isn't with that. Next step is to re balance 9" props n maybe try original P1 versions. Pulling my hair out!

 

[Pilot_FML]

????????
Your comment about "now" trying to balance props really surprised me.

Having balanced motors AND balanced props are the first things to check when jello rears its ugly head.

Good isolation mounts are next, followed lastly by ND filters. The filters might improve SMALL or intermittent jello but often come at the expense of slightly blurring your video.

If you haven't done things in the order I've suggested, chances are good you haven't found the root cause(s) of the jello problem...

P.S. Gimbals aren't designed to eliminate most induced vibrations. Their primary purpose is to create a "flying tripod" to stabilize aerial video.

Really bad jello. I posted this clip because it shows how the vibration instantly stops when I kill the motors after hand catch. The gimbal is still powered so it proves the fault isn't with that. Next step is to re balance 9" props n maybe try original P1 versions. Pulling my hair out!

 

[ScouseMouse]

Thanks but I did actually say "re-balance" the 9" props as I have done this before flying a few weeks ago. I installed these pretty much as soon as I got the P1 and have never flown with the standard props so I am going to give them a try as well.
I was interested to read about balancing motors? This is something I have taken for granted and never considered so it is another avenue for me to look into. Thanks for your comments.

 

[ScouseMouse]

An update. I have installed a 25Mw Vtx with direct connection to my SJ4000.
Because I am doing this on a tight budget I originally bought an Easycap video capture device which, when connected to my Android phone should have sent the feed from my Vrx straight to the phone. However, my phone won't actually power up the Easycap from an OTG cable so that idea is shelved for now. In frustration I bought a 7" monitor but promptly managed to break the LCD screen!
I guess this is all to be expected when doing a project on the cheap!