Everything I have from 3DR sucks...

[SolidSalad]

OK, I'm going to be brutally honest here. From my (limited) experience to date, products from 3DR suck. Both mine and my brothers ArduPilot randomly refuse to arm. I don't mean I've tweaked a setting some place, I mean like it has a flight, swaps batteries, and decides "nope". 30 mins later, same pack, same everything, and no worries. Then later, again, just nothing. This has occurred to both my ArduPilot and my brothers. We have been through all the logical troubleshooting steps. Everything is as it should be, and yet more often than not, my flying day is soured by these unexplained failures.

So, that was the ArduPilot voodoo. On to the Minum OSD stuff. Again, both myself and my brother have found these to be pretty crap. My brothers has some bazaar start-up issues, while mine quietly ignores some settings (no RSSI, for instance) no matter what it is told.

Between my brother and myself, we have about 30 years R/C flying experience, and no other brand of product has caused this much frustration. I would tell any newcomer to Quads to get a cheap flight controller and steer clear of 3DR products. Sorry guys, but there it is.

SolidSalad.

 

[mackar]

Always nice when a newly registered member on a forum writes posts like this ;-)

 

[SolidSalad]

Always nice when a newly registered member on a forum writes posts like this ;-)

Well, I did only register here to express my frustration with my 3DR stuff. What is the point of a public forum if not to express opinions?

In all honesty, its probably less the product produced by 3DR, and more the actual software behind it, but so far, nothing works as indicated, from compass calibration, to arming, to the MinumOSD. Even calibrating the ESCs took several attempts before the APM would enter calibration mode. Keeping in mind, my brother has the same issues, and can not get his OSD to function at all.

When it works, it is great. Stability mode was, well, stable. Acro was more responsive, and felt more intuitive. Once I got the ESCs calibrated, both modes were smooth and fun to fly FPV. With the exception of RTL set as a (to date untested) failsafe, I've not had the chance to try other modes.
Sadly, its unreliability so far has very much soured my experience with APM and the MinumOSD. If others have had similar experience, and found a solution, I'd love to hear about it.

 

[dazzab]

Sadly, its unreliability so far has very much soured my experience with APM and the MinumOSD. If others have had similar experience, and found a solution, I'd love to hear about it.

I'd have to agree that lack of a good OSD solution really lets down 3DR products. It's almost like it was an after thought. I've found that users of APM and Pixhawk are much more focussed on using a ground station so OSD isn't a priority in that case. While a ground station gives you great functionality it's not much help for FPV or when you are using a monitor for taking photos. Having said that, I highly recommend using a ground station with live telemetry as it gives you logging on your side in case your copter goes missing or you need to know where it's last location was. Not to mention much more functionality that I won't go in to now. But for me OSD is a requirement, I just couldn't imagine flying without a telemetry display on my Tx.

Have you tried the official help forums at 3DR or even the blog and DIYDrones? I'd be amazed if someone there couldn't sort out your issues.

 

[Old Man]

Although SolidSalad starts out with a pretty negative first post he does have a point relative to the performance to the earlier 3dR GPS equipment. I have one like that too and it has made for some pretty frustrating days. OTH, should he do just a tiny bit more research he would find that 3dR has left that GPS unit behind and moved to the uBlox GPS, which is phenominal. I have yet to experience an acquisition period of more than 60 seconds, with 30 seconds or less being the norm unless I was in a "small window" box canyon. For now, get that old GPS up off the deck if that's where it is and shield the bottom of it from EMI coming off the rest of the boards. Do that only until changing to the uBlox unit.

As for the OSD, I agree, it could be a lot better with considerable improvements in the flight instruments. The MinimOSD has the bank attitudes working backwards. Perhaps with Branson's investment and Guinn's marketing knowledge they'll start focusing a little more on being user friendly. So Solid, the critism should be taken for what it might be, a plea for a better product. Some of that has already been done but there's still room for improvemnt in other areas.

 

[jdennings]

I have to agree that the OSD sucks. I can still remember hours spent on soldering and wiring getting the thing to work, plus the whole flashing enchilada. That said, once it did work it never stopped working. Which doesn't mean of course that your experience is unique. That osd does feel like an after thought. In fact, I think it is. Having a groundstation nearby with Mission Planner running really helps in these cases.

As far as arming inconsistent behavior, never had this except when pre-arming checks failed, you may want to look into any message from Mission Planner after failed arm. Weird compass behavior or GPS lack of fix? Gyros or accels failed to initialize because of possible initial calibration issues? Or maybe it all has to do with ESC calibration: If not successful or something's not right, the FC won't arm. With respect to ESC calibration, sometimes (and I still can't for the life of me figure out in which cases) all in one calibration just does not work for some ESCs. This is pointed out in the wiki/manual, where in that case one by one calibration is required. Yeah, not great on that front ...

 

[Old Man]

The earlier GPS units are quite sensitive to EMI radiated from power lines. I had quite a few pre-arm failures until I started moving farther away from power lines before trying to initialize everything.

One thing 3dR does I've seen before is in their documentation. It's an engineering company and they let their engineers write the Wiki documentation. A UAV company had some of the worst tech manuals and written instructions that millions of $$ could buy that was not corrected until they took that task away from the engineers and hired people that were trained in communications simplification. Engineers like to write for other engineers and do not understand that the rest of the world is not completely populated with nothing but engineers. Instructions have to be written in a simplified language were task requirements are not made excessively difficult though complicated equations and specialized language. If 3dR wanted to jump to the head of the line cleaning up the Wiki and making it less challenging to navigate would be the first place to start.

 

[R_Lefebvre]

SolidSalad, Arducopter provides you all the information you need to figure out why it is not arming. So complaining that it won't arm, without providing that information, makes this more of a rant than a plea for help. Where are your logs? What is Mission Planner telling you? Nobody can help you without info.

Old Man suggests you are using an old MediaTek GPS, but I don't see any mention of that. Is this true? If it is, then just throw it out, it's almost worthless.

 

[Bartman]

Good post, bad post.

MultiRotorForums.com exists to keep conversations flowing without regard for what this shop or that manufacturer wants us to post. So, with that in mind, the post is ok and we'll get through the problems and help our community here to understand the problems better.
On the other hand, making a first post with the thread title "Everything I have from 3DR Sucks" is pretty immature and I hope future posts by the OP will be more well mannered.
We can discuss what we want, we're big people, but the title is inappropriate.
Thanks,
Bart

PS, [MENTION=17068]SolidSalad[/MENTION]; , listen to what R_Lefebvre has to say about 3DR hardware.....he's a 3DR developer, a master grade user, and will provide you with whatever help you might need to get it worked out.

 

[Dawolf]

I am new to 3dr and the whole PX4/Pixhawk platform myself. I've been using mostly Naza for the last 7 months or so and the differences are huge. The Naza stuff seems alot more plug and play, but limited to what you can do, and tweak. I will admit, that I got into the PX4 world to tweak and tune. Therefore there is a learning curve. My first experience has also been tainted because of bad equipment. I was under the impression that 3dr had "superior" products that are at the very least Quality Controlled. For the premium anyway, I would expect that, and I also expect good support should something go wrong. In my case, I wasted 2 weeks messing around with a new build trying to get ESCs to arm, only to have it be a very defective Pixhawk be the culprit. This thing was toast and how it ever made it out of the factory is beyond me.
I got the new Pixhawk and everything fired up and worked as I expected it to. I love the Arducopter platform, but I have to say, I am very tempted to try out the Fixhawk or HK etc. I don't see the downside. I already took lumps after buying the premium product. It can't be worse.!
OK so I had to vent a little here too!....Thanks Bartman!.....and Welcome fellow Noob SolidSalad, looking forward to some more positive stuff in the future....but for now, I am with you on the 3dr sucking as well.

 

[R_Lefebvre]

I can't really speak to the relative quality between the brands. But the customer support is very different. With most of the clones, if the same thing happened, you'd just throw it in the trash and buy another as they offer no support whatsoever.

 

[Old Man]

Dawolf,

There was a period where the support at 3dR had fallen quite low but that issue was recognized and appears to be on a positive track of improvement. That's what I've experienced anyway, and saw the bad with a massive change to the good after some initial issues. That 3dR responds to customers at all is a vast improvement over the most popular company, and generally anywhere from China in general. In my situation they went over the top in correcting some defective product issues with over night shipping, pre-paid return mailers, and setting up conference calls around my work schedule.

I can truly appreciate where you're coming from but have to ask, were these experiences before or after July/August of this year? That's around the time Vu Tran took over the 3dR customer service area over there and he's done some pretty good things since his arrival.

 

[DennyR]

The fact remains that all of the 3DR codes have a very long list of revisions that address a great many problems. Problems that invariably result in a crash. At the end of the day the problems seem to get worse as the complexity gets greater. Maybe someone needs to step back and fixed the basics so that it can work to the same level of reliability and accuracy as say the DJI Naza.

The Walkera series of multi rotors look quite nice but I had reservations about the 3DR code they use which it seems was well founded based on the number of reported crashes already. If Walkera have the internal talent to develop the code in house then it will work one day. They definitely have the skills to make well designed models. It has taken DJI a very long time to get where they are, they do churn out some good reliable products and that is due to the advanced testing procedures that are now in place.

Part timers have no place in this business, the learning curve costs users money and degrades the publics perception of safety. Open source is great if you want to spend your time farting about with limited horizons and get a kick out of DIY, but it sucks in the real world of commercial production. Sorry but these are the proven facts. It is a bit like a round of golf. You always imagine that the next will be much better, but without a big step in how you go about it, it will not improve much.

When you sell a few components to a DIY fanatic then it is his or her responsibility to make it work. No risk to the seller at all once someone puts a soldering iron to it. But when you sell a complex hi-volume ready to fly machine like a Phantom 2+ then you have a different set of standards to adhere to. Not just the design but the production costing and quality. The buyer is probably highly inexperienced so you have to not only make it work properly but also make it goof proof. Light years away from any open source project.

 

[R_Lefebvre]

Denny, the fact remains that you still have no idea what you're talking about as regards 3DR and Arducopter. This is just another in a rant in a long history of same, where the facts are either incorrect or just made up.

Long list of revisions: Can you show me another flight control system which does revisions at a lower rate? Arducopter's last release was 3.1.5, 5 months ago.

Problems that result in a crash: Can you show me an example of a crash due to a bug in a stable release (not to be confused with user error, as there are many).

Walkera: Same deal, can you show examples of crashes that are due to code errors, as opposed to user errors?

DJI Advanced testing measures: Is that was led to the A2 double-release debacle?

Part-Timers: Can you list how many of the developers are part time?

 

[Old Man]

FYI, I work with a military sUAS that uses a sophisticated auto pilot and flight control system. Software upgrades are common. The upgrade process for any flight control system is a mandatory requirement unless it was perfect the first time and no added features are ever desired. One only needs to look at the revisions made to iOS software to recognize that software is never stable. Not stable unless you fired the software development staff in any case.


Sent from my iPhone using Tapatalk

 

[R_Lefebvre]

One only needs to look at the revisions made to iOS software to recognize that software is never stable. Not stable unless you fired the software development staff in any case.

Exactly. It will never be "stable" unless you stop working on it. And if you stop working on it, that does not mean it's better. It just means you're ignoring problems.

 

[jdennings]

I got the new Pixhawk and everything fired up and worked as I expected it to. I love the Arducopter platform, but I have to say, I am very tempted to try out the Fixhawk or HK etc. I don't see the downside. I already took lumps after buying the premium product. It can't be worse.!
OK so I had to vent a little here too!....Thanks Bartman!.....and Welcome fellow Noob SolidSalad, looking forward to some more positive stuff in the future....but for now, I am with you on the 3dr sucking as well.

My experience with more than a half dozen boards from 3DR and others is that they are neither better, nor worse. When it comes to upsides/downsides, imho:

What you get with paying a 3DR premium is, in a nutshell:
1. You support development and the developpers
2. You get boards earlier than clones, as it takes the latter months to clone. A moot point now.
3. You get, at least in theory, better support with the APM forum. T

With respect to 3, there's some support on other forums including this one, some is very good (but definitely not from the clone vendors, it's either inexistant or sub-par), some is borderline if not plain ignorant***. It's the good old "don't believe everything you read on the net", you take risks. The 3DR guys on their own support forum are on average way more knowledgeable and reliable. That said there are issues with that forum too in terms of actual response.

I approach 1. with a mixed approach when it comes to my wallet. I want to support the dev community and 3DR, so I do it by happily buying stuff from them. But there's a limit to how far I am ready to kiss 3DR's VCs in the background and add to their cash cow registers vs the developpers. There's also a limit to how far I want to support 3DR's tendancy to label the work of many developpers in the open source community as their own, and use it for their own enrichment beyond a reasonable amount. 3DR benefits tremendously from it's sponsorship of development, so I am not willing to infinitely add to that (the temporary lock-out of clone boards in Mission Planner was a good example that shows keeping things in check is healthy imho). And so I go both ways with my purchasing decisions. This video from Sparkfun also particularly resonates: https://www.youtube.com/watch?v=xGhj_lLNtd0 ...
I know this is controversial, and many disagree with my approach. At the end of the day you've got to do what you think feels right for you ...

*** (Sorry DennyR, but with all due respect that post of yours above is a good example. Yes, there are many *beta* revs when new features are introduced, and lots of testing including some resulting in crashes, as is the case in the closed labs of all FC manufacturers. But that has nothing to do with stable releases. And these crashes typically happen testing beta features so advanced there is no other FC that can even come close to offering, e.g. in flight switching of dcm to ekf inertial navigation (fixed in beta release candidate rc12, or auto spline waypoint where velocity crosses 0, etc ...). The community of beta testers is huge and it's all done in the open so you hear about it more. That's a good thing, I'd rather have 1000 testers than a dozen. If anything that makes Ardupilot FC's and stable releases more reliable than fc's from other manufactuers, not less ...
).

 

[Old Man]

jdennings,

I truly support your position with #3. Some of the DIY forums get pretty thin in the response department, some do pretty well. It's kind of hit and miss with some of it, and I feel for those that go to the source and get ignored. If you're going to host a forum, be there for the people you wanted to support. Yes, it becomes time consuming and invasive but that should have been recognized from the beginning.

Unlike DJI, 3dR does not put beta versions on the market with the intent of marketing more product through the mandatory use of the beta version. If people think back to the beginning of the year they will remember DJI put the A2 controller out labelled as beta but failed to assure people were aware what they were getting. The advertising provided the appearance the A2 was a vetted unit with beta thrown in for good measure. Many buyers and many thousands of $$ later it still isn't up to snuff, and that fails to mention the double/triple beta back to back software releases a few months ago. 3dR makes it clear that if you are not into software development you should not upload the beta versions. If you do you became a willing test pilot. Everyone has the option of only uploading the proven releases and to date I have not had a problem with any of those. In fact some of my stuff has worked better after every one.

3dR products are designed to interact with products from other companies. It might take a little thought and some work but they will happen. Not so with the other company. All their stuff is designed to work with their other stuff and outside products can be quite difficult to tie into the base systems and components. They even make propeller choices a near proprietary decision.

The real drawback for most people where 3dR is concerned is one has to work for it a little. Might have to read a few pages of Wiki or do something different from that other major brand name where everything is easy and supposedly all plug and play, when it happens to work the way it's supposed to. Once some time is spent getting involved with the 3dR line things become easier. You still have to work for it but you have a lot more to work with. Yea, it can be hard but the rewards come from what you put into it.

 

[mackar]

The real drawback for most people where 3dR is concerned is one has to work for it a little. Might have to read a few pages of Wiki or do something different from that other major brand name where everything is easy and supposedly all plug and play, when it happens to work the way it's supposed to. Once some time is spent getting involved with the 3dR line things become easier. You still have to work for it but you have a lot more to work with. Yea, it can be hard but the rewards come from what you put into it.

This is not a drawback for me A big part of this hobby is to find info how this stuff works, build, test and hopefully be able to fly.
I've been flying with Naza-M based F450 before and I cannot say that it has been working without problems. Sure the Pixhawk isn't as easy to setup as a Naza but one of the reasons is that you have so many more functions.

 

[dazzab]

In my view there is quite a difference between revisions of a mature system that's been out there for years and revisions to develop functionality. It wasn't long ago that the APM code didn't have issues with vibrations. Until they implemented code that worked like other controllers. Then the users were totally caught out by vibration issues. It wasn't that long ago they discovered that light and heat on the baro caused issues. They wrote code to deal with that. It wasn't that long ago they had issues with possible fly aways that they determined could be from GPS glitches so they wrote code to deal with that. They really added a lot of sanity checks. Now they are going further and going with dual sensors etc. It's' all very impressive.

I think if Randy and Tridge were on this forum they would be the first to say that the APM/Pixhawk has been a very fast development project. Apparently it's been going for about 5 years with about 50 people contributing to the code. But it didn't really take off until about two years ago. By comparison DJI and others had software out there actually flying copters for commercial systems earlier. Like it or not, they were ahead and worked out some issues long before 3DR even knew what the issues were. That's just normal.

What 3DR and the open source community has done with the Pixhawk and open source development is truly revolutionary. Keep your eye on them as they are the future. Chris Anderson is a visionary and the people that work around him are just amazing. World wide there are very talented people sharing and developing amazing ideas and code.

I have expressed here that in my experience the Pixhawk is not ready for prime time. But that depends on what you consider prime time. For my specific needs other flight controllers work better for me right now. For those not comfortable with the DIY mentality or enthusiasm then there are closed source out of the box experiences like the Phantom. It's really quite amazing.

I'm moving on to some mapping and more scientific projects next year. No way I'll be using DJI for that. It will be Pixhawk/Arducopter/Arduplane as it's very advanced in that area. I also suspect that with the Canberra UAV team winning the Outback Challenge recently that perhaps a few of us may be able to tie Tridge down and make him play with multicopters more. The Randy/Tridge team is quite amazing along with all the others like Rob that are in there coding away on projects that just boggle my brain.

We live in interesting times.