I bought my APM 2.0 before it was actually available, and spent a long time waiting for it to finally arrive..mounted it on my home brew quad and went into tuning hell...I never got it to where it would fly better than the KK board I had on it previously, so I removed the APM 2.0 and went back to the KK board.
I had some time off around the holidays, so decided I could revisit this a bit. Big code changes, more advise on how to tune properly and now it flies well...still working up to a mission, but I think I'm getting close.
I notice that there is a recommendation for a "power protection kit" that is a zener diode and a fuse that should be soldered in place. I have been searching and I can't find out WHY this kit is required...I can see threads where it's discussed how to install it, that it's available, etc, but I've hit my limit on time for trying to find out WHY it's required...
Can anyone comment?
BTW, finding accurate and up to date information on the APM is my single biggest frustration with the purchase. You find things on the wiki or in the forum. The 2.8.1 tuning guide is by far the best thing I've located to date and it was a link in the wiki that I almost didn't notice.
I know how it works and I know how to install it...I just haven't been able to locate a discussion on what problem it's trying to solve.
I mean, are people blowing the board up with to much voltage? Voltage spikes taking things down? What motivated the change?
I'm trying to understand the reasons behind the addition to determine if my quad is at risk...and I've been completely unsuccessful.
I'm also not finding the power protection kit on the DIYDrones store, although the instructions do provide part numbers of the devices so I can likely source them another place.
Seems this shipped with the later boards...
The 'kit' was never listed in the store that I remember but 3DR would send the parts to you if requested.
In essence, failure analysis was showing that the input voltage was being exceeded by builders. A major problem with this hobby is that a number of people coming in do not understand how fragile and sensitive the components of the APM are to overvoltage. Often they are learning electronics while trying to learn RC also. The pico fuse and zener addition were an attempt to mitigate a overvoltage input mistake. Power reversal is also suspected in some situations. It is that simple.
With APM2.5+ and the introduction of the APM Power Module, these events should eventually fade.
I didn't receive a fuse and zener in my APM2.0 purchase. I use the 3DR PDB with all four ESCs connected and have never had a problem.
I am sorry to hear you had a poor experience with your APM2.0. After I sorted out my radio setup issues, it has been fantastic with only me to blame for any damage due to flying/learning to fly.
Not only comfortable reading one, drawing one and designing one...I'm an EE.
So, it's meant to protect against user error and not any specific event or weakness in the design (unless sensitivity to overvoltage is considered a weakness).
Seems to me an external power supply is not the best approach to resolving the issue. The low power levels that the board requires (not servos, etc) implies that a small one chip switcher would be a better solution, but I guess a clamp circuit represents an alternative.
I've got my quad flying very, very stable now...but was completely incapable of arriving at that level of stability the first time around. I've made no changes to the quad since that time, and now, rock solid and I'm happy with it's performance. I've got most of the things tuned now, just some final tweaks to Loiter and Alt Hold before I attempt a RTL and then a mission.
I have some frustration with the idea that I purchased something that was "polished" but when received was quite a bit rougher than I believed...I'm quite certain that the problem lies with my expectation rather than the product...at least from the 3DR perspective...I do lay some of the blame at their feet...and won't accept complete responsibility for it, but recognize my expectations were maybe a bit higher than they should have been...
I continue to experience frustration with the "instructions" for the product. They aren't current with the firmware revisions...they often reference things that are labeled differently (Altitude Rate P in MP referenced as THR_ALT_P in the wiki, etc), or just aren't mentioned at all. I can deal with most of that...but my single biggest frustration in all of it...there is no rational, step by step process to start with an untuned custom built copter and arrive at a tuned copter capable of utilizing all the features offered on the board in a reasonable length of time. The closest I've seen is the recent forum post (not a wiki entry!!!) by Dave C. and it only covers Acro Roll and Pitch. Not that I'm being critical of Dave C...I'm actually quite thankful and impressed with his work.
It would be nice to see something that said...start with Acro mode and get the Roll, Pitch and Yaw solid, then move to Stable mode and do the same, then move the Alt Hold, Loiter, RTL and finally a Mission. Perhaps a video or two to show what a stable flight looks like, and then what some of the common deviations from that look like. Dave C's example was just about perfect, if you ask me...I only wish it would continue down the path to completion. I recognize that Dave C isn't a part of 3DR...and this is only a hobby for him, so that makes his contribution that much better...I wish for more, but don't expect it.
I fully expected my tuning process to take a while...I'm not using an off the shelf quad they provide...however, I wouldn't have ever guessed it would take this long. I just wish that 3DR would step up a bit more on the documentation side of things.
As a company, 3DR is in that stage I liken Microsoft to have been in after they got the IBM PC (the dual floppy disc PC - mid 1980's) firmware deal. Tranforming from smallish tech company to a larger customer oriented business can be a tough transition. As the entire thing/product has been open source, I think it makes the job more difficult.
Documentation will always be behind the release until they centralize the process of development. There is an active effort on the part of 3DR to improve the situation. This has been repeated several times by Chris Anderson.
The tuning variables... I don't even want to think about the changes in them. The Code Monkeys (an affectionate term - I love and appreciate the efforts of the development team) have the hardest job in the world. As folk come and go, keeping track of all the various builds and tweaks is a Herculean job. Happy 2013 to the Development Team!
Glad to see another EE trained hobbyist. There were so many questions regarding the power input of the AMP2.x, I felt compelled to massage the schematics for the basic power to help ease some of the questions. Not everyone is comfortable with Eagle or even has access to a PC (if they are running with tablets, netbooks, etc). Thus the simplified power schematic.
3DR is looking to staff some engineers. If you leave in southern California (or want to) and are ready to change positions, give them a shout.
My 1st quad was built from their published 'B' model drawings. I essentially cloned their dimensions with variations in materials. My 2nd will have shorter arms ('C' model has shorter arms) but a larger central body. The motor-to-motor moments will be the same as the 'B' version to flatten the tuning curve.
William, thanks for the feedback, and I agree that we should work harder on a great tuning guide. I just wanted to be sure you understand that 3DR doesn't make the software or the documentation. Both are written and maintained by volunteers in this community, and distributed for free. (3DR just sells hardware). That's what an open source community is all about.
If you'd like to help with the improvement of the documentation, we'd love to have you! Please PM me with an email address and I'll put you on the dev list to bring you up to speed.
Thanks for the offer, but I'll pass. I make it a policy to not provide volunteer work to a for profit business.
I'd even settle for a guide that listed the steps to follow to arrive at a good tune. I've searched the site for one, I've searched OFF the site for one...if it's there, it's not obvious or even easy to find.
WRT the documentation...It could be vastly improved with a simple rationalization of the terms from the User interface and the detailed instruction in the Wiki. Terms utilized in the Wiki should agree with terms in MP...simple as that. I shouldn't have to speculate if Loiter P is equal to loiter_lat/loiter_lon...
There are a number of items that are available as part of the code, and in MP, that are simply not documented, or not documented well in the wiki.
In the development world where I live, this simply isn't allowed...documentation must, without exception, must be done PRIOR to release. Not just change logs, but actual, this is how you use it, documentation...
Finally, I'm not all that interested in being a contributor to the code, documentation or hardware design side of things...My life is full enough right now, room for additional projects and time commitments just isn't there. This is my HOBBY...I don't make money on it, I'm playing with toys! (a toy is anything I own that doesn't make me money). This is what I do to get away from work...