After some off topic discussion on the Autotune support thread I read the current Tuning page on the wiki
http://copter.ardupilot.com/wiki/tuning/
Hidden away in the text of that page I found this little nugget which was new to me:
My copter suddenly crashed for no good reason: The I2C port, used to talk to the compass, has a rare but very serious flaw which will block code execution To avoid any issues be sure to solder the compass securely.
Now, two points about this:
- What has this got to do with tuning?
- Just, WHAT?! What is this about?
I have an external compass which I bought like everyone else after 3.0.1 came out. It's plugged in to the I2C port with the cable that came with it. That's what I thought we were supposed to do. And yet this seems to be saying that we should be soldering the compass (which bit? to where? how). And if we don't, our copters might fall out of the sky.
I can't find anything else about this, anywhere. Just that one little sentence on a wiki page about an entirely unrelated topic. It sounds like it is a fairly major thing to not know about!
I've certainly had crashes "for no good reason" but I always assume there is a good reason as I know that arducopter is now basically cleverer than me: most recently I had what I diagnosed as a brownout due to premature failure of a dying Lipo, leading to a crash from 150'. It would be nice to know if using a compass and plugging it in, as I thought we were supposed to do, can cause my copter to fall out of the sky and if so what the alternative is!
Thanks,
Harry
Replies
Andke,
Thanks, I'll give that a try when I get a chance. But, if there is a problem then such a test wouldn't necessarily identify it. I would have thought that bad / intermittent data on a port from a dodgy connection could have more of an effect in terms of constant interrupts than simply removing it.
Graham, yes that's fair enough. But just to emphasise: I don't have a problem (that I know of) with my quads - they fly well. I had a crash a while back but I just used that for illustration, I'm not suggesting that it was caused by anything other than a bad battery (just that it could have been).
My point was to ask a question about the documentation, which is what people learning about the system will (hopefully) now read and to point out that this part is inconsistent or incomplete, on an issue that sounds like it could be important.
Why is this comment there "apropos of nothing" in the tuning page?
Somebody who knows has clearly written that comment in the page, who knows about an issue that if it is real could be critical. But they have written it somewhere that it's unlikely to be found (I read up on tuning in the days before this new documentation site was written, so hadn't read this page before), and they have only given enough information to raise concern in those who happen to read it without any explanation of what if anything the issue is, who it might affect, and what if anything should be done about it.
I'm just raising a concern that this seems to be a problem with the documentation and that it would be good if this bit could be moved, and clarified or removed as appropriate. For all I know there is no problem - this seems to be the most likely, given that copters don't generally fall out of the sky - in which case this text should be removed or at least qualified. But as someone reading those docs, I don't know.
Hi Harry,
Have you tried contacting support at 3DR, I have found them to be excellent to deal with, with very good response times, DIY Drones is not the be all and end all for advice, sometimes you need to look inside the square for a change.
No offence intended to anybody that gives support to forum members.
This could be good!
Can't wait for the diagnosis, where's the popcorn?